Abstract
In recent years, the fruit industry has become an important part of agricultural development, and fruit harvesting is a key stage in the production process. However, picking fruits during the harvest season is always a major challenge. In order to solve the challenges of time-consuming, costly, and inefficient fruit picking, researchers have conducted a lot of studies on automatic fruit picking equipment. Existing picking technologies still require further research and development to improve efficiency and reduce fruit damage. Aiming at the efficient and non-destructive picking of fruits, this paper reviews machine vision and mechanical fruit picking technology and the current research status, including the current application status, equipment structure, working principle, picking process, and experimental results. As a promising tool, machine vision technology has been widely researched and applied due to its low hardware cost and rich visual information. With the development of science and technology, automated fruit picking technology integrates information technology, integrates automatic perception, transmission, control, and operation, etc., saves manpower costs, and continuously promotes the development of modern agriculture in the direction of refinement of equipment technology, automation, and intelligence. Finally, the challenges faced by automated fruit picking are discussed, and future development is looked forward to with a view to contributing to its sustainable development.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
With the dramatic increase in the world's population, the fruit industry is under pressure to increase acreage and production (Eigenbrod and Gruda 2015; Horrigan et al. 2002; Gongal et al. 2015). Due to increased social diversity and an aging population, the number of people involved in the processing of agricultural products is decreasing, and the labor required for fruit picking accounts for 60–70% of the entire growing process, making inefficient picking and high labor costs a major challenge for fruit farmers (Fess et al. 2011; Sibhatu et al. 2015; He and Schupp 2018; Vougioukas 2019). In recent years, there has been a significant increase in interest in automated fruit picking technology, which reduces the cost of picking while increasing the efficiency of picking and has great potential for future development (Hua et al. 2023).
The twenty-first century is a critical period for the transition from agricultural mechanization to intelligent automated machinery, and industrial intelligent automation is of indelible importance to the development of modern agriculture in terms of scale, diversification, and precision (An et al. 2022; Yang and Zhang 2014; Lad et al. 2022). According to the United Nations, the world population will reach 9.7 billion in 2050. This means that the world’s annual agricultural production will have to increase by about 60% to meet people’s needs (Walker 2016; Mahmud et al. 2023). China, for example, has been the world’s largest producer and consumer of fruit since 1993 (Chen et al. 2017). The most recent data in Fig. 1A shows that the area of orchards and production have increased every year. Fruit harvesting has become one of the most time-consuming and labor-intensive aspects of fruit production operations (Vrochidou et al. 2022; Jatoi et al. 2017). In recent years, researchers around the world have been working on mechanized and intelligent fruit picking techniques (Ayaz et al. 2019).
As early as the 1980s, American scholars Schertz and Brown proposed the use of robots for fruit picking, and more research results have emerged successively (Jia et al. 2020). However, previous studies have focused on the current various intelligent fruit and vegetable picking robots and their performance characteristics (Wang et al. 2022), vision control technology (Zhao et al. 2016), identification and localization (Li et al. 2022a, b), image processing technology (Hua et al. 2023), and the study of key structures (end-effector and robotic arm, etc.) (Li et al. 2022a, b; Cai et al. 2010; Lu et al. 2022; Oliveira et al. 2022). It can be concluded from this literature that these techniques have been successfully used for picking the corresponding fruits or vegetables, but all of them suffer from low accuracy and efficiency in judging the ripeness of the fruits, as well as a high level of damage. Although extensive and in-depth research has been carried out in the field of fruit picking at home and abroad and a large number of agricultural fruit picking equipment has been developed, a detailed and systematic summary of the latest research results in this field has not yet been made, and the existing review articles only collect and compare the picking equipment from one or a few specific perspectives, which is not conducive to the comprehension of the latest technology of agricultural picking equipment.
In recent years, extensive development and detailed studies based on various fruit picking techniques have been conducted Fig. 1B. Statistics on the keywords “fruit picking technology” and “fruit picking” from the Web of Science (https://www.webofscience.com/) show a rapid increase in the number of publications and citation frequency of the literature in the last 10 years. Among them, the citation frequency has increased significantly from 2019 to 2021. In this paper, we will review the picking equipment from two aspects of mechanization and machine vision-based fruit picking technology, respectively, and analyze the structure, process, picking efficiency, fruit damage level, and picking environment of the equipment. Summarize the future development trend of automated picking equipment in order to promote the modernization of the fruit industry and the promotion of high technology as a reference.
This review is organized as follows: Sect. 2 introduces mechanized fruit picking technologies, describing the characteristics and picking efficiency of seven picking technologies. Section 3 focuses on the principles and application status of machine vision-based picking technology and presents the current research status of automatic fruit picking equipment for apples, strawberries, tomatoes, citrus, and strawberries, respectively. In addition, agricultural drone fruit picking technology is introduced. Section 4 focuses on the challenges and future trends of automated fruit picking technology. Section 5 concludes the paper. The framework of this work is shown in Fig. 2.
2 Mechanical automatic fruit picking technology and equipment
In 1968, research on mechanical fruit and vegetable harvesting equipment commenced, and the United States took the lead in studying mechanical vibratory shaking and pneumatic vibratory shaking types of picking machinery (Navas et al. 2021; Bao et al. 2022). However, while these machines can complete the basic task of picking, they cause significant damage to the fruit and suffer from low picking efficiency, severely limiting their usefulness (Atanda et al. 2011). With the continuous development of science and technology worldwide, mechanical fruit harvesting equipment has undergone rapid development (Yuan and Chen 2014). Current research on mechanical fruit harvesting equipment can be broadly categorized into two types: human–machine cooperative harvesting and machine vision-based fruit picking equipment (Silwal 2016). In the former, the picking machine is operated by a worker to complete the harvesting task, while in the latter, fruit identification, positioning, and picking actions are all performed by the equipment itself (Bechar and Vigneault 2017; Yin 2020).
This section provides a comprehensive overview of the technologies used for mechanical harvesting of fruits, including vibratory, pneumatic, impact, comb and brush, push and shear, and shear technologies (Chen 2021). An extensive search was conducted to gather relevant information, and a comprehensive assessment of the current state of research and issues in the field was provided to assist scholars in their research projects.
2.1 Vibratory shaking type
Vibratory harvesting machinery utilizes a vibration device to transmit vibrations. By setting the optimal vibration frequency and amplitude, the vibration device transmits vibrations to the fruit tree, causing it to vibrate and the fruit to move with variable acceleration, leading to the fruit falling off and completing the harvest (Wu et al. 2022; Gupta et al. 2016). However, high vibration frequencies can cause damage to the fruit (Kou et al. 2022). The vibratory device of the vibratory shaking picker mainly generates vibrations and comes in two forms: the eccentric wheel type and the crank-slider mechanism type Fig. 3. Experiments on citrus harvesting using these two forms demonstrated that the harvesting rate using the crank-slider type ranged from 50 to 55%, while the eccentric type yielded 70 to 75% (Zhu et al. 2021a, b, c, d). The shaking method can be classified based on the position of the vibration: branch-trunk vibration and crown vibration. In branch trunk vibration, the clamping mechanism grips the trunk or branch, causing the eccentric wheel or crank-slider device to vibrate the tree to separate the fruit (Whitney et al. 1988).
Vibratory fruit harvesting equipment is the most widely used form of mechanical harvesting for forest fruits due to its high efficiency and low cost (Afsah-Hejri et al. 2022). However, the vibrations produced by the equipment can cause damage to the fruit (Hinsch et al. 1993). As a result, both domestic and foreign research on vibratory shaking harvesting equipment has focused mainly on blueberries, wine grapes, apples, and other fruits.
Simple berry harvesting equipment was first developed by the French in the 1940s, and by the 1960s, picking highbush blueberries with vibration was gaining importance (Calnitsky 2017). Initially, the main reliance was on manual picking and simple tools, which led to the loss of fruit due to slippage and consequent economic losses (Elik et al. 2019). Typical research shows that the United States OXBO Manufacturing Company has 25 years of experience with blueberry and raspberry harvesters, and the use of the rotary vibratory picking method can minimize plant damage (Huffman 2014; Arak 2021). BEI Company in the United States is in an international leading position in the field of blueberry picker research; its LBT Harvester, BEI Tracks Blueberry Harvester, Rotary Harvester, and other series of vibration blueberry pickers in the United States are widely used (Zhao et al. 2018; Yarborough and Hergeri 2010; Yu et al. 2014; DeVetter et al. 2019). Bao et al. (2014) conducted extensive research on automatic blueberry picking technology and developed a vibration strategy based on the blueberry harvesting equipment, which is 10 times more efficient than manual picking. A hand-pushed dwarf shrub blueberry picker and a vibratory harvester with an efficiency of 12 kg/h were developed for dwarf shrub blueberries, both of which are hand-held, easy to operate, have low manufacturing and maintenance costs, and are significantly more efficient than manual picking (Guo et al. 2012). Li et al. (2020a, b, c) designed a blueberry picker using a vibrator consisting of a crank linkage and a double rocker mechanism. It was 10.6 times more efficient than manual picking, with high efficiency, low fruit breakage, and labor savings. Another type of longitudinal vibration picker for berry bush fruits, adding a spring system as a buffer, established and analyzed the vibration model of the picking location and the lateral branches of the berry bushes and obtained the optimal vibration relationship during the picking process, with a harvest rate of 95.5% and an average error of 0.0095 (Hou et al. 2023; Chen et al. 2021a, b).
In recent years, the growing demand for high-quality wine grapes has posed a significant challenge to the development of the industry (Keller 2010). Wine grape harvesting has been mechanized in places such as Europe and the US, for example, in France, 70% of wine grapes are harvested by machine (Sarig 2012; Jones et al. 2014). The New Holland Braud series of self-propelled grape harvesters, which was launched in 1975 under the umbrella of the Case New Holland Industrial Group, has been an industry leader for more than 40 years, and experiments have shown that a single Braud series harvester can replace up to 60 harvesters at a harvest cost of around a quarter of the cost of manual harvesting (Douthie 2019; Pezzi and Martelli 2015; Fornari et al. 2021). Therefore, the development of mechanized harvesting is considered an inevitable trend in the industry (Fu et al. 2022). Yuan et al. (2020) developed a crankshaft vibratory threshing and harvesting unit with a flexible clamping vibratory mechanism, drive train, and frame, which had a harvesting success rate of 93.06% and a breakage rate of 4.57%, which is significantly more efficient than manual harvesting. Another method is to use a 4R mechanism combined with a planar hinge mechanism to achieve variable speed and directional movement during wine grape harvesting, with an average harvest rate of 85.55% (Zhu et al. 2023). It is worth noting that damage during harvest is still a major disadvantage of vibratory grape harvesting, and future research should focus on how to reduce fruit damage.
For apple picking, most of them are still manually based with low picking efficiency (Zhu et al. 2021a, b, c, d; Zhang et al. 2021a, b). Due to factors such as the large size of apples and the thin skin and thick flesh, automated robotic picking has been realized in Europe and the United States (Ghahremani 2020). However, there is a lack of research on mechanized apple picking. Figure 4A shows a hydraulically controlled vibratory apple picker. During the picking operation, the vibratory picking device clamps the trunk of a high-acid apple tree and vibrates, causing the fruit to be dislodged under inertia for picking, with a net picking rate of 95.9% and a damage rate of 1.3%, which suggests that the machine meets the requirements for picking high-acid apples (Shang et al. 2023).
Mechanized harvesting techniques using vibroseis in the fruit industry have been extensively researched in a large body of literature and have developed considerably over the last few decades. However, because of the tender and fragile nature of fruits, damage to both fruit and fruit trees often occurs. Therefore, more efforts in the future should be focused on efficient and non-destructive fruit harvesting, and the selection of the optimal vibration frequency can effectively reduce the damage to the fruit. Actuator selection is still an important aspect we are very concerned about. According to the characteristics of the fruit, the choice of flexible materials for the damage situation has also improved. With the continuous development of automation and mechanization technology, improve the applicability of orchard machinery and the reliability of mechanical equipment to achieve low fruit damage or even lossless harvesting, so as to meet the needs of the fruit market.
2.2 Pneumatic type
Pneumatic fruit harvesting machines have been a popular method for fruit harvesting since their inception (Elfferich et al. 2022). These machines utilize powerful air flow to induce movement in the fruit, which allows them to be dislodged from the tree once the force applied is greater than the resistance of the fruit stalk to the branch (Brown and Sukkarieh 2021). The air flow is generated by powerful fans attached to large tractors, and the direction of the air flow is controlled by a guiding device.
Pneumatic fruit picking machinery was first researched in the USA, particularly for citrus harvesting. Pneumatic mechanical harvesting methods were proposed in Fig. 4B. The early experiments were not very efficient due to outdated technology, low-powered equipment, and low efficiency (Jutras et al. 1963). By the 1970s, air-powered harvesting experiments were conducted with the strongest airflow speed of 43.8 m/s, where continuous impact on citrus trees for 10 s achieved a net harvesting rate of 85–90% for ripe citrus. However, continuous airflow speed during harvesting could cause mild damage to the citrus trees and, in severe cases, affect citrus yields the following year (Whitney and Patterson 1972). To improve the operational performance of the air-suction floor date picker, a prototype was designed, built, and tested to optimize its key parameters (Zhang et al. 2021a, b).
Compared with vibratory harvesting, pneumatic harvesting has the advantages of high picking efficiency, less impurity, and convenient operation. It is worth noting that pneumatic harvesting machinery is not in direct contact with the fruit plant and will not cause physical damage to the fruit skin and trunk, but because of its high-speed airflow on the fruit skin, branches, and leaves and other unavoidable damage, the application of large-scale harvesting is less. The future development of this technology should focus on reducing energy consumption, reducing damage, analyzing the correlation between the spatial structure of fruit trees and harvesting efficiency, and adapting the growth characteristics of fruit trees to mechanized operations by improving their growth characteristics.
2.3 Percussive type
Impact fruit harvesting technology uses mechanical springs or electromagnetic excitation to generate high-intensity input excitation that acts directly on the fruit tree. Make the tree branches produce instantaneous acceleration, produce vibration, and make the fruit fall off. Harvesting methods directly acting on the impact of the fruit tree effectively avoid damage to the fruit in the harvesting process, but damage to the fruit tree and branches affects the yield of the second year (Burks et al. 2005).
Studies have shown that 80–90% of sweet cherries can be harvested by both shock and vibratory shaking, so the performance of the two methods is close to each other (Erdoǧan et al. 2003; Liu et al. 2018). However, increasing the frequency of vibration during harvest may cause damage to the fruit. Therefore, it is important to control the shock or vibration frequency to avoid fruit damage (Norton et al. 1962). In order to minimize damage to the fruit tree caused by shock and to ensure that the fruit is smoothly removed from the canopy, both tree shapes were tested. The results showed that both shapes had a fruit picking rate of 90–95% with minimal damage to fruit and branches (Pellerin et al. 1978). In order to address the challenges associated with mechanical harvesting of sweet cherries and to reduce the damage caused to the fruit surface during dislodgement, an impact cherry picker was designed, characterized by the installation of an impact device on both sides of the branch. While the machine was able to achieve high efficiency (1480 kg/h) and high net picking rates, it was found that the use of two impact devices increased damage to the branches and affected fruiting rates the following year (Peterson et al. 2003). A spring-loaded impact trunk shaker was designed for apple trees, and its performance was evaluated. Test results showed that the shaker was capable of delivering up to 1151 J of energy at a speed of 5.16 m/s (Pacheco and Rehkugler 1980). Impact harvesting equipment seems to be used efficiently, but it can also lead to damage to the fruit surface and branches, which can seriously affect fruit production in the coming year and cause losses to the fruit grower. This method is suitable for fruit trees with thick stems and hard fruits with hard skin, such as walnuts, olives, and cherries (Yuan et al. 2022).
Thus, it seems that the impact fruit harvesting technology causes irreversible damage to fruit trees during harvesting, and future research should be directed towards continuously analyzing the causes of damage to fruits and plants during the harvesting process of impact machinery, optimizing the mechanism of the harvesting machinery, improving the harvesting efficiency, and reducing the impact damage. Strengthen the exploration of new technologies to replace the use of impact fruit harvesting methods to ensure fruit quality.
2.4 Brush type
The comb fruit harvester uses a combing device to act on the canopy of the fruit tree, increasing the rotation and oscillation of the combing device to dislodge the fruit (Castro-García et al. 2012). This equipment can be used for fruits such as goji berries and blueberries. In terms of blueberry harvesting, the current state-of-the-art technology is best typified by vibratory mechanical harvesting technology. According to the growth characteristics of blueberry fruits, a combing device is used to act directly on the canopy of the blueberry plant, and the fruits are dislodged through the combing device. Guo et al. (2012) proposed a hand-pushed dwarf bush blueberry harvester for dwarf bush blueberry harvesting, with a single harvesting capacity of 12 kg/h, a fruit damage rate of 10%, and a harvesting net rate of 86%. Peterson et al. (1997) developed an experimental mechanical blueberry harvester (V45) with an inclined double spike drum vibrator to ensure the reliability of harvesting ripe blueberries. Tests showed that the efficiency of the blueberry harvester was significantly higher than that of a rotary harvester and that the quality of the fruit was comparable to that of the hand-harvested blueberries, with a guaranteed quality. Figure 4C shows a comb vibratory harvester developed for field harvesting of olive trees, where parameters such as the position of picked fruit on the capture frame, the position of unseparated fruit on the tree, the degree of tree damage, and ground speed were considered during the analysis (Guirado et al. 2016). In order to analyze the force distribution within the citrus canopy, two different field experiments were carried out using a comb vibratory picker with a complete machine structure, as shown in Fig. 4D. The first experiment was designed to investigate the effect of fruit position on the forces applied, while the second experiment investigated the distribution of forces and accelerations along the length of the branch. The results showed that the forces applied to the fruit within the canopy were higher than those applied to the fruit at the edges (Savary et al. 2011).
In order to harvest apples, Hu (2020) proposed a flexible comb harvesting method using a flexible comb harvesting platform. The principle of fruit detachment during the harvesting process was investigated using a fruit-branch system dynamics model, and the motions and forces involved were analyzed. It was found that the rigid-flexible coupling model was the most effective for fruit detachment. Comb-type apple harvesting equipment has been shown to be effective in reducing fruit damage, thereby significantly increasing harvest and damage rates (Zhao 2022). The spiral brush apple harvester was developed for machine harvesting of apple trees, and this machine can efficiently pick, transport, and collect apples in batches. The results showed that it is suitable for apple harvesting as it achieves the highest efficiency. It can be seen that comb-brush harvesting of fruit is widely used in the small berry sector, and harvesting efficiency is significantly higher.
Compared with vibration, impact, and other harvesting methods, brush-type harvesting machinery greatly reduces the damage to the fruit and plant, and the harvesting efficiency of the fruit has improved. For the development of this technology, it should be focused on the continuous optimization of the brush structure, selecting suitable brush materials for the fruit so that the brush-type harvester achieves the best harvesting effect, studying the characteristics of the bonding force between the branch and the fruit stalk, analyzing the factors affecting fruit shedding, and conducting field harvesting experiments to determine the best harvesting machinery parameters.
2.5 Push-shear type
The push-shear picking technique is less automated and is suitable for large orchards with flat terrain. However, a push-shear cherry picker based on the pruning principle can quickly pick cherries with a wide range of stem profiles. Field trials have shown that the picker can pick more than 9.7 kg of cherries in half an hour, with a stalk picking success rate of more than 98% and a damage rate of less than 1%. The researchers invented a push-shear cherry picking device that can be recommended for cherry trees, improving operational efficiency and saving manpower. The device drops cherries onto a picking net for easy sorting and storage (Fig. 4E) (Yu and Ampazidis 2022; Xu et al. 2021).
Push-and-shear harvesting technology is also currently facing gradual elimination because it requires manual assistance to complete harvesting, has a low degree of automation, is inefficient, and is not suitable for large-scale production. Currently, with the development of computer technology, many people will turn their attention to agriculturally intelligent robots to reduce the dependence on manual labor and improve overall productivity, which is also the future trend of automated fruit harvesting.
2.6 Shear type
In order to solve the difficulties and safety hazards of manual fruit picking for a wide range of fruits with high growth, such as apples, pears, peaches, etc., Gong (2020) invented a high-level fruit shear picker that can pick fruits quickly and non-destructively, improves the efficiency of picking, and reduces the cost. Another invention is a shear-type woody fruit picker (Li 2021), which harvests high branches by adjusting a telescopic rod and using a clamping and shearing device to shear off the fruit, with the picking principle shown in Fig. 4F. Although these inventions are effective in practice, their overall operation and structure are complicated, and each mechanism works independently. The operation is cumbersome, which limits their use. However, it solves the dilemma of the high cost, low efficiency, fruit damage, and worker injury of traditional manual ladder climbing.
Although shear-type harvesting machinery can complete the basic harvesting tasks, there are still drawbacks, such as the low degree of automation and cumbersome structure. Future research should focus on simplifying the structure and reducing the cost, and we believe that the main thing should be to strengthen the degree of automation, improve flexibility, and choose intelligent actuators to ensure that the fruit will be harvested smoothly, avoiding the impact of the harvesting process on the surrounding flowers and foliage.
3 Fruit picking technology based on machine vision
In recent years, although mechanized fruit harvesting technology has matured, it is more damaging to branches and fruits, less efficient, and still requires manual involvement in the process to achieve fruit harvesting. Therefore, there is an urgent need to develop an efficient and damage-free fruit picking technology. This section describes the principles of machine vision technology, the current status of research in the agricultural field and the research progress of fruit picking equipment applying machine vision technology.
3.1 Principles of machine vision picking technology
Machine vision technology is used to mimic human visual functions, using a control system to analyse and process image information from objective objects for practical applications in harvesting, measurement and control (Kamkar et al. 2020; Pathare et al. 2013; Erol et al. 2007). This technology involves mechanics, computer technology, image processing, image recognition and localisation, artificial intelligence, signal processing and many other fields (Chen and Gong 2015). As fresh fruit products require good eating and appearance quality, selective harvesting methods are needed to ensure that ripe fruit can be harvested quickly and without damage (Paturi and Cheruku 2021; De Corato 2020). The basic structure of a fruit harvesting device based on machine vision technology consists of an autonomous mobile platform, a multi-degree-of-freedom robotic arm, a force feedback system with flexible end-effectors, a multi-sensor machine vision system, a drive control system, an intelligent decision-making system and auxiliary hardware and software (Duan et al. 2021; Jia et al. 2020; Tang et al. 2020a, b).
The initial phase of picking focuses on visual sensing perception techniques and learning crop information (Zou et al. 2012; Zhao et al. 2016), including camera components to recognize target fruit information (Wang et al. 2019), fruit localization, target background recognition (features such as colour, shape, texture and pose), 3D reconstruction, robot behavior planning based on visual localization mechanisms and vision (Fang 2019). For smooth fruit picking, vision systems have multiple sensing capabilities such as vision-mechanical cooperative control, vision recognition, coordinated vision-mechanical positioning and fault tolerance (Davidson et al. 2020). The control system should be collaborative and use a vision servo-controlled picking mechanism to perform the operation of pinching and cutting ripe fruit stalks (Ronzhin et al. 2022). The end-effector is required to accurately receive commands from the control system and collaborate with the robot arm to complete the fruit picking (Arad et al. 2020).
With the rapid development of computer and automation technology and the application and popularity of agricultural high technology, robotics is gradually entering the field of agricultural production (Wakchaure et al. 2023; Mao et al. 2021). It is worth noting that at present, there are still some challenges to be solved in the process of fruit harvesting robot picking machine vision technology. Due to the complexity of the crop-growing environment, such as branches and leaves, neighboring fruits, and some uncertain factors, this may cause inaccurate recognition and positioning, affecting the picking efficiency (Xiong et al. 2018; Wang et al. 2017). To address this challenge, Li et al. proposed a multi-armed apple harvesting robotic system, which utilizes a fruit accurate recognition and localization algorithm based on multi-task deep convolutional neural network (DCNN) technology to improve the recognition rate and localization accuracy of potentially occluded fruits (Li et al. 2023a, b), and at the same time, determining the shape and size of the fruits, the growing environment, the planting method, the biological characteristics, etc. are also beneficial for accurate fruit harvesting (Li et al. 2020a, b, c). In addition, the path planning and obstacle avoidance speed of harvesting robots in the field will also have a great impact on the operation speed of harvesting robots, and the use of machine learning combined with artificial neuron network technology to improve the recognition and processing ability of harvesting robots in complex environments will be the key to the development of navigation and positioning technology in the future (Li et al. 2020a, b, c; Wang and Liu 2020; Kondratenko et al. 2022; Zhou et al. 2022). In harvesting, for fresh fruits, picking robot arm picking is easy to break the fruit skin, affecting the commercialization of fresh fruits, and there is a phenomenon of missed picking; therefore, for different fruits, the research and development of a specialized, lightweight, flexible picking robot arm is imminent (Vrochidou et al. 2022; Zhou et al. 2021).
3.2 Current status of machine vision technology research
With the development of automation technology, investment in artificial intelligence research is increasing. Machine vision, a branch of AI, is now widely used for recognition and identification purposes in various work environments, such as the fruit picking equipment in Fig. 5A. This technology simulates human visual functions by capturing and processing images through a camera and then uploading the images to a personal computer for practical applications and control. Machine vision technology combines expertise from several fields such as image processing, machine automation, optics, vision sensors, virtual control, and computer applications (Gao et al. 2020). In recent years, with the rapid development of industrial intelligence, machine vision technology, which has the advantages of convenience, accuracy, speed, and intelligence, has been widely used in various fields such as industrial inspection, vision robotics, intelligent agriculture, and unmanned driving and has received more and more attention (Dong and Han 2021; Sun et al. 2018; Ren et al. 2022; Pérez et al. 2016; Mavridou et al. 2019). The development of computer vision technology has mainly included target detection and recognition, image segmentation, pose estimation, behavior analysis, etc., and the commonly used algorithms include convolutional neural networks (CNN), support vector machines (SVM), and deep learning (ResNet, YOLO) (Yang et al. 2021; Leo et al. 2017; Sharma et al. 2020; Srivastava et al. 2021). The development of machine vision technology, on the one hand, is thanks to the improvement of computer and camera performance; on the other hand, it is also inseparable from the optimization and innovation of the core algorithms, and currently, in the context of the AI big data era, the deep learning algorithms of 5G deep fusion will increase the accuracy of machine vision exponentially, so improving the accuracy of AI algorithms is the focus of future research and the difficulty(Ren and Wang 2022; Tang et al. 2023; Mahmood et al. 2022). Although machine vision technology has been better developed in various fields, there is still a gap between the diversity and complexity of research objects and meeting the needs of practical applications (Arrieta et al. 2020). Vision technology used in the field of automated fruit picking faces a huge amount of data, redundant information, feature space dimensions, and other characteristics (Cubero et al. 2016). A single simple feature extraction algorithm is difficult to meet the algorithm’s requirements for universality because of the field environment of light and shade conditions, the crop’s environment changes, and the accuracy of the recognition and positioning will be affected (Peng et al. 2023; Fu et al. 2021; Zhu et al. 2020; Shekhar et al. 2020). Future research and development should further improve autonomous navigation, fruit localization, and identification, among other issues.
The introduction of the technology of machine vision, instead of the traditional manual inspection methods, greatly improves the quality of the products put on the market and increases the production efficiency (Syam and Sharma 2018; Tao et al. 2018). It is widely used in food and beverage, cosmetics, pharmaceuticals, building materials and chemicals, metal processing, electronics manufacturing, packaging, automotive manufacturing and other industries (Malik et al. 2023; Kabbour and Luque 2020). It can be broadly divided into two directions, one is non-destructive testing and the other is vision robotics. For example, Zhu et al. (2021a, b, c, d) reviewed the research progress in the application of machine vision technology in food processing, presented the challenges and future trends. Dowlati et al. (2012) applied machine vision technology to fish quality assessment and gave an outlook of future development. Penumuru et al. (2020) introduced a machine vision robot using machine vision and introduces a generic approach to automatic material recognition using machine vision and machine learning techniques to improve the cognitive capabilities of machine tools deployed in Industry 4.0 as well as material handling equipment such as robots. Li et al. (2023a, b) provides an overview of the current state of machine vision technology in furniture manufacturing and summarises the challenges faced by machine vision. In recent years, the scale of research on machine vision technology in the field of smart agriculture has been increasing, especially in the field of smart fruit and vegetable picking (Sharma et al. 2020; Shaikh et al. 2022; Saleem et al. 2021). The picking equipment typically consists of mobile platforms, PCs, machine vision components, control system cameras, end-effectors, and robotic arms, as shown in Fig. 5B. The operation process mainly relies on the detection and recognition of external features such as fruit color, texture, and shape, and operates the end-effector via image information processing on the PC system (Chen et al. 2022).
The key to the future development of vision technology in the field of fruit picking is still based on fruit recognition, positioning, and fruit separation (Yang et al. 2023a, b; Tian et al. 2020). Strengthen the research on the biomechanical characteristics of the picking object, optimize the recognition algorithm, develop adaptive scene analysis algorithms and adaptive control systems to adapt to new types of scene analysis algorithms and adaptive control systems, integrate 5G and IoT technologies, and realize the development of the picking robot’s operational capabilities for multi-scene and multi-crop types (Rong et al. 2021; Riaz et al. 2022; Liu et al. 2023a, b). Research and development of advanced materials suitable for picking, microsensors, actuators, and soft machinery applied to the picking process, and the gradual realization of strong adaptability and efficient autonomous harvesting (Zhang et al. 2023; Wang and Chortos 2022).
3.3 Application status of machine vision picking equipment
In recent years, machine vision technology, as an important branch of artificial intelligence, has been at the core of achieving intelligent perception, which is one of the inevitable key technologies for the development of smart agriculture, effectively liberating the labor force and also improving the quality and yield of crop products. Some representative research results are presented in Sect. 3.3, reviewing the current status of harvesting fruits such as tomato, apple, citrus, strawberry, and kiwi. In addition, in order to explore the challenges of harvesting fruits growing at high altitudes, several machine vision-based UAV harvesting techniques are presented.
3.3.1 Apple picking
In 1985, France developed the first apple-picking robot, which can basically pick an apple in 10 s in a test environment. Then later, Japan, the United States, Israel, and other agriculturally developed countries also joined the research on apple picking robots, which was also quite successful (Karkee et al. 2021; Tillett 1993). For example, Israel’s apple picking robot on the Gala apple picking test has 12 picking robotic arms, the use of a “3 claw” design, similar to catching dolls, and the apple one by one to pick down is 8–10 times more than the manual apple picking (Hohimer et al. 2019; Bergerman et al. 2016). Apple picking, as one of the advantageous industries in American agriculture, is even more automated. As shown in Fig. 6A, six hands are installed on the fruit picking robot, picking about 30 apples in a minute, from identification to picking in only 1.5 s. With satellite navigation combined with visual analysis, it can accurately identify ripe apples on the tree, and it adopts a vacuum suction type for all-weather picking (Zhang et al. 2018).
To evaluate the efficiency of different harvesting maneuvers, Bu et al. (2022) developed an apple harvesting device with an integrated vision system, flexible end-effector, and manipulator (Fig. 6B) and evaluated it using two harvesting maneuvers with a harvesting success rate of more than 80% and a damage rate of 0, but with a longer harvesting cycle. In order to improve the efficiency of harvesting equipment in various environments, Zhao et al. (2022) proposed a fast, high-precision harvesting device with advanced vision capabilities. Tests were conducted under different light conditions to evaluate the performance of the mobile platform, the reliability of the vision system, and the harvesting efficiency of the robotic arm. The results show that it is an efficient harvesting device that can identify ripe fruits at an average speed of about 1 s, with a picking speed of about 25 s/pc and a picking success rate of more than 95%.
To improve harvesting efficiency and reduce fruit damage, De-An et al. (2011) developed an apple picking robot (Fig. 6C). During harvesting, the pressure value on the apple surface was continuously monitored to prevent damage caused by excessive clamping force, and experimental results showed that the average recognition time was 15.4 s and the picking success rate was 77%. In order to solve the problem of damage caused to branches and fruits during harvesting, Davidson et al. (2016) developed an automatic apple harvester using a low-cost “non-sensor” system, and in order to facilitate the detection of clustering and occlusion of fruits, the machine vision system incorporates the Circular Hough Transform (Circular Hough Transform) and the Circular Block Analysis method. Using a six-degree-of-freedom (6DOF) serial link design for the manipulator and end-effector, the picking results showed that 95 out of 100 fruits were picked, with an average positioning and picking time of 1.2 and 6.8 s per fruit, respectively, and an accuracy of up to 90% could be achieved, with a success rate of more than 95% of the picking, proving the efficiency and accuracy of the picking of the fruits. Therefore, the study of apple picking robots is one of the research hotspots and difficulties in the field of agricultural robotics, which involves a number of disciplines such as mechanics, electronics, control, artificial intelligence, etc., and has a wide range of development prospects.
3.3.2 Strawberry picking
Strawberry is a labor-intensive crop whose harvesting mechanization is hindered by its own physiological characteristics, such as inconsistent ripening, thin and soft skin, and the irregular arrangement of each cluster of fruits after ripening (Baur and Iles 2023; Yarborough and Hergeri 2010). Therefore, the research and development of a non-destructive and efficient strawberry picking robot are imminent (Zhu et al. 2023; Abasi et al. 2018). In recent years, the German company Organifarms designed the “BERRY” strawberry picking robot, which is the most typical and can automatically detect the location of strawberries, ripeness, and quality and complete the picking machine collection work (Parsa et al. 2023). The strawberry picking robot from the Belgian company Octinion can identify a strawberry in 5 s, is equipped with multiple cameras with machine vision capabilities, can automatically generate 3D images and locate them, and can determine on its own whether a strawberry is ripe or not (Bogue 2020; Woo et al. 2020; Van Henten 2019; Verbiest et al. 2021).
Currently, strawberry picking still faces problems such as low efficiency, high damage rate, inaccurate recognition and localization, and high cost, and in recent years, relevant researchers have been committed to improving these problems with a view to designing efficient and damage-free strawberry picking robots (Yu et al. 2020; Rehman et al. 2022). Ling et al. (2021) designed a strawberry intelligent picking robot that can accurately identify and locate the position of ripe strawberries and harvest the fruits by gripping and twisting the fruit stalks. The experimental results showed that the average speed of discrimination of strawberries by this machine was 1 s, the rate of misjudgment of the fruits was 7%, and the success rate of the picking was 90%. To address the severe labor shortage in Western agriculture, Octinion has developed an autonomous strawberry picking device. Three RGB cameras were used for 3D vision and strawberry detection based on color differences, and picking was done by a robotic arm, which showed a picking speed of 4 s/pc with significantly higher efficiency and lower damage rates (De Preter et al. 2018).
To overcome the effects of the picking environment, Xiong et al. (2020) designed a porous strawberry continuous picking robot with a Hokuyo radar for navigation sensing at the front position and proposed a new obstacle separation algorithm to enable the picking system to pick bunches of strawberries. The results showed that the success rate of the first picking ranged from 50.0 to 97.1%, and the success rate of the second picking increased to 75.0 to 100.0%, with the failure being attributed to the limitations of the vision system as well as the lack of flexibility of the gripper. Meanwhile, to improve factors such as low picking efficiency and slow picking cycle, Xiong et al. (2019) designed a cable-driven strawberry picking robot as shown in Fig. 6D. Field experiments showed that the picking success rate was 96.8%, the average cycle time for picking a single strawberry consecutively was 7.5 s, or 10.6 s if all procedures were included, and the average picking success rate in a farm environment was 53.6%, or 59.0% if “success with damage” was included. To enable picking robots to work in complex environments, Huang et al. (2020) developed an automated strawberry picking device that utilized a human–robot cooperation approach for target recognition and a robotic arm controller for picking. Feng et al. (2015) developed a strawberry harvester that utilized an image segmentation algorithm based on the OHTA color space, and harvesting experiments were evaluated for performance. The device was found to be highly accurate in detecting strawberry stems, with an accuracy of 93% and more than 90% accuracy in determining the maturity and shape quality of strawberries against a black and white background. As shown in Fig. 6E, Han et al. (2012) developed an autonomous robot for harvesting strawberries cultivated in a tabletop system, where strawberry detection was based on a stereoscopic CCD color camera and laser device, respectively. The strawberry detection was carried out based on the 3D image and distance information obtained from the stereo CCD color camera and laser device, respectively. A DC servo motor-driven end-effector was designed to pick strawberries without any damage in less than 7 s, which greatly improves the problem of long harvesting cycles and damages. Hayashi et al. (2010) developed a strawberry harvesting robot that was designed based on the concepts of nighttime operation, pedal handling, and task sharing with the workers. The experiments showed that the machine vision device recognized the stalks with an accuracy of 60% and successfully picked a fruit in 11.5 s.
The above study shows that although the existing strawberry picking technology has gradually matured, it is still some distance from being practical and commercialized. The existing strawberries are planted in a regular and disorderly manner, which leads to an increase in the difficulty of identification and positioning of the harvesting robot and an increase in the picking cycle. Therefore, in order to make strawberry picking more efficient, the planting of strawberries should be regular and orderly, and the intelligent recognition and positioning of the harvesting robot, the design of the mechanical structure, path planning, and control system need further research and improvement.
3.3.3 Tomato picking
As one of the three major world-trade fruits and vegetables, tomatoes contain rich nutritional value (Iloh et al. 2020; Diop and Jaffee 2005). The global tomato planting area and production have increased significantly, but they are facing problems such as concentrated ripening and low picking efficiency (Li et al. 2018). In response to the above problems, the development of tomato picking robots basically solves the problem of time-consuming and labor-intensive tomato picking (Tian et al. 2022a, b; Bachche 2015).
Israeli agribusiness research and development of the Metomotion tomato automatic picking machine is the most typical and applicable to greenhouse tomato picking (Hughes et al. 2022; Saraiva et al. 2023). The development of a multi-purpose robotic system, GRoW, equipped with the most advanced robotics and automation technology, can achieve efficient and non-destructive tomato picking (Xie et al. 2022a, b; Cheng et al. 2023). Japanese agri-tech company Inaho has developed a tomato picking robot that can realize fully automated picking and can be used for up to 12 h on a single charge, working around the clock (de Bourgogne 2021). U.S. greenhouse company Appharvest, through the collection of a large number of tomato image data, can develop a tomato picking robot that can identify 50 kinds of tomato varieties and determine the maturity of tomatoes (Childers 2020). In recent years, the development of large language models such as ChatGPT has set off a revolution in the field of robotics, and the first tomato picking robot co-developed by ChatGPT and humans has appeared, advancing the development of fruit and vegetable picking robots (Gill and Kaur 2023; AlZu’bi et al. 2022).
In order to improve the efficiency of tomato picking, Feng et al. (2015) designed a tomato intelligent picking robot, as shown in Fig. 6F. The robot used an image recognition algorithm to detect ripe tomatoes in the field of view and determined the spatial position based on the linear laser positioning principle. The results showed that the picking success rate was as high as 83.9%, but the picking time was long, about 24 s/pc. In order to realize the robot’s ability to pick ripe tomatoes in a greenhouse, a picking robot for solar greenhouses with a vertical trellis was designed to automatically detect and identify the fruits during the picking and collection process, resulting in an efficient and damage-free picking of tomatoes, which may be inaccurately identified and located due to external factors (Yu et al. 2022). In order to achieve automated tomato picking, a tomato picking robotic arm was modeled, and a recognition method was developed to create a stereo vision model using the circular Hough transform and RGB color space. Fifty tests were conducted in a simulated environment with a success rate of 78% (Zhou et al. 2018). Yaguchi et al. (2016) developed a stereoscopic camera tomato picking device using infinite rotating joints for gripping, which allows depth measurements to be taken in direct sunlight. The device was evaluated through tomato picking competitions and on-farm experiments, with a picking speed of 80 s/fruit and a success rate of 60%, which is inefficient and much lower than manual picking speeds. Yasukawa et al. (2017) designed an infrared image- and specular reflection-based tomato picking robot for indoor fruit detection and recognition, with an accuracy in real-world image evaluation of 88.1%.
A dual-arm tomato harvesting robot for greenhouses was designed using binocular vision sensors with a 95% success rate in detecting ripe tomatoes, with only a few missed detections due to leaf shading. Tests showed that the positioning error was less than 10 mm and the harvesting success rate was 87.5%, which was significantly higher than the manual harvesting efficiency when applied to actual harvesting (Ling et al. 2019). Rong et al. (2022) designed an integrated adsorption and gripping manipulator in order to accurately identify the position of the tomato fruits and estimate the grasping posture to improve the success and efficiency of the robotic harvesting and developed an optimal sorting algorithm and an optimal sorting algorithm for the fruits. Optimal sorting algorithm and fruit nearest-neighbor localization algorithm, and designed directional grasping and sequential picking control strategies. Evaluation results showed that the string and fruit recognition accuracy based on the YOLOv5m was 90.2% and 97.3%, respectively, and the success rate of fruit harvesting was increased to 72.1%, with an average harvesting time for individual fruits reaching 14.6 s. Gao et al. (2022) developed a pneumatic finger end-effector for cherry tomato picking equipment that uses a combination of gripping and rotation to pick ripe fruits consistently and stably, and field tests showed that the average cycle time for picking a single cherry tomato was 6.4 s. The picking success rates for pickable cherry tomatoes in different directions were 84% (right), 83.3% (rear), 79.8% (left), and 69.4% (front).
Currently, strawberry picking has been automated by robots, but from the existing literature, it can be seen that in the current shortage of labor, the research and development of picking robots basically improves the problem of low efficiency in strawberry picking, but because of its complex growth environment, there is still a long period of picking and inaccurate identification and positioning problems.
3.3.4 Citrus picking
Citrus is rich in resources and excellent varieties, and in recent years, there has been an increasing demand for citrus (Liu et al. 2022). Currently, citrus harvesting is still a labor-intensive and time-intensive task, with the risk of injury to the fruit growers (Khatri et al. 2021). Automated citrus harvesting can reduce the labor risk and improve the picking efficiency of the fruit growers (Ferreira et al. 2018; Blasco et al. 2019).
Spain’s Agri-Tech invented a citrus picking robot that can pick 60 citrus a minute, compared to 8 a minute by hand. The robot can also sort the picked citrus by size via a video recorder (Zhao et al. 2020; Harman and Sklar 2022). To address the challenges of commercializing citrus harvesting robots, Yin et al. (2023) provide a fully integrated, autonomous, and innovative solution for citrus harvesting robots to overcome the difficulties of harvesting citrus due to its natural growth characteristics, propose a new visual estimation of the fruit pose, and design a new end-effector, which allows the robot to harvest citrus continuously with an overall success rate of 87.2% and an average picking time of 10.9 s/pc.
For the automated picking of citrus, Yang et al. (2019) proposed a citrus robot, which had a success rate of 83.6% and 91.9% in identifying obstacles and fruit ripeness, respectively, with an error of 5.9 mm, a processing time of 0.4 s for a single image frame, and a success rate of 80.51% and 75.79% in picking ripe citrus and avoiding obstacles, respectively. In order to improve the low automation of picking, a citrus fruit picking robot was developed, which was divided into a control part and a mechanical part. The picking results showed that the average picking time was 5.4 s and the maximum picking height was 1.85 m. The robot has the ability to identify, localize, pick, sort, and box citrus (Liu et al. 2019). As shown in Fig. 6G, Wang et al. (2019) developed a citrus harvesting device that uses a tongue-shaped end-effector to randomly harvest citrus fruits along the direction of the stalk. The results of laboratory citrus stalk cutting and harvesting experiments showed that the average cutting success rate was as high as ~ 98% over the range of deflection angles [− 50°, 50°], and the harvesting rate was ~ 89% in the optimal position. The harvesting results in the natural environment showed that the harvesting rate was as high as 74% in the optimal position. Yang et al. (2023a, b) developed a lightweight, Raspberry Pi-based platform. A citrus intelligent recognition picking robot was developed based on the Raspberry Pi platform, and a citrus deep recognition picking system was designed. A deep convolutional neural network (YOLOv4-tiny) was used to verify the validity of the citrus dataset, with a recognition rate of 98%, which can achieve the recognition of citrus three-dimensional positional coordinates and accurate picking, and 20 picking tests were conducted, with an average success rate of picking of more than 90%.
With the rapid development of modern agricultural technology, the automation of citrus production and harvesting is an inevitable trend. It should be noted that although the citrus automated harvesting robot has successfully realized the identification and harvesting of citrus, the automated harvesting of citrus still faces some challenges, such as adaptability to different varieties and growing environments, stability of the robot, etc., but these challenges will be gradually overcome with the continuous progress of technology.
3.3.5 Kiwifruit picking
Kiwifruit is a nutritious fruit with a unique flavor and high economic returns (Barman et al. 2021). At present, kiwifruit harvesting is mainly done manually, which requires a lot of manpower for picking, sorting, and packing due to the relatively concentrated harvesting period and short harvesting cycle (Sarkar 2021; Barbole et al. 2022). Therefore, it is necessary to promote the automation of kiwifruit harvesting to make up for the problem of labor shortage and to achieve the mechanization, informationization, and standardization of the kiwifruit industry (Ren et al. 2023).
The automatic kiwifruit picker developed by Robotics Plus of New Zealand picks ripe kiwifruit from the fruit rack through four machine picking arms and then transmits them to the collection box through the tube bag under the machine arm to realize the work of picking, sorting, packaging, etc. The robot can pick kiwifruit from the fruit rack through four machine-picking arms (Zhang et al. 2019; Karkee et al. 2019). As shown in Fig. 6H, northwest A&F University in China has developed a kiwifruit-picking robot. After the robot starts working, the camera will accurately identify and locate the fruit, after which the bionic manipulator is able to quickly clamp and pick without damage, combining with the hardening characteristics of kiwifruit harvesting to achieve efficient kiwifruit collection (Fu 2023).
Mu et al. (2020) developed a kiwifruit picking robot (Fig. 6I), which proposed an automated picking method based on kiwifruit characteristics, where the end-effector approached the fruit from below, wrapped and grasped the fruit with two bionic fingers, and then bending the fingers would allow separation of the fruit from the trunk. Tests showed an average picking time of 4–5 s for 240 fruits with a picking success rate of 94.2%, demonstrating the potential of the grasp-pick-gather end-effector. A compact and lightweight kiwifruit picker was developed based on kiwifruit characteristics. It recognizes and locates ripe fruits and picks them by clamping and twisting. The design of the machine lays the foundation for further research in automatic sorting technology (Gao et al. 2013). Williams et al. (2019) developed a multi-armed kiwifruit harvesting device with a novel harvesting mechanism to efficiently harvest kiwifruit from the canopy. Field trials demonstrated a harvest rate of 51% with an average harvest cycle time of 5.5 s/unit, but with long wastage times and fruit losses of up to 23.4%. With the new improved vision system and two manipulators, the success rate for harvesting fruit was 86.0% and 55.8% for kiwifruit, with a cycle time of 2.78 s per unit (Williams et al. 2020).
As shown in Fig. 6J, Mu et al. (2017) proposed a picking method with “grab-pick-slide” and designed an end-effector with bionic fingers, information sensing, and machine vision for non-destructive kiwifruit picking. Tests showed a picking success rate of 90%, an average loss rate of 10%, and an average picking cycle time of 4 s. The device has great utility in achieving non-destructive fruit picking. As shown in Fig. 6K, He et al. (2022) proposed a two-arm cooperative method for mechanical picking of kiwifruit in orchards. The method consists of three steps: determining the picking position, collision detection, and a continuous picking cycle. The picking test of the dual-arm collaborative platform showed an average picking success rate of 86.67% and a collision detection time of 3.95 ± 0.83 s for each fruit.
To improve the efficiency of kiwifruit harvesting, a multi-mechanical-arm kiwifruit harvesting device was proposed by Barnett et al. (2020). Ten harvesting tests were conducted, all of which yielded good results, and the use of a multi-mechanical arm was the most effective method for completing the harvesting operation. In order to automate kiwifruit harvesting, Chen et al. (2012) designed an end-effector for a kiwifruit harvesting device that was mounted on the front end of the robotic arm. In addition, the end-effector mechanism, sensing system, and control system were designed. Tests showed that the end-effector had a gripping success rate of 100%, a picking success rate of 90%, and an average picking cycle time of 9 s/pc.
Kiwifruit harvesting is a labour-intensive task, and it is clear from existing research that automated kiwifruit harvesting has become a need and a trend for the future as kiwifruit production becomes more and more significant.
3.3.6 Drone picking techniques
Existing picking robots are mostly designed with ground mobile platforms, which cannot complete the picking operations targeting higher-growing fruit trees and inter-mountain orchards (Xie et al. 2022a, b; Hussain et al. 2022). Currently, using UAVs as mobile carriers to carry end-picking end-effectors is one of the solutions to high-altitude fruit picking, which has the advantages of high harvesting efficiency and flexibility (Varadaramanujan et al. 2017; Suryawanshi et al. 2022).
There are fewer studies on fruit picking by drones, with the apple picking drone developed by Israeli agricultural drone manufacturer Tevel Aerobotics being the most typical (Neupane et al. 2023; https://www.tevel-tech.com/, 2022). As shown in Fig. 6L, this UAV combines flight technology, robotics, and AI technology. It is also equipped with a gripper, a camera, a protective frame, etc., and has achieved more than 90% accuracy in picking ripe fruits in trials (Maor 2022). Meanwhile, another achievement of Tevel’s fruit picking UAV system is fruit grading. When the fruit picking UAV is selectively picking fruits, the harvesting result is significantly better than manual picking and 10% higher than manual grading harvesting (Vrochidou et al. 2022; Eminoglu and Yegul 2022). In 2020, Tevel proposed a harvesting solution based on UAV picking technology, as shown in Fig. 6M, where a robotic arm mounted on the UAV mainly recognizes the location and ripeness of fruits through visual sensors. When ripe fruits are detected within a certain range, the UAV flies near the fruit trees, grabs the fruits with the robotic arm, and picks them by rotating and twisting them. The collected fruits end up in a collection device (Maor 2023).
Jagadeeswaran has proposed a smart coconut and palm-cutting drone for efficient fruit harvesting. The quadcopter UAV is equipped with a vision system, a slider crank mechanism to lift a payload of 1 kg, and a cutter on the top to separate the fruits from the tree (Jagadeeswaran et al. 2021). Chu et al. (2022) proposed a novel device for harvesting red pine fruits divided into a drone and a harvesting mechanism, and the tests have shown that this harvesting device is highly automated, simple to operate, and cost-effective. Tang et al. (2020a, b) developed a bionic snake mouth harvesting mechanism and an unmanned harvesting device based on visual positioning that have high harvesting efficiency, low labor costs, and can effectively prevent fruit damage and personnel injuries during the harvesting process. Yan et al. (2021) developed a fruit tree picking UAV that uses a micro-hydraulic rod to move a mobile blade, which, together with a fixed blade, accurately separates the fruit stalks from the fruit tree through probe detection. This UAV enables non-destructive fruit picking, reduces labor costs, and improves picking efficiency. Zhu et al. (2021a, b, c, d) proposed an autonomous lychee picking device based on a quadcopter UAV. The system uses an identification control system to detect ripe lychee fruits and an actuator-end scissor assembly to cut the stalks, which allows for non-destructive fruit picking that is cost-effective, safe, and reliable.
From the existing literature, it can be seen that the existing research on picking UAVs is mostly in the design stage, many designs are idealized and not tested on the ground, and the related control problems have to be further solved. Compared with ground-picking robots, picking UAVs have higher requirements for range, identification and positioning, and obstacle avoidance.
4 Challenges and future trends
As mentioned earlier, most fruit picking experimental studies have shown that fruit picking equipment based on mechanization and machine vision demonstrates unique advantages and development prospects in the agricultural field. In recent years, many research teams have conducted extensive scientific studies on automated fruit picking technologies and have made significant breakthroughs in key technologies. However, experiences gained from picking experiments have concluded that low efficiency, a high damage rate, low automation, inaccurate identification and positioning, and the high cost of complex structures are the main challenges of current fruit picking technologies. In order to address these challenges, it is crucial to reduce costs, increase identification and positioning accuracy, and improve efficiency.
4.1 Mechanised fruit picking technology
Mechanical fruit picking equipment can cause greater damage to fruit and branches during harvesting. Vibratory picking techniques may cause the fruit to fall directly to the ground, damaging the fruit skin and branches. Chen et al. (2021a, b) designed a vibratory blueberry picker, and experimental results showed that the picker caused 6.7% damage to branches. It has been shown in the literature that pneumatic picking may cause damage to the fruit surface, branches, and leaves due to larger air currents (Afsah-Hejri et al. 2022). Junming et al. (2021) conducted harvesting trials on apples using an impact vibratory picking machine, which achieved an efficiency of more than 90% but caused damage to the plant. Zhang et al. (2015) designed a comb-and-brush berry picking machine to harvest mature fruits with a picking rate of 86.70% and a fruit damage rate of 8.62%, pointing out that comb-and-brush harvesting can cause different degrees of damage to branches and leaves. The push-shear harvesting technique caused less damage to fruits, but the net harvesting rate was lower (Yu and Ampazidis 2022). Shear harvesting may cause damage to the fruits as the stalks are not cut at a good distance, which tends to destroy the fruits (Xu et al. 2023). Although the use of orchard harvesting machinery has become widespread, damage to fruits and stems during harvesting still occurs, which constrains the development of this equipment. To solve this problem, future research and development should focus on reducing the damage rate and improving the picking efficiency, such as by adding some flexible materials as actuators, giving priority to materials that cause less damage to fruits when selecting materials, and determining the optimal picking parameters through a large number of experiments. In the future, fruit harvesting machinery may develop in the direction of low damage, high efficiency, and low cost. Equipment’s ability to operate in complex environments, overcoming the effects of external factors, or effectively responding to or avoiding these situations through means such as sensors and software analysis. In addition, researchers and developers should have the expertise to solve problems in a timely manner and to effectively improve and maintain the performance of vision systems.
Complex orchard environments pose a challenge to achieving a high level of automation in fruit picking, and picking machinery requires a large amount of information to ensure that the task is completed (Jia et al. 2022; Yoshida et al. 2022). However, due to the high vibration frequency, air velocity, impact frequency, and combing speed in complex environments, unripe fruits may be picked prematurely, leading to economic losses. Researchers have designed human–machine-assisted devices, but most of them are used for semi-automatic picking and still require human intervention to identify and locate ripe fruits. As a result, a large amount of manual labor is still required during fruit picking. Picking machinery is complex and expensive to maintain, which may lead to increased equipment damage. The internal structure of the machinery is cumbersome and requires several mechanical parts to work together, which is relatively costly. Safe management and maintenance of equipment are critical, but neglecting both can lead to problems, deterioration, and reduced harvesting efficiency (Duckett et al. 2018). In addition, the complexity of the equipment can lead to high maintenance costs and operator difficulties (Ampatzidis et al. 2014). Therefore, it is urgent to accelerate the development towards automation, intelligence, and user-friendliness so that harvesting equipment can be developed in the direction of simple operation, low failure rate, safety, and reliability.
In view of the above problems, automatic fruit harvesting machinery is one of the difficulties and hot spots in the field of agricultural high technology. Its harvesting efficiency has been a difficult problem for researchers. Future research and development work should be devoted to improving the harvesting efficiency of harvesting machinery, such as through a large number of experiments to determine the optimal operating parameters of harvesting machinery, to maximize the possibility of overcoming some of the difficulties in harvesting, and to improve the timeliness of the equipment for harvesting fruits. It is also helpful. In addition, most of the currently available harvesting equipment is designed for specific fruits or shapes with poor versatility, which increases the production cost. To address this issue, future R&D efforts should focus on upgrading existing harvesting equipment to be multifunctional and adaptable to different fruits and environments, and such multifunctional equipment would only require replacing the corresponding modules, which reduces costs and improves efficiency. For the design of actuators, such as shear mechanisms, flexible materials should be used, which can greatly avoid the damage caused to the fruits.
Complex orchard environments can reduce the efficiency of harvesting fruit. Existing mechanized harvesting technology requires manual assistance to complete the harvesting task, which is not fully automated and therefore imposes strict requirements on the fruit-growing environment. In order to achieve a high degree of automation of harvesting equipment, it is necessary to develop standardized planting specifications for different types of fruits. To achieve fruit planting standardization, scaling, specialization, and factory, it will help to reduce the complexity of the harvesting operation and improve efficiency, which will lay the foundation for the development of harvesting machinery and is also the future development trend in this field.
Harvesting machinery has a complex structure and high maintenance costs, increasing the risk of equipment damage (El-Termezy et al. 2022). Its structure involves several mechanical components, leading to high costs. Neglecting safety management and maintenance can lead to equipment problems, deterioration, and loss of efficiency. The complexity of the equipment can also lead to high maintenance costs and difficulties for operators (Bechar and Vigneaul 2016). As orchard harvesting machinery technology continues to mature, harvesting efficiency and fruit quality have improved. However, the structure of harvesting equipment often consists of numerous auxiliary components, which increases manufacturing and maintenance costs. In order to better serve the processing of agricultural products, future harvesting equipment should focus on simplicity, high precision, high automation, and low cost. For example, it is possible to combine the harvesting process with the grading and transport of fruits to reduce the costs required for segmented harvesting.
4.2 Machine vision picking technology
The lower recognition efficiency of vision systems used for fruit picking can affect the accuracy of fruit picking (Wan and Goudos 2020). In addition, the recognition efficiency of picking equipment can be reduced by a variety of external factors, including the complexity of the fruit-growing environment, such as whether the fruit is obscured by debris, branches, or overlapping growth (Xiao et al. 2023; Karkee et al. 2018). Natural conditions, such as the light, color, shape, texture of the fruit, inclement weather, noise, and geography, can also negatively affect the recognition efficiency, making it difficult for the visual recognition system to accurately identify the location of the fruit, and these problems seriously affect the picking efficiency (Tang et al. 2020a, b; Liu et al. 2020). Some studies have shown that Zhuang et al. (2023) introduced an artificial potential field (APF) for path planning of a robot based on a six-degree-of-freedom robot for situations such as obstacle obstruction and combined it with the A* algorithm, which is highly adaptive to obstacle avoidance path planning and is able to complete the obstacle avoidance path planning in a faster and more reasonable way. In addition, the use of deep learning techniques can improve the target recognition performance of harvesting robots in complex environments, but there are still many uncontrollable influencing factors, and the stability of visual recognition is also problematic (Ghasemi et al. 2022; Wu et al. 2021). Therefore, in order to maximize the potential of harvesting equipment, future R&D efforts should focus on improving the equipment’s ability to operate in complex environments, overcoming the effects of external factors, or effectively responding to or avoiding these situations through means such as sensors and software analysis. In addition, researchers and developers should have the expertise to solve problems in a timely manner and to effectively improve and maintain the performance of vision systems.
Picking cycle time and efficiency are key factors in measuring the performance of fruit picking equipment (Li et al. 2011). Although the equipment must ensure that fruit damage is minimized and does not affect the next year’s yield, it is less efficient and not as effective as manual labor (Connor et al. 2014). Wang et al. (2023) pointed out that in the case of cherry tomato harvesting, factors such as inefficient recognition, long processing time, and susceptibility to subjective factors are often encountered, which limit the accuracy of the harvesting robots in complex scenarios and robustness, most of which require manual participation. A self-built target detection algorithm for hollies is proposed, and the results show that the average accuracy rate reaches 95.2%, and the improved model can perform real-time target recognition and maturity detection for hollies. Therefore, to improve the efficiency of harvesting equipment, future research should focus on enhancing infrastructure maintenance to ensure optimal component interaction, improving the response time of the recognition and localization system, increasing the harvesting speed of the end-effector, and constructing finer target recognition and detection methods.
Future research and development of fruit harvesting equipment should focus on simplifying structures, improving harvesting efficiency, and reducing manufacturing and maintenance costs (Bac et al. 2014). Existing equipment involves multiple fields such as physical, mechanical, electronic information, and intelligent system control, requiring expensive hardware and software facilities to achieve high efficiency, and maintenance of complex equipment requires specialized personnel, leading to increased operating costs (Zhang et al. 2020). Data suggests that the world’s first AI-picking robot, Robocrop, cost a staggering £700,000 to develop. Meanwhile, new research from Interact Analysis suggests that the robotic harvesting market is in the early stages of growth but has huge potential—valued at $236 million in 2022 but set to rise to $6.8 billion by 2030 (https://interactanalysis.com/). Reducing R&D costs has become the trend for future harvesting robots, and improving the relevant supporting facilities of harvesting equipment can reduce costs. The various supporting facilities of harvesting equipment should be fully utilized to improve their maintenance. In order to achieve intelligent, efficient, and low-cost harvesting, operators should monitor the working status of each facility in real time and find the best parameters through experiments. In addition, future development should focus on miniaturization, intelligence, and user-friendliness to fully replace manual harvesting and contribute to rural revitalization.
Existing fruit picking equipment is designed for specific types or shapes of fruits and is therefore less versatile (Li et al. 2022a, b). Most fruit picking is done in a short period of time, which poses a challenge for storage and extended shelf life. In order to reduce manufacturing costs and better serve agriculture, fruit picking equipment must be highly versatile to pick a wide range of fruit types (Tian et al. 2020). Mechanical damage during fruit harvesting often leads to a significant reduction in fruit quality and economic losses. It has been shown that during harvesting, fruit damage due to factors such as abrasion or friction accelerates water loss, and bacteria can penetrate into the fruit, leading to rapid decay and spoilage (Komarnicki et al. 2017; Tensaw 2020; Sudheer and Indira 2007). Therefore, measures need to be taken to minimize damage to the fruits, and the materials used for the pusher and collection device should be improved; the use of soft materials can minimize damage caused by direct contact between the fruits and the end of the pusher, avoiding direct dropping of the fruits and reducing the impact of environmental factors. At present, the development and promotion of various intelligent fruit picking equipment is accelerating this process, greatly reducing costs and improving efficiency. This is also the future development trend of automated fruit picking.
5 Conclusions
This paper reviews the progress of the application of mechanized technology and machine vision-based technology in the field of fruit harvesting. It clearly points out the categories, harvesting principles, and harvesting efficiencies of existing mechanized harvesting technologies at home and abroad, and the use of this technology for fruit harvesting needs to take into account the damage to the fruit and the degree of automation. In order to achieve efficient and damage-free fruit harvesting, machine vision-based fruit harvesting equipment has become a hot research topic in recent years. In order to obtain detailed information about the environment and fruits, the harvesting equipment is equipped with various types of vision sensors and image analysis algorithms. Fruit picking such as apples, strawberries, tomatoes, citrus, and kiwifruit is reviewed in detail, and the use of drone picking technology can achieve fast and efficient picking for high-growth fruit trees and inter-mountain orchards.
In addition, on the basis of the overview of the current situation of the application of the above picking equipment, this paper summarizes some of the difficulties that still exist in the practical application of mechanized technology and machine vision technology in the field of fruit picking, such as low efficiency, high cost, high damage, and difficulties in identification and positioning. In view of the above problems in practical application, the future development of the two picking technologies is worth looking forward to. In summary, with the rapid development of computer and automation technology, the application of agricultural robots, and the development of a new generation of information technology such as the Internet of Things and the accelerated deployment of 5G networks, there is an urgent need to design a new machine vision algorithm to improve the feature extraction capability, the feature selection capability, and the feature classification capability, and the researchers have found that the 5G deep fusion deep learning algorithm combines the efficiency and robustness of machine vision with the flexibility of human vision, and the combined machine vision harvesting system not only has the ability to detect in complex environments, but also improves greatly in real-time. In the face of manual picking labor, damage and other problems are endless. The fruit of the automated harvesting of accurate and non-destructive problems needs to be solved. In recent years, non-contact image sensors accordingly came out, not only to detect the size and shape of the fruit but also to the fruit of the damage to the appearance of the analysis of the fruit, suitable for a variety of types of fruit harvesting and sorting, so that efficient and accurate harvesting of fruits has become a reality. It is believed that with the continuous progress of technology and the support of government policy, the research of automatic fruit harvesting equipment will be a direction with broad prospects.
References
Abasi S, Minaei S, Jamshidi B, Fathi D (2018) Dedicated non-destructive devices for food quality measurement: a review. Trends Food Sci Technol 78:197–205. https://doi.org/10.1016/j.tifs.2018.05.009
Afsah-Hejri L, Homayouni T, Toudeshki A, Ehsani R, Ferguson L, Castro-García S (2022) Mechanical harvesting of selected temperate and tropical fruit and nut trees. Hortic Rev 49:171–242. https://doi.org/10.1002/9781119851981.ch4
AlZu’bi S, Mughaid A, Quiam F, Hendawi S (2022) Exploring the capabilities and limitations of ChatGPT and alternative big language models. Artif Intell Appl. https://doi.org/10.47852/bonviewAIA3202820
Ampatzidis YG, Vougioukas SG, Whiting MD, Zhang Q (2014) Applying the machine repair model to improve efficiency of harvesting fruit. Biosyst Eng 120:25–33. https://doi.org/10.1016/j.biosystemseng.2013.07.011
An Z, Wang C, Raj B, Eswaran S, Raffik R, Debnath S, Rahin SA (2022) Application of new technology of intelligent robot plant protection in ecological agriculture. J Food Qual 2022:1–7. https://doi.org/10.1155/2022/1257015
Arad B, Balendonck J, Barth R, Ben-Shahar O, Edan Y, Hellström T et al (2020) Development of a sweet pepper harvesting robot. J Field Robot 37(6):1027–1039. https://doi.org/10.1002/rob.21937
Arak M (2021) Cultivation technology for lowbush blueberry cultivation in milled peat field plantations
Arrieta AB, Díaz-Rodríguez N, Del Ser J, Bennetot A, Tabik S, Barbado A et al (2020) Explainable artificial intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI. Inf Fusion 58:82–115. https://doi.org/10.1016/j.inffus.2019.12.012
Atanda SA, Pessu PO, Agoda S, Isong IU, Ikotun I (2011) The concepts and problems of post-harvest food losses in perishable crops. Afr J Food Sci 5(11):603–613
Ayaz M, Ammad-Uddin M, Sharif Z, Mansour A, Aggoune EHM (2019) Internet-of-Things (IoT)-based smart agriculture: toward making the fields talk. IEEE Access 7:129551–129583. https://doi.org/10.1109/ACCESS.2019.2932609
Bac CW, Van Henten EJ, Hemming J, Edan Y (2014) Harvesting robots for high-value crops: state-of-the-art review and challenges ahead. J Field Robot 31(6):888–911. https://doi.org/10.1002/rob.21525
Bachche S (2015) Deliberation on design strategies of automatic harvesting systems: a survey. Robotics 4(2):194–222. https://doi.org/10.3390/robotics4020194
Bao YD, Guo YL, Guo S (2014) The current situation and development trend of blueberry picking machinery. Energy Sav Technol 03:228–230
Bao Y, Yuan N, Zhao Y, Wu L (2022) Recent patents for collection device of fruit harvesting machine. Recent Patents Eng 16(4):96–108. https://doi.org/10.2174/1872212116666220107115125
Barbole DK, Jadhav PM, Patil SB (2022) A review on fruit detection and segmentation techniques in agricultural field. In: Second international conference on image processing and capsule networks: ICIPCN 2021 2. Springer International Publishing, pp 269–288. https://doi.org/10.1007/978-3-030-84760-9_24
Barman M, Das AB, Badwaik LS (2021) Effect of xanthan gum, guar gum, and pectin on physicochemical, color, textural, sensory, and drying characteristics of kiwi fruit leather. J Food Process Preserv 45(5):e15478. https://doi.org/10.1111/jfpp.15478
Barnett J, Duke M, Au CK, Lim SH (2020) Work distribution of multiple Cartesian robot arms for kiwifruit harvesting. Comput Electron Agric 169:105202. https://doi.org/10.1016/j.compag.2019.105202
Baur P, Iles A (2023) Replacing humans with machines: a historical look at technology politics in California agriculture. Agric Hum Values 40(1):113–140. https://doi.org/10.1007/s10460-022-10341-2
Bechar A, Vigneaul C (2016) Agricultural robots for field operations: concepts and components. Biosyst Eng 149:94–111. https://doi.org/10.1016/j.biosystemseng.2016.06.014
Bechar A, Vigneault C (2017) Agricultural robots for field operations. Part 2: operations and systems. Biosyst Eng 153:110–128. https://doi.org/10.1016/j.biosystemseng.2016.11.004
Bergerman M, Billingsley J, Reid J, van Henten E (2016) Robotics in agriculture and forestry. In: Springer handbook of robotics. pp 1463–1492. https://doi.org/10.1007/978-3-319-32552-1_56
Blasco J, González GMG, Chueca P, Cubero S, Aleixos N (2019) Advances in automated in-field grading of harvested crops. In: Robotics and automation for improving agriculture. Burleigh Dodds Science Publishing, pp 215–232
Bogue R (2020) Fruit picking robots: has their time come? Ind Robot Int J Robot Res Appl 47(2):141–145. https://doi.org/10.1108/IR-11-2019-0243
Brown J, Sukkarieh S (2021) Design and evaluation of a modular robotic plum harvesting system utilizing soft components. J Field Robot 38(2):289–306. https://doi.org/10.1002/rob.21987
Bu L, Chen C, Hu G, Sugirbay A, Sun H, Chen J (2022) Design and evaluation of a robotic apple harvester using optimized picking patterns. Comput Electron Agric 198:107092. https://doi.org/10.1016/j.compag.2022.107092
Burks T, Villegas F, Hannan M, Flood S, Sivaraman B, Subramanian V, Sikes J (2005) Engineering and horticultural aspects of robotic fruit harvesting: opportunities and constraints. HortTechnology 15(1):79–87. https://doi.org/10.21273/HORTTECH.15.1.0079
Cai W, Chen Y, Zou X, Wu J, Xu D (2010) Design and simulation of virtual prototype for picking manipulator’s end-effector. In 2010 3rd international conference on computer science and information technology, vol 7. IEEE, pp 246–249. https://doi.org/10.1109/ICCSIT.2010.5565181
Calnitsky N (2017) Harvest histories: a social history of Mexican farm labour in Canada since 1974 (Doctoral Dissertation, Carleton University)
Castro-García S, Blanco-Roldán GL, Jiménez-Jiménez F, Muñoz-Tejada R, Gil-Ribes JA (2012) Table olive fruit and tree suitability to mechanical harvesting methods. In: International conference of agricultural engineering CIGR-AgEng. pp 8–12
Chen ZB (2021) Current status of research on vibratory forest fruit harvesting technology. Mech Eng 01:21–24
Chen GZ, Gong SR (2015) Computer vision and pattern recognition technology in the field of agricultural production. Jiangsu Agric Sci. https://doi.org/10.15889/j.issn.1002-1302.2015.08.133
Chen J, Wang HU, Jiang HR, Gao H, Lei WL, Dang GR (2012) Design of an end-effector for a kiwifruit picking robot. J Agric Mach 43(10):151–154. https://doi.org/10.6041/j.issn.1000-1298.2012.10.027
Chen J, Chen C, Yao D (2017) Analysis on the comparative advantage and export competitiveness of China’s fruit products. In: International conference on transformations and innovations in management (ICTIM 2017). Atlantis Press, pp 476–486. https://doi.org/10.2991/ictim-17.2017.36
Chen JY, Wang Y, Liang DT, Xu WH, Chen Y (2021a) Design and test of miniaturised axial vibration blueberry picker. Mech Des. https://doi.org/10.13841/j.cnki.jxsj.2021.04.006
Chen J, Wang Y, Liang D, Xu W, Chen Y (2021b) Design of a buffered longitudinal vibratory picking mechanism for berry shrub fruits. Trans ASABE 64(4):1165–1171. https://doi.org/10.13031/trans.14119
Chen PL, Zhu LX, Zhang SA (2022) Research progress of fruit precision recognition algorithm for picking robots. Modern Agricultural Equipment (02):8–13+42
Cheng C, Fu J, Su H, Ren L (2023) Recent advancements in agriculture robots: benefits and challenges. Machines 11(1):48. https://doi.org/10.3390/machines11010048
Childers B (2020) Rethinking the Appalachian economy: how modern technology can transform agriculture in mountainous regions. Ky J Equine Agric Nat Resour L 13:389
Chu Y, Cheng HY, Meng LP, Chen L, Li H, Liu B, Cui TH, Li JP, Fu XX (2022) A preliminary study on the use of drones for picking red pine cones. Forestry Machinery and Woodworking Equipment (01):64–67. https://doi.org/10.13279/j.cnki.fmwe.2022.0009
Connor DJ, Gómez-del-Campo M, Rousseaux MC, Searles PS (2014) Structure, management and productivity of hedgerow olive orchards: a review. Sci Hortic 169:71–93. https://doi.org/10.1016/j.scienta.2014.02.010
Cubero S, Lee WS, Aleixos N, Albert F, Blasco J (2016) Automated systems based on machine vision for inspecting citrus fruits from the field to postharvest—a review. Food Bioprocess Technol 9:1623–1639. https://doi.org/10.1007/s11947-016-1767-1
Davidson JR, Silwal A, Hohimer CJ, Karkee M, Mo C, Zhang Q (2016) Proof-of-concept of a robotic apple harvester. In: 2016 IEEE/RSJ international conference on intelligent robots and systems (IROS). IEEE, pp 634–639. https://doi.org/10.1109/IROS.2016.7759119
Davidson J, Bhusal S, Mo C, Karkee M, Zhang Q (2020) Robotic manipulation for specialty crop harvesting: a review of manipulator and end-effector technologies. Glob J Agric Allied Sci 2(1):25–41. https://doi.org/10.35251/gjaas.2020.004
de Bourgogne RM (2021) Smart farming technology in Japan and opportunities for EU companies. ECOS
De Corato U (2020) Improving the shelf-life and quality of fresh and minimally-processed fruits and vegetables for a modern food industry: a comprehensive critical review from the traditional technologies into the most promising advancements. Crit Rev Food Sci Nutr 60(6):940–975. https://doi.org/10.1080/10408398.2018.1553025
De Preter A, Anthonis J, De Baerdemaeker J (2018) Development of a robot for harvesting strawberries. IFAC-PapersOnLine 51(17):14–19. https://doi.org/10.1016/j.ifacol.2018.08.054
De-An Z, Jidong L, Wei J, Ying Z, Yu C (2011) Design and control of an apple harvesting robot. Biosyst Eng 110(2):112–122. https://doi.org/10.1016/j.biosystemseng.2011.07.005
DeVetter LW, Yang WQ, Takeda F, Korthuis S, Li C (2019) Modified over-the-row machine harvesters to improve northern highbush blueberry fresh fruit quality. Agriculture 9(1):13. https://doi.org/10.3390/agriculture9010013
Diop N, Jaffee SM (2005) Fruits and vegetables: global trade and competition in fresh and processed product markets. In: Aksoy MA, Beghin JC (eds) Global agricultural trade and developing countries. World Bank, pp 237–257
Dong J, Han Q (2021) Research on high speed robot sorting system based on machine vision technology. In 2021 IEEE 4th international conference on information systems and computer aided education (ICISCAE). IEEE, pp 554–558. https://doi.org/10.1109/ICISCAE52414.2021.9590676
Douthie S (2019) Hi-tech harvest in Victoria’s King valley. Aust N Z Grapegrow Winemak 666:41
Dowlati M, de la Guardia M, Mohtasebi SS (2012) Application of machine-vision techniques to fish-quality assessment. Trends Anal Chem 40:168–179. https://doi.org/10.1016/j.trac.2012.07.011
Duan JL, Wang ZR, Ye L, Yang Z (2021) Research progress and development trend of fruit picking robot arm motion planning. J Intell Agric Equip (in English and Chinese) 2(2):11. https://doi.org/10.12398/j.issn.2096-7217.2021.02.002
Duckett T, Pearson S, Blackmore S, Grieve B, Chen WH, Cielniak G et al (2018) Agricultural robotics: the future of robotic agriculture. arXiv Preprint. https://arxiv.org/abs/1806.06762. https://doi.org/10.48550/arXiv.1806.06762
Eigenbrod C, Gruda N (2015) Urban vegetable for food security in cities. A review. Agron Sustain Dev 35:483–498. https://doi.org/10.1007/s13593-014-0273-y
Elfferich JF, Dodou D, Della Santina C (2022) Soft robotic grippers for crop handling or harvesting: a review. IEEE Access 10:75428–75443. https://doi.org/10.1109/ACCESS.2022.3190863
Elik A, Yanik DK, Istanbullu Y, Guzelsoy NA, Yavuz A, Gogus F (2019) Strategies to reduce post-harvest losses for fruits and vegetables. Strategies 5(3):29–39. https://doi.org/10.7176/JSTR/5-3-04
El-Termezy G, Abd El Hamid S, Sabry H (2022) Development of a fruits harvesting machine. Middle East J Agric Res 11(01):01–10. https://curresweb.com/index.php/MEJAR1/article/view/3
Eminoglu MB, Yegul U (2022) Smart farming application in fruit harvesting. Research & reviews in agriculture, forestry and aquaculture, 45.
Erdoǧan D, Güner M, Dursun E, Gezer İ (2003) Mechanical harvesting of apricots. Biosyst Eng 85(1):19–28. https://doi.org/10.1016/S1537-5110(03)00024-2
Erol A, Bebis G, Nicolescu M, Boyle RD, Twombly X (2007) Vision-based hand pose estimation: a review. Comput Vis Image Underst 108(1–2):52–73. https://doi.org/10.1016/j.cviu.2006.10.012
Fang YM (2019) Research and application of 3D reconstruction methods based on stereo vision and machine learning (Master’s Thesis, South China Agricultural University). https://doi.org/10.27152/d.cnki.ghanu.2019.000964
Feng Q, Wang X, Wang G, Li Z (2015) Design and test of tomatoes harvesting robot. In: 2015 IEEE international conference on information and automation. IEEE, pp 949–952. https://doi.org/10.1109/ICInfA.2015.7279423
Ferreira MD, Sanchez AC, Braunbeck OA, Santos EA (2018) Harvesting fruits using a mobile platform: a case study applied to citrus. Engenharia Agricola 38:293–299. https://doi.org/10.1590/1809-4430-Eng.Agric.v38n2p293-299/2018
Fess TL, Kotcon JB, Benedito VA (2011) Crop breeding for low input agriculture: a sustainable response to feed a growing world population. Sustainability 3(10):1742–1772. https://doi.org/10.3390/su3101742
Fornari M, Chiomento JLT, De Nardi FS, dos Santos Trentin N, dos Santos Trentin T, Amado TJC (2021) Mechanized grape harvesting in Brazil: an agronomic view and new challenges Colheita mecanizada de uva no Brasil: uma visão agronômica e novos desafios. Braz J Dev 7(6):58182–58193. https://doi.org/10.34117/bjdv7n6-290
Fu L (2023) Strategic short note: intelligent sensing and robotic picking of kiwifruit in orchard. In: IoT and AI in agriculture: self-sufficiency in food production to achieve society 5.0 and SDG’s globally. Springer Nature Singapore, Singapore, pp 283–288. https://doi.org/10.1007/978-981-19-8113-5
Fu Y, Yang G, Pu R, Li Z, Li H, Xu X et al (2021) An overview of crop nitrogen status assessment using hyperspectral remote sensing: current status and perspectives. Eur J Agron 124:126241. https://doi.org/10.1016/j.eja.2021.126241
Fu J, Ji C, Liu H, Wang W, Zhang G, Gao Y et al (2022) Research progress and prospect of mechanized harvesting technology in the first season of ratoon rice. Agriculture 12(5):620. https://doi.org/10.3390/agriculture12050620
Gao H, Wang H, Chen J (2013) Research and design of a kiwifruit picking robot. Agric Mech Res. https://doi.org/10.13427/j.cnki.njyi.2013.02.038
Gao JJ, Qu ZH, Song YQ (2020) Machine vision technology research and application status and development trend. China Media Technology. https://doi.org/10.19483/j.cnki.11-4653/n.2020.07.001
Gao J, Zhang F, Zhang J, Yuan T, Yin J, Guo H, Yang C (2022) Development and evaluation of a pneumatic finger-like end-effector for cherry tomato harvesting robot in greenhouse. Comput Electron Agric 197:106879. https://doi.org/10.1016/j.compag.2022.106879
Ghahremani M (2020) An economic analysis of a robotic harvest technology in New Zealand fresh apple industry: a dissertation presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Agribusiness, Massey University School of Agriculture and Environment, Manawatu, New Zealand (Doctoral dissertation, Massey University). http://hdl.handle.net/10179/16658
Ghasemi Y, Jeong H, Choi SH, Park KB, Lee JY (2022) Deep learning-based object detection in augmented reality: a systematic review. Comput Ind 139:103661. https://doi.org/10.1016/j.compind.2022.103661
Gill SS, Kaur R (2023) ChatGPT: vision and challenges. Internet of Things Cyber Phys Syst 3:262–271. https://doi.org/10.1016/j.iotcps.2023.05.004
Gong BL (2020) A shear picking device for high level fruits: C N Patent CN111631011A [P]. 2020-09-08
Gongal A, Amatya S, Karkee M, Zhang Q, Lewis K (2015) Sensors and systems for fruit detection and localization: a review. Comput Electron Agric 116:8–19. https://doi.org/10.1016/j.compag.2015.05.021
Guirado RRS, Jiménez FJ, Blanco-Roldan GL, García SC, Ruiz FJC, Ribes JA (2016) Vibration parameters assessment to develop a continuous lateral canopy shaker for mechanical harvesting of traditional olive trees. Span J Agric Res 14(2):3. https://doi.org/10.5424/sjar/2016142-7909
Guo YL, Bao YD, He PZ, Wang HB (2012) Design and testing of a hand-pushed lowbush blueberry picker. J Agric Eng 28(7):40–45. https://doi.org/10.3969/j.issn.1002-6819.2012.07.007
Gupta SK, Ehsani R, Kim NH (2016) Optimization of a citrus canopy shaker harvesting system: mechanistic tree damage and fruit detachment models. Trans ASABE 59(4):761–776. https://doi.org/10.13031/trans.59.10819
Han KS, Kim SC, Lee YB, Kim SC, Im DH, Choi HK, Hwang H (2012) Strawberry harvesting robot for bench-type cultivation. J Biosyst Eng 37(1):65–74. https://doi.org/10.5307/JBE.2012.37.1.065
Harman H, Sklar EI (2022) Multi-agent task allocation techniques for harvest team formation. In: International conference on practical applications of agents and multi-agent systems. Springer International Publishing, Cham, pp 217–228. https://doi.org/10.1007/978-3-031-18192-4_18
Hayashi S, Shigematsu K, Yamamoto S, Kobayashi K, Kohno Y, Kamata J, Kurita M (2010) Evaluation of a strawberry-harvesting robot in a field test. Biosyst Eng 105(2):160–171. https://doi.org/10.1016/j.biosystemseng.2009.09.011
He L, Schupp J (2018) Sensing and automation in pruning of apple trees: a review. Agronomy 8(10):211. https://doi.org/10.3390/agronomy8100211
He Z, Ma L, Wang Y, Wei Y, Ding X, Li K, Cui Y (2022) Double-arm cooperation and implementing for harvesting kiwifruit. Agriculture 12(11):1763. https://doi.org/10.3390/agriculture12111763
Hinsch RT, Slaughter DC, Craig WL, Thompson JF (1993) Vibration of fresh fruits and vegetables during refrigerated truck transport. Trans ASAE 36(4):1039–1042. https://doi.org/10.13031/2013.28431
Hohimer CJ, Wang H, Bhusal S, Miller J, Mo C, Karkee M (2019) Design and field evaluation of a robotic apple harvesting system with a 3D-printed soft-robotic end-effector. Trans ASABE 62(2):405–414. https://doi.org/10.13031/trans.12986
Horrigan L, Lawrence RS, Walker P (2002) How sustainable agriculture can address the environmental and human health harms of industrial agriculture. Environ Health Perspect 110(5):445–456. https://doi.org/10.1289/ehp.02110445
Hou J, He Z, Liu D, Zhu Z, Long Z, Yue X, Wang W (2023) Mechanical damage characteristics and nondestructive testing techniques of fruits: a review. Food Sci Technol 43:e001823. https://doi.org/10.1590/fst.001823
Hu GR (2020) Research on the key technology of comb-type apple harvesting (Master’s Thesis, Northwest Agriculture and Forestry University). https://doi.org/10.27409/d.cnki.gxbnu.2020.001028
Hua X, Li H, Zeng J, Han C, Chen T, Tang L, Luo Y (2023) A review of target recognition technology for fruit picking robots: from digital image processing to deep learning. Appl Sci 13(7):4160. https://doi.org/10.3390/app13074160
Huang Z, Sklar E, Parsons S (2020) Design of automatic strawberry harvest robot suitable in complex environments. In: Companion of the 2020 ACM/IEEE international conference on human-robot interaction. pp 567–569. https://doi.org/10.1145/3371382.3377443
Huffman WE (2014) Agricultural labor: demand for labor. Encycl Agric Food Syst 1:105–122
Hughes J, Lida F, Birrell SA (2022) Field robotics for harvesting. In: Digital agritechnology: robotics and systems for agriculture and livestock production, 69. https://doi.org/10.1016/B978-0-12-817634-4.00009-4
Hussain S, Fatima K, Cheema MJM, Saleem SR, Iqbal T (2022) Applications of robotics and UAVs in orchards for fruit picking. Environ Sci Proc 23(1):29. https://doi.org/10.3390/environsciproc2022023029
Iloh EC, Nwokedi M, Onyebukwa CF, Ekeocha Q (2020) World Trade Organization’s trade liberalization policy on agriculture and food security in West Africa. Regional Development in Africa, 10. https://doi.org/10.5772/intechopen.86558
Jagadeeswaran P, Duraisamy RT, Elavarasan M, Tamilarasu P, Yogeshwaran P (2021) A smart coconut and palm cutting drone. Inf Technol Ind 9(3):204–208
Jatoi MA, Jemrić T, Sito S (2017) Mechanized pre & post-harvest practices of berry crops cultivation. Glasnik Zaštite Bilja 40(4):84–93. https://doi.org/10.31727/gzb.40.4.7
Jia W, Zhang Y, Lian J, Zheng Y, Zhao D, Li C (2020) Apple harvesting robot under information technology: a review. Int J Adv Rob Syst 17(3):1729881420925310. https://doi.org/10.1177/1729881420925310
Jia W, Wei J, Zhang Q, Pan N, Niu Y, Yin X et al (2022) Accurate segmentation of green fruit based on optimized mask RCNN application in complex orchard. Front Plant Sci. https://doi.org/10.3389/FPLS.2022.955256
Jones JE, Kerslake FL, Close DC, Dambergs RG (2014) Viticulture for sparkling wine production: a review. Am J Enol Vitic 65(4):407–416. https://doi.org/10.5344/ajev.2014.13099
Junming H, Weixue H, Wei W, Hongjie Z, Zhi, R (2021) Fruit vibrantion harvesting technology and its damage :a review. INMATEH-Agric Eng 63(1)
Jutras PJ, Coppock GE, Patterson JM (1963) Harvesting citrus fruit with an oscillating air blast. Trans ASAE 6(2):192–194. https://doi.org/10.13031/2013.40863
Kabbour M, Luque R (2020) Furfural as a platform chemical: from production to applications. Biomass Biofuels Biochem. https://doi.org/10.1016/B978-0-444-64307-0.00010-X
Kamkar S, Ghezloo F, Moghaddam HA, Borji A, Lashgari R (2020) Multiple-target tracking in human and machine vision. PLoS Comput Biol 16(4):e1007698. https://doi.org/10.1371/journal.pcbi.1007698
Karkee M, Silwal A, Davidson JR (2018) Mechanical harvest and in-field handling of tree fruit crops. In: Automation in tree fruit production: principles and practice. CABI, Wallingford UK, pp 179–233. https://doi.org/10.1079/9781780648507.0179
Karkee M, Gord J, Sallato B, Whiting MD (2019) Optimizing fruit production efficiencies through mechanization. In: Achieving sustainable cultivation of temperate zone tree fruits and berries. Burleigh Dodds Science Publishing, pp 347–372. https://doi.org/10.19103/AS.2018.0040.10
Karkee M, Zhang Q, Silwal A (2021) Agricultural robots for precision agricultural tasks in tree fruit orchards. In: Innovation in agricultural robotics for precision agriculture: a roadmap for integrating robots in precision agriculture. pp 63–89. https://doi.org/10.1007/978-3-030-77036-5_4
Keller M (2010) Managing grapevines to optimise fruit development in a challenging environment: a climate change primer for viticulturists. Aust J Grape Wine Res 16:56–69. https://doi.org/10.1111/j.1755-0238.2009.00077.x
Khatri S, Shrestha S, Pokharel KP (2021) Evaluation of manual fruit harvesters and storability characteristics of harvested sweet orange under ordinary room storage condition. Sustain Food Agric 2(2):84–91. https://doi.org/10.26480/sfna.02.2021.84.91
Komarnicki P, Stopa R, Kuta L, Szyjewicz D (2017) Determination of apple bruise resistance based on the surface pressure and contact area measurements under impact loads. Comput Electron Agric 142:155–164. https://doi.org/10.1016/j.compag.2017.08.028
Kondratenko Y, Atamanyuk I, Sidenko I, Kondratenko G, Sichevskyi S (2022) Machine learning techniques for increasing efficiency of the robot’s sensor and control information processing. Sensors 22(3):1062. https://doi.org/10.3390/s22031062
Kou X, Wang D, Zhou JB, Tang JY (2022) Analysis of the current situation of domestic research on forest fruit picking equipment. Forestry Machinery and Woodworking Equipment (06), pp 15–21. https://doi.org/10.13279/j.cnki.fmwe.2022.0092
Lad AM, Bharathi KM, Saravanan BA, Karthik R (2022) Factors affecting agriculture and estimation of crop yield using supervised learning algorithms. Mater Today Proc 62:4629–4634. https://doi.org/10.1016/j.matpr.2022.03.080
Leo M, Medioni G, Trivedi M, Kanade T, Farinella GM (2017) Computer vision for assistive technologies. Comput Vis Image Underst 154:1–15. https://doi.org/10.1016/j.cviu.2016.09.001
Li CL (2021) A shear type woody fruit picker: C N Patent CN112166822A [P]. 2021-01-05
Li P, Lee SH, Hsu HY (2011) Review on fruit harvesting method for potential use of automatic fruit harvesting systems. Procedia Eng 23:351–366. https://doi.org/10.1016/j.proeng.2011.11.2514
Li B, Lecourt J, Bishop G (2018) Advances in non-destructive early assessment of fruit ripeness towards defining optimal time of harvest and yield prediction—a review. Plants 7(1):3. https://doi.org/10.3390/plants7010003
Li Y, Iida M, Suyama T, Suguri M, Masuda R (2020a) Implementation of deep-learning algorithm for obstacle detection and collision avoidance for robotic harvester. Comput Electron Agric 174:105499. https://doi.org/10.1016/j.compag.2020.105499
Li ZP, Zhang C, Wang BN, Yu DY, Wang HB (2020b) Design study of a blueberry picker based on vibration strategy. For Eng. https://doi.org/10.16270/j.cnki.slgc.2020.02.009
Li Z, Guo R, Li M, Chen Y, Li G (2020c) A review of computer vision technologies for plant phenotyping. Comput Electron Agric 176:105672. https://doi.org/10.1016/j.compag.2020.105672
Li Z, Yuan X, Wang C (2022a) A review on structural development and recognition–localization methods for end-effector of fruit–vegetable picking robots. Int J Adv Rob Syst 19(3):17298806221104906. https://doi.org/10.1177/17298806221104906
Li Y, Feng Q, Li T, Xie F, Liu C, Xiong Z (2022b) Advance of target visual information acquisition technology for fresh fruit robotic harvesting: a review. Agronomy 12(6):1336. https://doi.org/10.3390/agronomy12061336
Li R, Zhao S, Yang B (2023a) Research on the application status of machine vision technology in furniture manufacturing process. Appl Sci 13(4):2434. https://doi.org/10.3390/app13042434
Li T, Xie F, Zhao Z, Zhao H, Guo X, Feng Q (2023b) A multi-arm robot system for efficient apple harvesting: perception, task plan and control. Comput Electron Agric 211:107979. https://doi.org/10.1016/j.compag.2023.107979
Ling X, Zhao Y, Gong L, Liu C, Wang T (2019) Dual-arm cooperation and implementing for robotic harvesting tomato using binocular vision. Robot Auton Syst 114:134–143. https://doi.org/10.1016/j.robot.2019.01.019
Ling X, Liu JT, Liang CY, Wang XD (2021) Intelligent strawberry picking robot design and experiment. Mod Agric Equip 01:46–50
Liu TH, Luo G, Ehsani R, Toudeshki A, Zou XJ, Wang HJ (2018) Simulation study on the effects of tine-shaking frequency and penetrating depth on fruit detachment for citrus canopy-shaker harvesting. Comput Electron Agric 148:54–62. https://doi.org/10.1016/j.compag.2018.03.004
Liu J, Lin C, Guo SC, Peng ZY (2019) Design and study of a machine for picking citrus fruits. Packag Eng. https://doi.org/10.19554/j.cnki.1001-3563.2019.17.009
Liu G, Nouaze JC, Touko Mbouembe PL, Kim JH (2020) YOLO-tomato: a robust algorithm for tomato detection based on YOLOv3. Sensors 20(7):2145. https://doi.org/10.3390/s20072145
Liu S, Lou Y, Li Y, Zhang J, Li P, Yang B, Gu Q (2022) Review of phytochemical and nutritional characteristics and food applications of Citrus L. fruits. Front Nutr 9:968604. https://doi.org/10.3389/fnut.2022.968604
Liu Z, Zhang X, Sun Y, Zhou Y (2023a) Advanced controls on energy reliability, flexibility, resilience, and occupant-centric control for smart and energy-efficient buildings—a state-of-the-art review. Energy Build. https://doi.org/10.1016/j.enbuild.2023.113436
Liu Y, Zheng H, Zhang Y, Zhang Q, Chen H, Xu X, Wang G (2023b) “Is this blueberry ripe?”: a blueberry ripeness detection algorithm for use on picking robots. Front Plant Sci 14:1198650. https://doi.org/10.3389/fpls.2023.1198650
Lu R, Dickinson N, Lammers K, Zhang K, Chu P, Li Z (2022) Design and evaluation of end effectors for a vacuum-based robotic apple harvester. J ASABE. https://doi.org/10.13031/ja.14970
Mahmood MR, Matin MA, Sarigiannidis P, Goudos SK (2022) A comprehensive review on artificial intelligence/machine learning algorithms for empowering the future IoT toward 6G era. IEEE Access 10:87535–87562. https://doi.org/10.1109/ACCESS.2022.3199689
Mahmud NA, Sabil A, Hisham NN, Siraj S, Adnan NA, Amin NDM (2023) Sustainable living: alternative green structure module design for home self-food production. In: IOP conference series: earth and environmental science, vol 1205, no. 1. IOP Publishing, p 012085. https://doi.org/10.1088/1755-1315/1205/1/012085
Malik S, Muhammad K, Waheed Y (2023) Nanotechnology: a revolution in modern industry. Molecules 28(2):661. https://doi.org/10.3390/molecules28020661
Mao W, Liu Z, Liu H, Yang F, Wang M (2021) Research progress on synergistic technologies of agricultural multi-robots. Appl Sci 11(4):1448. https://doi.org/10.3390/app11041448
Maor Y (2022) Apparatus, systems and methods for harvesting and thinning using aerial drones for orchards, plantations and greenhouses: C N Patent Israel: CN109640621B [P]. 2022-05-27
Maor Y (2023) U.S. Patent No. 11,846,946. U.S. Patent and Trademark Office, Washington, DC
Mavridou E, Vrochidou E, Papakostas GA, Pachidis T, Kaburlasos VG (2019) Machine vision systems in precision agriculture for crop farming. J Imaging 5(12):89. https://doi.org/10.3390/jimaging5120089
Mu L, Liu Y, Cui Y, Liu H, Chen L, Fu L, Gejima Y (2017) Design of end-effector for kiwifruit harvesting robot experiment. In: 2017 ASABE annual international meeting. American Society of Agricultural and Biological Engineers, p 1. https://doi.org/10.13031/aim.201700666
Mu L, Cui G, Liu Y, Cui Y, Fu L, Gejima Y (2020) Design and simulation of an integrated end-effector for picking kiwifruit by robot. Inf Process Agric 7(1):58–71. https://doi.org/10.1016/j.inpa.2019.05.004
Navas E, Fernández R, Sepúlveda D, Armada M, Gonzalez-de-Santos P (2021) Soft grippers for automatic crop harvesting: a review. Sensors 21(8):2689. https://doi.org/10.3390/s21082689
Neupane C, Pereira M, Koirala A, Walsh KB (2023) Fruit sizing in orchard: a review from caliper to machine vision with deep learning. Sensors 23(8):3868. https://doi.org/10.3390/s23083868
Norton R, Claypool L, Leonard S, Adrian P, Fridley R, Charles F (1962) Mechanical harvesting of sweet cherries: 1961 tests show promise and problems. Calif Agric 16(5):8–10
Oliveira F, Tinoco V, Magalhães S, Santos FN, Silva MF (2022) End-effectors for harvesting manipulators-state of the art review. In: 2022 IEEE international conference on autonomous robot systems and competitions (ICARSC). IEEE, pp 98–103. https://doi.org/10.1109/ICARSC55462.2022.9784809
Pacheco A, Rehkugler GE (1980) Design and development of a spring activated impact shaker for apple harvesting. Trans ASAE 23(4):826–0830. https://doi.org/10.13031/2013.34671
Parsa S, Debnath B, Khan MA, AG E (2023) Modular autonomous strawberry picking robotic system. J Field Robot. https://doi.org/10.1002/rob.22229
Pathare PB, Opara UL, Al-Said FAJ (2013) Colour measurement and analysis in fresh and processed foods: a review. Food Bioprocess Technol 6:36–60. https://doi.org/10.1007/s11947-012-0867-9
Paturi UMR, Cheruku S (2021) Application and performance of machine learning techniques in manufacturing sector from the past two decades: a review. Mater Today Proc 38:2392–2401. https://doi.org/10.1016/j.matpr.2020.07.209
Pellerin RA, Millier WF, Lakso AN, Rehkugler GE, Throop JA, Allport TE (1978) Apple harvesting with an inertial vs. impulse trunk shaker on open-center and central-leader trees—part I. Trans ASAE 21(3):407–0413. https://doi.org/10.13031/2013.35314
Peng K, Ma W, Lu J, Tian Z, Yang Z (2023) Application of machine vision technology in citrus production. Appl Sci 13(16):9334. https://doi.org/10.3390/app13169334
Penumuru DP, Muthuswamy S, Karumbu P (2020) Identification and classification of materials using machine vision and machine learning in the context of industry 4.0. J Intell Manuf 31(5):1229–1241. https://doi.org/10.1007/s10845-019-01508-6
Pérez L, Rodríguez Í, Rodríguez N, Usamentiaga R, García DF (2016) Robot guidance using machine vision techniques in industrial environments: a comparative review. Sensors 16(3):335. https://doi.org/10.3390/s16030335
Peterson DL, Wolford SD, Timm E, Takeda F (1997) Fresh market quality blueberry harvester. Trans ASAE 40(3):535–540. https://doi.org/10.13031/2013.21298
Peterson DL, Whiting MD, Wolford SD (2003) Fresh–market quality tree fruit harvester part I: sweet cherry. Appl Eng Agric 19(5):539. https://doi.org/10.13031/2013.15313
Pezzi F, Martelli R (2015) Technical and economic evaluation of mechanical grape harvesting in flat and hill vineyards. Trans ASABE 58(2):297–303. https://doi.org/10.13031/trans.58.10997
Rehman A, Saba T, Kashif M, Fati SM, Bahaj SA, Chaudhry H (2022) A revisit of internet of things technologies for monitoring and control strategies in smart agriculture. Agronomy 12(1):127. https://doi.org/10.3390/agronomy12010127
Ren J, Wang Y (2022) Overview of object detection algorithms using convolutional neural networks. J Comput Commun 10(1):115–132. https://doi.org/10.4236/jcc.2022.101006
Ren Z, Fang F, Yan N, Wu Y (2022) State of the art in defect detection based on machine vision. Int J Precis Eng Manuf Green Technol 9(2):661–691. https://doi.org/10.1007/s40684-021-00343-6
Ren X, Huang B, Yin H (2023) A review of the large-scale application of autonomous mobility of agricultural platform. Comput Electron Agric 206:107628. https://doi.org/10.1016/j.compag.2023.107628
Riaz AR, Gilani SMM, Naseer S, Alshmrany S, Shafiq M, Choi JG (2022) Applying adaptive security techniques for risk analysis of internet of things (IoT)-based smart agriculture. Sustainability 14(17):10964. https://doi.org/10.3390/su141710964
Rong J, Wang P, Yang Q, Huang F (2021) A field-tested harvesting robot for oyster mushroom in greenhouse. Agronomy 11(6):1210. https://doi.org/10.3390/agronomy11061210
Rong J, Wang P, Wang T, Hu L, Yuan T (2022) Fruit pose recognition and directional orderly grasping strategies for tomato harvesting robots. Comput Electron Agric 202:107430. https://doi.org/10.1016/j.compag.2022.107430
Ronzhin A, Ngo T, Vu Q, Nguyen V, Ronzhin A, Ngo T et al (2022) Theoretical foundations to control technological and robotic operations with physical manipulations of agricultural products. In: Ground and air robotic manipulation systems in agriculture. pp 89–113. https://doi.org/10.1007/978-3-030-86826-0_5
Saleem MH, Potgieter J, Arif KM (2021) Automation in agriculture by machine and deep learning techniques: a review of recent developments. Precis Agric 22:2053–2091. https://doi.org/10.1007/s11119-021-09806-x
Saraiva R, Dias I, Grego J, Oliveira M (2023) Greenhouse tomato technologies and their influence in Mediterranean region. In: Tomato cultivation and consumption-innovation, sustainability and health. pp 1–27. https://doi.org/10.5772/intechopen.112273
Sarig Y (2012) Mechanical harvesting of fruit-past achievements, current status and future prospects. In: International symposium on mechanical harvesting and handling systems of fruits and nuts 965. pp 163–169. https://doi.org/10.17660/ActaHortic.2012.965.21
Sarkar P (2021) Use of shaking mechanism and robotic arm in fruit harvesting: a comprehensive review. J Crop Weed 17(2):01–09
Savary SU, Ehsani R, Salyani M, Hebel MA, Bora GC (2011) Study of force distribution in the citrus tree canopy during harvest using a continuous canopy shaker. Comput Electron Agric 76(1):51–58. https://doi.org/10.1016/j.compag.2011.01.005
Shaikh TA, Rasool T, Lone FR (2022) Towards leveraging the role of machine learning and artificial intelligence in precision agriculture and smart farming. Comput Electron Agric 198:107119. https://doi.org/10.1016/j.compag.2022.107119
Shang SQ, Li CP, He XN, Wang DW, Wang HQ, Yang S (2023) Design and testing of a vibratory picker for high acid apples. J Agric Mach. https://doi.org/10.13427/j.cnki.njyi.2023.07.001
Sharma A, Jain A, Gupta P, Chowdary V (2020) Machine learning applications for precision agriculture: a comprehensive review. IEEE Access 9:4843–4873. https://doi.org/10.1109/ACCESS.2020.3048415
Shekhar H, Seal S, Kedia S, Guha A (2020) Survey on applications of machine learning in the field of computer vision. In: Emerging technology in modelling and graphics: proceedings of IEM graph 2018. Springer Singapore, pp 667–678. https://doi.org/10.1007/978-981-13-7403-6_58
Sibhatu KT, Krishna VV, Qaim M (2015) Production diversity and dietary diversity in smallholder farm households. Proc Natl Acad Sci USA 112(34):10657–10662. https://doi.org/10.1073/pnas.1510982112
Silwal A (2016) Machine vision system for robotic apple harvesting in fruiting wall orchards. Washington State University, Pullman
Srivastava S, Divekar AV, Anilkumar C, Naik I, Kulkarni V, Pattabiraman V (2021) Comparative analysis of deep learning image detection algorithms. J Big Data 8(1):1–27. https://doi.org/10.1186/s40537-021-00434-w
Sudheer KP, Indira V (2007) Post harvest technology of horticultural crops, vol 7. New India Publishing, New Delhi
Sun X, Gu J, Tang S, Li J (2018) Research progress of visual inspection technology of steel products—a review. Appl Sci 8(11):2195. https://doi.org/10.3390/app8112195
Suryawanshi S, Baraskar T, Umbrani K, Chitnis A (2022) Using drone technology for fruit orchard management and waste reduction. In 2022 6th international conference on computing, communication, control and automation (ICCUBEA). IEEE, pp 1–9. https://doi.org/10.1109/ICCUBEA54992.2022.10010999
Syam N, Sharma A (2018) Waiting for a sales renaissance in the fourth industrial revolution: machine learning and artificial intelligence in sales research and practice. Ind Mark Manag 69:135–146. https://doi.org/10.1016/j.indmarman.2017.12.019
Tang YC, Zou XJ, He JH, Zhu LX, Chen MY, Gao YH (2020a) A visually positioned bionic snake-mouth harvesting mechanism and unmanned harvesting device: C N Patent CN211240896U [P]. 2020-08-14
Tang Y, Chen M, Wang C, Luo L, Li J, Lian G, Zou X (2020b) Recognition and localization methods for vision-based fruit picking robots: a review. Front Plant Sci 11:510. https://doi.org/10.3389/fpls.2020.00510
Tang B, Chen L, Sun W, Lin ZK (2023) Review of surface defect detection of steel products based on machine vision. IET Image Proc 17(2):303–322. https://doi.org/10.1049/ipr2.12647
Tao F, Cheng J, Qi Q, Zhang M, Zhang H, Sui F (2018) Digital twin-driven product design, manufacturing and service with big data. Int J Adv Manuf Technol 94:3563–3576. https://doi.org/10.1007/s00170-017-0233-1
Tensaw M (2020) Assessment of postharvest handling practices, loss and quality of mengo (Mangifera indica L.) ecotypes fruit in Harari region, Ethiopia (Doctoral Dissertation, Haramaya University)
Tian H, Wang T, Liu Y, Qiao X, Li Y (2020) Computer vision technology in agricultural automation—a review. Inf Process Agric 7(1):1–19. https://doi.org/10.1016/j.inpa.2019.09.006
Tian Z, Ma W, Yang Q, Duan F (2022a) Application status and challenges of machine vision in plant factory—a review. Inf Process Agric 9(2):195–211. https://doi.org/10.1016/j.inpa.2021.06.003
Tian C, Dong HL, Luo FS, Shi WT (2022b) Design of an aerial fruit picking and collection machine based on pneumatic transmission. Modern Manufacturing Technology and Equipment (01), pp 55–57. https://doi.org/10.16107/j.cnki.mmte.2022.0016
Tillett ND (1993) Robotic manipulators in horticulture: a review. J Agric Eng Res 55(2):89–105. https://doi.org/10.1006/jaer.1993.1035
Van Henten EJ (2019) Automation and robotics in greenhouses. In: Achieving sustainable greenhouse cultivation. Burleigh Dodds Science Publishing, pp 359–378
Varadaramanujan S, Sreenivasa S, Pasupathy P, Calastawad S, Morris M, Tosunoglu S (2017) Design of a drone with a robotic end-effector. In: Proceedings of the 30th Florida conference on recent advances in robotics, Boca Raton, FL, USA. pp 11–12
Verbiest R, Ruysen K, Vanwalleghem T, Demeester E, Kellens K (2021) Automation and robotics in the cultivation of pome fruit: where do we stand today? J Field Robot 38(4):513–531. https://doi.org/10.1002/rob.22000
Vougioukas SG (2019) Agricultural robotics. Annu Rev Control Robot Auton Syst 2:365–392. https://doi.org/10.1146/annurev-control-053018-023617
Vrochidou E, Tsakalidou VN, Kalathas I, Gkrimpizis T, Pachidis T, Kaburlasos VG (2022) An overview of end effectors in agricultural robotic harvesting systems. Agriculture 12(8):1240. https://doi.org/10.3390/agriculture12081240
Wakchaure M, Patle BK, Mahindrakar AK (2023) Application of AI techniques and robotics in agriculture: a review. Artif Intell Life Sci. https://doi.org/10.1016/j.ailsci.2023.100057
Walker RJ (2016) Population growth and its implications for global security. Am J Econ Sociol 75(4):980–1004. https://doi.org/10.1111/ajes.12161
Wan S, Goudos S (2020) Faster R-CNN for multi-class fruit detection using a robotic vision system. Comput Netw 168:107036. https://doi.org/10.1016/j.comnet.2019.107036
Wang J, Chortos A (2022) Control strategies for soft robot systems. Adv Intell Syst 4(5):2100165. https://doi.org/10.1002/aisy.202100165
Wang L, Liu M (2020) Path tracking control for autonomous harvesting robots based on improved double arc path planning algorithm. J Intell Rob Syst 100:899–909. https://doi.org/10.1007/s10846-020-01257-2
Wang C, Tang Y, Zou X, SiTu W, Feng W (2017) A robust fruit image segmentation algorithm against varying illumination for vision system of fruit harvesting robot. Optik 131:626–631. https://doi.org/10.1016/j.ijleo.2016.11.177
Wang Y, Yang Y, Yang C, Zhao H, Chen G, Zhang Z et al (2019) End-effector with a bite mode for harvesting citrus fruit in random stalk orientation environment. Comput Electron Agric 157:454–470. https://doi.org/10.1016/j.compag.2019.01.015
Wang Z, Xun Y, Wang Y, Yang Q (2022) Review of smart robots for fruit and vegetable picking in agriculture. Int J Agric Biol Eng 15(1):33–54. https://doi.org/10.25165/j.ijabe.20221501.7232
Wang C, Wang C, Wang L, Wang J, Liao J, Li Y, Lan Y (2023) A lightweight cherry tomato maturity real-time detection algorithm based on improved YOLOV5n. Agronomy 13(8):2106. https://doi.org/10.3390/agronomy13082106
Whitney JD, Patterson JM (1972) Development of a citrus removal device using oscillating forced air. Trans ASAE 15(5):849–855. https://doi.org/10.13031/2013.38024
Whitney JD, Churchill DB, Hedden SL, Smerage GH (1988) Trunk shakers for citrus harvesting—part I: measured trunk shaker and tree trunk motion. Appl Eng Agric 4(2):93–101. https://doi.org/10.13031/2013.26588
Williams HA, Jones MH, Nejati M, Seabright MJ, Bell J, Penhall ND, MacDonald BA (2019) Robotic kiwifruit harvesting using machine vision, convolutional neural networks, and robotic arms. Biosyst Eng 181:140–156. https://doi.org/10.1016/j.biosystemseng.2019.03.007
Williams H, Ting C, Nejati M, Jones MH, Penhall N, Lim J et al (2020) Improvements to and large-scale evaluation of a robotic kiwifruit harvester. J Field Robot 37(2):187–201. https://doi.org/10.1002/rob.21890
Woo S, Uyeh DD, Kim J, Kim Y, Kang S, Kim KC et al (2020) Analyses of work efficiency of a strawberry-harvesting robot in an automated greenhouse. Agronomy 10(11):1751. https://doi.org/10.3390/agronomy10111751
Wu Z, Chen Y, Zhao B, Kang X, Ding Y (2021) Review of weed detection methods based on computer vision. Sensors 21(11):3647. https://doi.org/10.3390/s21113647
Wu D, Ding D, Cui B, Jiang S, Zhao E, Liu Y, Cao C (2022) Design and experiment of vibration plate type camellia fruit picking machine. Int J Agric Biol Eng 15(4):130–138. https://doi.org/10.25165/j.ijabe.20221504.6971
Xiao X, Wang Y, Jiang Y (2023) Review of research advances in fruit and vegetable harvesting robots. J Electr Eng Technol. https://doi.org/10.1007/s42835-023-01596-8
Xie D, Chen L, Liu L, Chen L, Wang H (2022a) Actuators and sensors for application in agricultural robots: a review. Machines 10(10):913. https://doi.org/10.3390/machines10100913
Xie J, Wu JZ, Li YG, Liang JT (2022b) A review of research on picking drones. Manuf Autom 10:72–75
Xiong J, Lin R, Liu Z, He Z, Tang L, Yang Z, Zou X (2018) The recognition of litchi clusters and the calculation of picking point in a nocturnal natural environment. Biosyst Eng 166:44–57. https://doi.org/10.1016/j.biosystemseng.2017.11.005
Xiong Y, Peng C, Grimstad L, From PJ, Isler V (2019) Development and field evaluation of a strawberry harvesting robot with a cable-driven gripper. Comput Electron Agric 157:392–402. https://doi.org/10.1016/j.compag.2019.01.009
Xiong Y, Ge Y, Grimstad L, From PJ (2020) An autonomous strawberry-harvesting robot: design, development, integration, and field evaluation. J Field Robot 37(2):202–224. https://doi.org/10.1002/rob.21889
Xu Q, Meng FB, Qu Y, Li AM, Xia K, Zhang YY, Xu HY (2021) A push-cut cherry picking device and fruit picking device: C N Patent CN112772141A [P]. 2021-05-11
Xu HQ, Li Y, Zhang JJ (2023) Research on fruit characteristics and mechanised harvesting of winter dates. China Agric Chem News. https://doi.org/10.13733/j.jcam.issn.2095-5553.2023.02.008
Yaguchi H, Nagahama K, Hasegawa T, Inaba M (2016). Development of an autonomous tomato harvesting robot with rotational plucking gripper. In: 2016 IEEE/RSJ international conference on intelligent robots and systems (IROS). IEEE, pp 652–657. https://doi.org/10.1109/IROS.2016.7759122
Yan SZ, Zhang YC, Zhao C, Jan Y, Chai YS (2021) A fruit tree picking drone: C N Patent CN214902269U [P]. 2021-11-30
Yang HW, Zhang LY (2014) Research on the development of agricultural mechanical automation in mechanical engineering. Appl Mech Mater 454:23–26. https://doi.org/10.4028/www.scientific.net/AMM.454.23
Yang CH, Liu YP, Wang Y, Xiong LY, Xu HB, Zhao WH (2019) Research on recognition and positioning system of citrus picking robot under natural environment. J Agric Mach. https://doi.org/10.6041/i.issn.1000-1298.2019.12.002
Yang L, Liu Y, Yu H, Fang X, Song L, Li D, Chen Y (2021) Computer vision models in intelligent aquaculture with emphasis on fish detection and behavior analysis: a review. Arch Comput Methods in Eng 28:2785–2816. https://doi.org/10.1007/s11831-020-09486-2
Yang Y, Han Y, Li S, Yang Y, Zhang M, Li H (2023a) Vision based fruit recognition and positioning technology for harvesting robots. Comput Electron Agric 213:108258. https://doi.org/10.1016/j.compag.2023.108258
Yang HY, Huang WJ, Li YH, Duan XG, Kang SY, Wang ZY et al (2023b) Intelligent recognition and harvesting of citrus based on Raspberry Pi platform. J Cent South Univ For Sci Technol. https://doi.org/10.14067/j.cnki.1673-923x.2023.08.019
Yarborough DE, Hergeri GB (2010) Mechanical harvesting of berry crops. Hortic Rev 16:255–282
Yasukawa S, Li B, Sonoda T, Ishii K (2017) Development of a tomato harvesting robot. In: 2017 International conference on artificial life and robotics (ICAROB), Miyazaki. pp 408–411
Yin XC (2020) Design and optimization of vibratory walnut picking machine (Master’s Thesis, Harbin University of Commerce). https://doi.org/10.27787/d.cnki.ghrbs.2020.000338
Yin H, Sun Q, Ren X, Guo J, Yang Y, Wei Y et al (2023) Development, integration, and field evaluation of an autonomous citrus-harvesting robot. J Field Robot. https://doi.org/10.1002/rob.22178
Yoshida T, Onishi Y, Kawahara T, Fukao T (2022) Automated harvesting by a dual-arm fruit harvesting robot. Robomech J 9(1):1–14. https://doi.org/10.1186/s40648-022-00233-9
Yu CP, Ampazidis Y (2022) Design and testing of a push-cut cherry picker. China Agric Chem News. https://doi.org/10.13733/j.jcam.issn.2095-5553.2022.10.002
Yu P, Li C, Takeda F, Krewer G, Rains G, Hamrita T (2014) Measurement of mechanical impacts created by rotary, slapper, and sway blueberry mechanical harvesters. Comput Electron Agric 101:84–92. https://doi.org/10.1016/j.compag.2013.12.001
Yu Y, Zhang K, Liu H, Yang L, Zhang D (2020) Real-time visual localization of the picking points for a ridge-planting strawberry harvesting robot. IEEE Access 8:116556–116568. https://doi.org/10.1109/ACCESS.2020.3003034
Yu FH, Zhou CQ, Yang X, Guo ZF, Chen CL (2022) Design and experimentation of a tomato picking robot for daylight greenhouses. J Agric Mach 01:41–49. https://doi.org/10.6041/j.issn.1000-1298.2022.01.005
Yuan ZY, Chen J (2014) Analysis of the development status, problems and countermeasures of fruit and vegetable picking robots. Sichuan Agric Agric Mach 06:16–18
Yuan PP, Zhu XL, You J, Han CJ, Zhang XJ, Guo H (2020) Development of a crankshaft vibratory threshing and harvesting device for wine grapes. J Agric Eng. https://doi.org/10.11975/j.issn.1002-6819.2020.09.008
Yuan YW, Bai SH, Niu K, Zhou LM, Zhao B, Wei LG, Xiong S, Liu LJ (2022) Research progress on mechanized harvesting technology and equipment for forest fruits. J Agric Eng. https://doi.org/10.11975/j.issn.1002-6819.2022.09.006
Zhang Z, Xiao H, Ding W, Mei S (2015) Mechanism simulation analysis and prototype experiment of Lycium barbarum harvest by vibration mode. Trans Chin Soc Agric Eng 31(10):20–28. https://doi.org/10.11975/j.issn.1002-6819.2015.10.003
Zhang Z, Pothula AK, Lu R (2018) A review of bin filling technologies for apple harvest and postharvest handling. Appl Eng Agric 34(4):687–703. https://doi.org/10.13031/aea.12827
Zhang Q, Karkee M, Tabb A (2019) The use of agricultural robots in orchard management. arXiv Preprint. https://arxiv.org/abs/1907.13114. https://doi.org/10.19103/AS.2019.0056.14
Zhang B, Xie Y, Zhou J, Wang K, Zhang Z (2020) State-of-the-art robotic grippers, grasping and control strategies, as well as their applications in agricultural robots: a review. Comput Electron Agric 177:105694. https://doi.org/10.1016/j.compag.2020.105694
Zhang K, Lammers K, Chu P, Li Z, Lu R (2021a) System design and control of an apple harvesting robot. Mechatronics 79:102644. https://doi.org/10.1016/j.mechatronics.2021.102644
Zhang FK, Ran JH, Li ZJ, Wang DW, Li P (2021b) Optimization of operational parameters for an air-priming floor picker for red dates. J Fruit Trees. https://doi.org/10.13925/j.cnki.gsxb.20200573
Zhang Y, Li P, Quan J, Li L, Zhang G, Zhou D (2023) Progress, challenges, and prospects of soft robotics for space applications. Adv Intell Syst 5(3):2200071. https://doi.org/10.1002/aisy.202200071
Zhao L (2022) Design and experimental study of a spiral comb apple harvester (Master’s Thesis, Northwest Agriculture and Forestry University). https://doi.org/10.27409/d.cnki.gxbnu.2022.002057
Zhao Y, Gong L, Huang Y, Liu C (2016) A review of key techniques of vision-based control for harvesting robot. Comput Electron Agric 127:311–323. https://doi.org/10.1016/j.compag.2016.06.022
Zhao Y, Yang C, Bao Y, Liu X, Guo Y (2018) Research status and development trend of the contact vibration small berry harvester. In: Advances in mechanical design: proceedings of the 2017 international conference on mechanical design (ICMD 2017). Springer Singapore, pp 969–984. https://doi.org/10.1007/978-981-10-6553-8_65
Zhao J, Yang Y, Zheng H, Dong Y (2020) Global agricultural robotics research and development: trend forecasts. J Phys Conf Ser 1693(1):012227. https://doi.org/10.1088/1742-6596/1693/1/012227
Zhou T, Zhang D, Zhou M, Xi H, Chen X (2018) System design of tomatoes harvesting robot based on binocular vision. In: 2018 Chinese automation congress (CAC). IEEE, pp 1114–1118. https://doi.org/10.1109/CAC.2018.8623150
Zhou L, Ren L, Chen Y, Niu S, Han Z, Ren L (2021) Bio-inspired soft grippers based on impactive gripping. Adv Sci 8(9):2002017. https://doi.org/10.1002/advs.202002017
Zhou H, Wang X, Au W, Kang H, Chen C (2022) Intelligent robots for fruit harvesting: recent developments and future challenges. Precis Agric 23(5):1856–1907. https://doi.org/10.1007/s11119-022-09913-3
Zhu Y, Ling ZG, Zhang YQ (2020) Advances and prospects in machine vision technology. J Graphol. https://doi.org/10.11996/JG.j.2095-302X.2020060871
Zhu QF, Lu RJ, Liu B, Lu J, Sun W (2021a) Research status and development trend of walnut picking machinery. Forestry and Grassland Machinery (01), pp. 45–53. https://doi.org/10.13594/j.cnki.mcjgjx.2021.01.010
Zhu QF, Lu RJ, Li FS (2021b) Research status and development trend of apple picking machinery. Forestry Machinery and Woodworking Equipment (05), pp 4–9+15. https://doi.org/10.13279/j.cnki.fmwe.2021.0053
Zhu L, Spachos P, Pensini E, Plataniotis KN (2021c) Deep learning and machine vision for food processing: a survey. Curr Res Food Sci 4:233–249. https://doi.org/10.1016/j.crfs.2021.03.009
Zhu H, Qi XK, Dong JS (2021d) An autonomous lychee picking device for drones: C N Patent CN113716057A [P]. 2021-11-30
Zhu XL, Yuan PP, You J, Han CJ (2023) Design and testing of a vibratory harvesting device for wine grapes based on the 4R space mechanism. Agricultural mechanization research design and testing of a vibratory harvesting device for wine grapes based on the 4R space mechanism. Agric Mech Res. https://doi.org/10.13427/j.cnki.njyi.2023.07.001
Zhuang M, Li G, Ding K (2023) Obstacle avoidance path planning for apple picking robotic arm incorporating artificial potential field and A* algorithm. IEEE Access. https://doi.org/10.1109/ACCESS.2023.3312763
Zou X, Zou H, Lu J (2012) Virtual manipulator-based binocular stereo vision positioning system and errors modelling. Mach Vis Appl 23:43–63. https://doi.org/10.1007/s00138-010-0291-y
Acknowledgements
We thank the support from the National Natural Science Foundation of China, the Joint Fund of Natural Science Foundation of Ningxia Hui Autonomous Region, the research on the heat and mass transfer mechanism of fresh wolfberry with stalk in vacuum pre-cooling, and the 2023 Ningxia University Postgraduate Innovation Programme. We also sincerely thank the anonymous reviewers for their critical comments on the manuscript and suggestions for improvement.
Funding
This study was supported by the National Natural Science Foundation of China (Grant No. 32260617), the Joint Fund of Natural Science Foundation of Ningxia Hui Autonomous Region (Grant No. 2022AAC03022), the Joint Fund of Natural Science Foundation of Ningxia Hui Autonomous Region (Grant No. 2023AAC03065), and the Postgraduate Innovation Project of Ningxia University 2023 (Grant No. CXXM202363).
Author information
Authors and Affiliations
Contributions
Jun Zhang: writing-original draft; software; resources; conceptualization; data curation. Ningbo Kang: guidance and financial support; project administration. Qianjin Qu: writing-review & editing; software; resources. Hongbo Zhang: writing-review & editing; visualization. Lianghuan Zhou: writing-review & editing; software; visualization.
Corresponding author
Ethics declarations
Competing interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Zhang, J., Kang, N., Qu, Q. et al. Automatic fruit picking technology: a comprehensive review of research advances. Artif Intell Rev 57, 54 (2024). https://doi.org/10.1007/s10462-023-10674-2
Accepted:
Published:
DOI: https://doi.org/10.1007/s10462-023-10674-2