Abstract
The recent development of faster and more accurate deep learning models has enabled researchers to utilize the potential of deep learning in robotics. Convolutional neural networks used for the process of semantic segmentation are being applied to improve the traditional robotic tasks by adding an additional level of intelligence, through the execution of context-aware tasks. Having that in mind, visual servoing can now be performed in a completely new manner, by exploiting only semantic and geometric knowledge about the environment. To carry out visual servoing, the mathematical model of the error between the images generated at the current and the desired mobile robot pose (i.e. position and orientation) in the image space needs to be adequately defined. In this paper, we propose the novel mathematical model for the weighted fitness function evaluation, which is utilized for the image registration process within the visual servoing framework. By weighting the classes by their importance in the desired image, the convergence domain of the initial error in the visual servoing process can be greatly extended. The experimental evaluation is carried out on the mobile robot RAICO (Robot with Artificial Intelligence based COgnition), where it is shown that weighted fitness function enables more robust intelligent visual servoing systems with a lower possibility of failure, easier real-world implementation, and feasible object driven navigation.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Mitić, M., Miljković, Z.: Neural network learning from demonstration and epipolar geometry for visual control of a nonholonomic mobile robot. Soft Comput. 18(5), 1011–1025 (2014)
Sünderhauf, N., Brock, O., Scheirer, W., Hadsell, R., Fox, D., Leitner, J., Upcroft, B., Abbeel, P., Burgard, W., Milford, M., Corke, P.: The limits and potentials of deep learning for robotics. Int. J. Robot. Res. 37(4–5), 405–420 (2018)
Bateux, Q., Marchand, E., Leitner, J., Chaumette, F.: Training deep neural networks for visual servoing. In: IEEE International Conference on Robotics and Automation (ICRA), pp. 1–8 (2018)
Sadeghi, F., Toshev, A., Jang, E., Levine, S.: Sim2Real viewpoint invariant visual servoing by recurrent control. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4691–4699 (2018)
Petrović, M., Mystkowski, A., Jokić, A., Đokić, L., Miljković, Z.: Deep Learning-based Algorithm for Mobile Robot Control in Textureless Environment. In: IEEE International Conference Mechatronic Systems and Materials (MSM), pp. 1–4 (2020)
Đokić, L., Jokić, A., Petrović, M., Miljković, Z.: Biologically inspired optimization methods for image registration in visual servoing of a mobile robot. In: 7th International Conference on Electrical, Electronics and Computing Engineering (IcETRAN 2020), 715–720 (2020)
Indraswari, R., Kurita, T., Arifin, A.Z., Suciati, N., Astuti, E.R.: Multi-projection deep learning network for segmentation of 3D medical images. Pattern Recogn. Lett. 125, 791–797 (2019)
Goli, P.: A new perceptually weighted cost function in deep neural network based speech enhancement systems. Hearing Balance Commun. 17(3), 191–196 (2019)
Cao, F., Che, S., Zhao, J.: A weighted hybrid training algorithm of neural networks for robust data regression. Int. J. Mach. Intell. Sensory Signal Process. 2(1), 51–66 (2017)
Acknowledgment
This work has been financially supported by the Ministry of Education, Science and Technological Development of the Serbian Government, through the project “Integrated research in macro, micro, and nano mechanical engineering – Deep learning of intelligent manufacturing systems in production engineering” (contract No. 451-03-9/2021-14/200105), by the Science Fund of the Republic of Serbia, grant No. 6523109, AI - MISSION4.0, 2020–2022, and by the Polish National Agency for Academic Exchange through the project: “Biologically inspired optimization algorithms for control and scheduling of intelligent robotic systems”, Grant No. PPN/ULM/2019/1/00354/U/00001.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Jokić, A., Petrović, M., Kulesza, Z., Miljković, Z. (2021). Visual Deep Learning-Based Mobile Robot Control: A Novel Weighted Fitness Function-Based Image Registration Model. In: Karabegović, I. (eds) New Technologies, Development and Application IV. NT 2021. Lecture Notes in Networks and Systems, vol 233. Springer, Cham. https://doi.org/10.1007/978-3-030-75275-0_82
Download citation
DOI: https://doi.org/10.1007/978-3-030-75275-0_82
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-75274-3
Online ISBN: 978-3-030-75275-0
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)