Abstract
This paper deals with the search-and-rescue tasks of a mobile robot with multiple interesting targets in an unknown dynamic environment. The problem is challenging because the mobile robot needs to search for multiple targets while avoiding obstacles simultaneously. To ensure that the mobile robot avoids obstacles properly, we propose a mixed-strategy Nash equilibrium based Dyna-Q (MNDQ) algorithm. First, a multi-objective layered structure is introduced to simplify the representation of multiple objectives and reduce computational complexity. This structure divides the overall task into subtasks, including searching for targets and avoiding obstacles. Second, a risk-monitoring mechanism is proposed based on the relative positions of dynamic risks. This mechanism helps the robot avoid potential collisions and unnecessary detours. Then, to improve sampling efficiency, MNDQ is presented, which combines Dyna-Q and mixed-strategy Nash equilibrium. By using mixed-strategy Nash equilibrium, the agent makes decisions in the form of probabilities, maximizing the expected rewards and improving the overall performance of the Dyna-Q algorithm. Furthermore, a series of simulations are conducted to verify the effectiveness of the proposed method. The results show that MNDQ performs well and exhibits robustness, providing a competitive solution for future autonomous robot navigation tasks.
摘要
本文研究未知动态环境下具有多个兴趣目标的移动机器人搜救任务问题。由于移动机器人需要搜救多个目标并避开障碍, 此类问题具有挑战性。为确保移动机器人合理避碰, 本文提出一种基于混合策略纳什均衡的Dyna-Q算法(MNDQ)。首先, 引入一种多目标分层结构以简化问题, 该结构将整个任务划分为多个子任务, 包括搜索目标和躲避障碍。其次, 提出基于动态风险相对位置的风险监测机制, 使机器人避免潜在碰撞和绕路。此外, 为提高采样效率, 提出了结合Dyna-Q和混合策略纳什均衡的强化学习方法(MNDQ)。根据混合策略纳什均衡, 智能体以概率的形式做出决策从而最大化期望回报, 提高Dyna-Q算法的整体性能。最后, 通过仿真实验验证所提方法的有效性。结果表明, 该方法具有良好的表现并为未来的机器人自主导航任务提供了解决思路。
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
Aggarwal S, Kumar N, 2020. Path planning techniques for unmanned aerial vehicles: a review, solutions, and challenges. Comput Commun, 149:270–299. https://doi.org/10.1016/j.comcom.2019.10.014
Banerjee C, Datta D, Agarwal A, 2015. Chaotic patrol robot with frequency constraints. Proc IEEE Int Conf on Research in Computational Intelligence and Communication Networks, p.340–344. https://doi.org/10.1109/ICRCICN.2015.7434261
Brito B, Floor B, Ferranti L, et al., 2019. Model predictive contouring control for collision avoidance in unstructured dynamic environments. IEEE Robot Autom Lett, 4(4):4459–4466. https://doi.org/10.1109/LRA.2019.2929976
Brito B, Everett M, How JP, et al., 2021. Where to go next: learning a subgoal recommendation policy for navigation in dynamic environments. IEEE Robot Autom Lett, 6(3):4616–4623. https://doi.org/10.1109/LRA.2021.3068662
Brockman G, Cheung V, Pettersson L, et al., 2016. OpenAI Gym. https://arxiv.org/abs/1606.01540
Chiu ZY, Richter F, Funk EK, et al., 2021. Bimanual regrasping for suture needles using reinforcement learning for rapid motion planning. Proc IEEE Int Conf on Robotics and Automation, p.7737–7743. https://doi.org/10.1109/ICRA48506.2021.9561673
Curiac DI, Banias O, Volosencu C, et al., 2018. Novel bioinspired approach based on chaotic dynamics for robot patrolling missions with adversaries. Entropy, 20(5):378. https://doi.org/10.3390/e20050378
Dong YS, Zou XJ, 2020. Mobile robot path planning based on improved DDPG reinforcement learning algorithm. Proc IEEE 11th Int Conf on Software Engineering and Service Science, p.52–56. https://doi.org/10.1109/ICSESS49938.2020.9237641
Faust A, Oslund K, Ramirez O, et al., 2018. PRM-RL: long-range robotic navigation tasks by combining reinforcement learning and sampling-based planning. Proc IEEE Int Conf on Robotics and Automation, p.5113–5120. https://doi.org/10.1109/ICRA.2018.8461096
Fudenberg D, Tirole J, 1991. Game Theory. MIT Press, Cambridge, USA.
Gaertner M, Bjelonic M, Farshidian F, et al., 2021. Collisionfree MPC for legged robots in static and dynamic scenes. Proc IEEE Int Conf on Robotics and Automation, p.8266–8272. https://doi.org/10.1109/ICRA48506.2021.9561326
Geng N, Meng QG, Gong DW, et al., 2019. How good are distributed allocation algorithms for solving urban search and rescue problems? A comparative study with centralized algorithms. IEEE Trans Autom Sci Eng, 16(1):478–485. https://doi.org/10.1109/TASE.2018.2866395
Greenwald A, Hall K, 2003. Correlated-Q-learning. Proc 20th Int Conf on Machine Learning, p.242–249.
Gregor M, Nemec D, Janota A, et al., 2018. A visual attention operator for playing Pac-Man. Proc ELEKTRO, p.1–6. https://doi.org/10.1109/elektro.2018.8398308
Hayamizu Y, Amiri S, Chandan K, et al., 2021. Guiding robot exploration in reinforcement learning via automated planning. Proc 31st Int Conf on Automated Planning and Scheduling, p.625–633. https://doi.org/10.1609/icaps.v31i1.16011
Hong LB, Wang Y, Du YC, et al., 2021. UAV search-and-rescue planning using an adaptive memetic algorithm. Front Inform Technol Electron Eng, 22(11):1477–1491. https://doi.org/10.1631/FITEE.2000632
Hu JL, Wellman MP, 2003. Nash Q-learning for general-sum stochastic games. J Mach Learn Res, 4:1039–1069.
Hubert T, Schrittwieser J, Antonoglou I, et al., 2021. Learning and planning in complex action spaces. Proc 38th Int Conf on Machine Learning, p.4476–4486.
Hwang KS, Lin JL, Huang HL, 2011. Dynamic patrol planning in a cooperative multi-robot system. Proc 14th FIRA RoboWorld Congress, p.116–123. https://doi.org/10.1007/978-3-642-23147-6_14
Hwang KS, Jiang WC, Chen YJ, 2015. Model learning and knowledge sharing for a multiagent system with Dyna-Q learning. IEEE Trans Cybern, 45(5):978–990. https://doi.org/10.1109/TCYB.2014.2341582
Jaderberg M, Czarnecki WM, Dunning I, et al., 2019. Human-level performance in 3D multiplayer games with population-based reinforcement learning. Science, 364(6443):859–865. https://doi.org/10.1126/science.aau6249
Lei XY, Zhang Z, Dong PF, 2018. Dynamic path planning of unknown environment based on deep reinforcement learning. J Rob, 2018:5781591. https://doi.org/10.1155/2018/5781591
Li CH, Fang C, Wang FY, et al., 2019. Complete coverage path planning for an Arnold system based mobile robot to perform specific types of missions. Front Inform Technol Electron Eng, 20(11):1530–1542. https://doi.org/10.1631/FITEE.1800616
Li HQ, Huang J, Cao Z, et al., 2023. Stochastic pedestrian avoidance for autonomous vehicles using hybrid reinforcement learning. Front Inform Technol Electron Eng, 24(1):131–140. https://doi.org/10.1631/FITEE.2200128
Li HR, Zhang QC, Zhao DB, 2020. Deep reinforcement learning-based automatic exploration for navigation in unknown environment. IEEE Trans Neur Netw Learn Syst, 31(6):2064–2076. https://doi.org/10.1109/TNNLS.2019.2927869
Li ZR, Lu C, Yi YT, et al., 2022. A hierarchical framework for interactive behaviour prediction of heterogeneous traffic participants based on graph neural network. IEEE Trans Intell Transp Syst, 23(7):9102–9114. https://doi.org/10.1109/TITS.2021.3090851
Liu SJ, Tong XR, 2021. Urban transportation path planning based on reinforcement learning. J Comput Appl, 41(1):185–190 (in Chinese). https://doi.org/10.11772/j.issn.1001-9081.2020060949
Liu YY, Yan SH, Zhao Y, et al., 2022. Improved Dyna-Q: a reinforcement learning method focused via heuristic graph for AGV path planning in dynamic environments. Drones, 6(11):365. https://doi.org/10.3390/drones6110365
Liu Z, Cao YQ, Chen JY, et al., 2023. A hierarchical reinforcement learning algorithm based on attention mechanism for UAV autonomous navigation. IEEE Trans Intell Transp Syst, 24(11):13309–13320. https://doi.org/10.1109/TITS.2022.3225721
Lu YL, Yan K, 2020. Algorithms in multi-agent systems: a holistic perspective from reinforcement learning and game theory. https://arxiv.org/abs/2001.06487
Lu YM, Kamgarpour M, 2020. Safe mission planning under dynamical uncertainties. Proc IEEE Int Conf on Robotics and Automation, p.2209–2215. https://doi.org/10.1109/ICRA40945.2020.9196515
Luo GY, Wang YT, Zhang H, et al., 2023. AlphaRoute: large-scale coordinated route planning via Monte Carlo tree search. Proc 37th AAAI Conf on Artificial Intelligence, p.12058–12067. https://doi.org/10.1609/aaai.v37i10.26422
Martins-Filho LS, Macau EEN, 2007. Patrol mobile robots and chaotic trajectories. Math Probl Eng, 2007:061543. https://doi.org/10.1155/2007/61543
McGuire KN, de Croon GCHE, Tuyls K, 2019. A comparative study of bug algorithms for robot navigation. Robot Auton Syst, 121:103261. https://doi.org/10.1016/j.robot.2019.103261
Mirchevska B, Hügle M, Kalweit G, et al., 2021. Amortized Q-learning with model-based action proposals for autonomous driving on highways. Proc IEEE Int Conf on Robotics and Automation, p.1028–1035. https://doi.org/10.1109/ICRA48506.2021.9560777
Nasar W, da Silva Torres R, Gundersen OE, et al., 2023. The use of decision support in search and rescue: a systematic literature review. ISPRS Int J Geo-Inform, 12(5):182. https://doi.org/10.3390/ijgi12050182
Nash JFJr, 1950. Equilibrium points in n-person games. Proc Natl Acad Sci USA, 36(1):48–49. https://doi.org/10.1073/pnas.36.1.48
Ng J, Bräunl T, 2007. Performance comparison of bug navigation algorithms. J Intell Robot Syst, 50(1):73–84. https://doi.org/10.1007/s10846-007-9157-6
Niroui F, Sprenger B, Nejat G, 2017. Robot exploration in unknown cluttered environments when dealing with uncertainty. Proc IEEE Int Symp on Robotics and Intelligent Sensors, p.224–229. https://doi.org/10.1109/IRIS.2017.8250126
Ohnishi M, Wang L, Notomista G, et al., 2019. Barrier-certified adaptive reinforcement learning with applications to brushbot navigation. IEEE Trans Robot, 35(5):1186–1205. https://doi.org/10.1109/TRO.2019.2920206
Osborne M, Rubinstein A, 1994. A Course in Game Theory. MIT Press, Cambridge, USA.
Padakandla S, 2021. A survey of reinforcement learning algorithms for dynamically varying environments. ACM Comput Surv, 54(6):127. https://doi.org/10.1145/3459991
Patle BK, Babu LG, Pandey A, et al., 2019. A review: on path planning strategies for navigation of mobile robot. Def Technol, 15(4):582–606. https://doi.org/10.1016/j.dt.2019.04.011
Pei M, An H, Liu B, et al., 2022. An improved Dyna-Q algorithm for mobile robot path planning in unknown dynamic environment. IEEE Trans Syst Man Cybern Syst, 52(7):4415–4425. https://doi.org/10.1109/TSMC.2021.3096935
Prado J, Marques L, 2014. Energy efficient area coverage for an autonomous demining robot. Proc 1st Iberian Robotics Conf, p.459–471. https://doi.org/10.1007/978-3-319-03653-3_34
Puterman ML, 1990. Markov decision processes. Handb Oper Res Manage Sci, 2:331–434. https://doi.org/10.1016/S0927-0507(05)80172-0
Rosenthal RW, 1973. A class of games possessing pure-strategy Nash equilibria. Int J Game Theory, 2(1):65–67. https://doi.org/10.1007/BF01737559
Roughgarden T, 2010. Algorithmic game theory. Commun ACM, 53(7):78–86. https://doi.org/10.1145/1785414.1785439
Roughgarden T, 2016. Twenty Lectures on Algorithmic Game Theory. Cambridge University Press, New York, USA. https://doi.org/10.1017/CBO9781316779309
Shi HB, Yang SK, Hwang KS, et al., 2018. A sample aggregation approach to experiences replay of Dyna-Q learning. IEEE Access, 6:37173–37184. https://doi.org/10.1109/ACCESS.2018.2847048
Sutton RS, Barto AG, 1999. Reinforcement learning. J Cognit Neurosci, 11(1):126–134. https://doi.org/10.1162/089892999563184
Sutton RS, Barto AG, 2018. Reinforcement Learning: an Introduction (2nd Ed.). MIT Press, Cambridge, USA.
Wakayama S, Ahmed NR, 2020. Auto-tuning online POMDPs for multi-object search in uncertain environments. Proc AIAA Scitech Forum. https://doi.org/10.2514/6.2020-0391
Wang BY, Liu Z, Li QB, et al., 2020. Mobile robot path planning in dynamic environments through globally guided reinforcement learning. IEEE Robot Autom Lett, 5(4):6932–6939. https://doi.org/10.1109/LRA.2020.3026638
Wu JF, Braverman V, Yang L, 2021. Accommodating picky customers: regret bound and exploration complexity for multi-objective reinforcement learning. Proc 35th Int Conf on Neural Information Processing Systems, p.13112–13124.
Wu YX, Li XJ, Liu JJ, et al., 2019. Switch-based active deep Dyna-Q: efficient adaptive planning for task-completion dialogue policy learning. Proc 33rd AAAI Conf on Artificial intelligence, p.7289–7296. https://doi.org/10.1609/aaai.v33i01.33017289
Wyrąbkiewicz K, Tarczewski T, Niewiara ł, 2020. Local path planning for autonomous mobile robot based on APF-BUG algorithm with ground quality indicator. In: Bartoszewicz A, Kabziński J, Kacprzyk J (Eds.), Advanced, Contemporary Control. Springer, Cham, p.979–990. https://doi.org/10.1007/978-3-030-50936-1_82
Yu Y, Tang J, Huang JY, et al., 2021. Multi-objective optimization for UAV-assisted wireless powered IoT networks based on extended DDPG algorithm. IEEE Trans Commun, 69(9):6361–6374. https://doi.org/10.1109/TCOMM.2021.3089476
Zhang YH, Chai ZJ, Lykotrafitis G, 2021. Deep reinforcement learning with a particle dynamics environment applied to emergency evacuation of a room with obstacles. Phys A Stat Mech Appl, 571:125845. https://doi.org/10.1016/j.physa.2021.125845
Zheng KY, Sung Y, Konidaris G, et al., 2021. Multiresolution POMDP planning for multi-object search in 3D. Proc IEEE/RSJ Int Conf on Intelligent Robots and Systems, p.2022–2029. https://doi.org/10.1109/IROS51168.2021.9636737
Zheng Z, Liu Y, Zhang XY, 2016. The more obstacle information sharing, the more effective real-time path planning? Knowl-Based Syst, 114:36–46. https://doi.org/10.1016/j.knosys.2016.09.021
Zou LX, Xia L, Du P, et al., 2020. Pseudo Dyna-Q: a reinforcement learning framework for interactive recommendation. Proc 13th Int Conf on Web Search and Data Mining, p.816–824. https://doi.org/10.1145/3336191.3371801
Author information
Authors and Affiliations
Contributions
Yang CHEN designed the research and drafted the paper. Yang CHEN and Dianxi SHI processed the data. Yang CHEN and Huanhuan YANG designed the simulations. Tongyue LI and Zhen WANG helped organize the paper. Yang CHEN and Dianxi SHI revised and finalized the paper.
Corresponding author
Ethics declarations
All the authors declare that they have no conflict of interest.
Additional information
Project supported by the National Natural Science Foundation of China (No. 91948303)
Rights and permissions
About this article
Cite this article
Chen, Y., Shi, D., Yang, H. et al. An anti-collision algorithm for robotic search-and-rescue tasks in unknown dynamic environments. Front Inform Technol Electron Eng 25, 569–584 (2024). https://doi.org/10.1631/FITEE.2300151
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1631/FITEE.2300151