Predator-Prey Reward Based Q-Learning Coverage Path Planning for Mobile Robot
Coverage Path Planning (CPP in short) is a basic problem for mobile robot when facing a variety of applications. <inline-formula> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula>-Learning based coverage path planning algorithms are beginning to be explore...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10064303/ |
| _version_ | 1797855316160806912 |
|---|---|
| author | Meiyan Zhang Wenyu Cai Lingfeng Pang |
| author_facet | Meiyan Zhang Wenyu Cai Lingfeng Pang |
| author_sort | Meiyan Zhang |
| collection | DOAJ |
| description | Coverage Path Planning (CPP in short) is a basic problem for mobile robot when facing a variety of applications. <inline-formula> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula>-Learning based coverage path planning algorithms are beginning to be explored recently. To overcome the problem of traditional <inline-formula> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula>-Learning of easily falling into local optimum, in this paper, the new-type reward functions originating from Predator-Prey model are introduced into traditional <inline-formula> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula>-Learning based CPP solution, which introduces a comprehensive reward function that incorporates three rewards including Predation Avoidance Reward Function, Smoothness Reward Function and Boundary Reward Function. In addition, the influence of weighting parameters on the total reward function is discussed. Extensive simulation results and practical experiments verify that the proposed Predator-Prey reward based <inline-formula> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula>-Learning Coverage Path Planning (PP-<inline-formula> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula>-Learning based CPP in short) has better performance than traditional BCD and <inline-formula> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula>-Learning based CPP in terms of repetition ratio and turns number. |
| first_indexed | 2024-04-09T20:21:43Z |
| format | Article |
| id | doaj.art-26a76fd3eafb4d17985684a977c1aa76 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-09T20:21:43Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-26a76fd3eafb4d17985684a977c1aa762023-03-30T23:01:21ZengIEEEIEEE Access2169-35362023-01-0111296732968310.1109/ACCESS.2023.325500710064303Predator-Prey Reward Based Q-Learning Coverage Path Planning for Mobile RobotMeiyan Zhang0https://orcid.org/0000-0002-0396-5786Wenyu Cai1https://orcid.org/0000-0002-8858-9221Lingfeng Pang2College of Electrical Engineering, Zhejiang University of Water Resources and Electric Power, Hangzhou, ChinaCollege of Electronics and Information, Hangzhou Dianzi University, Hangzhou, ChinaCollege of Electronics and Information, Hangzhou Dianzi University, Hangzhou, ChinaCoverage Path Planning (CPP in short) is a basic problem for mobile robot when facing a variety of applications. <inline-formula> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula>-Learning based coverage path planning algorithms are beginning to be explored recently. To overcome the problem of traditional <inline-formula> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula>-Learning of easily falling into local optimum, in this paper, the new-type reward functions originating from Predator-Prey model are introduced into traditional <inline-formula> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula>-Learning based CPP solution, which introduces a comprehensive reward function that incorporates three rewards including Predation Avoidance Reward Function, Smoothness Reward Function and Boundary Reward Function. In addition, the influence of weighting parameters on the total reward function is discussed. Extensive simulation results and practical experiments verify that the proposed Predator-Prey reward based <inline-formula> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula>-Learning Coverage Path Planning (PP-<inline-formula> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula>-Learning based CPP in short) has better performance than traditional BCD and <inline-formula> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula>-Learning based CPP in terms of repetition ratio and turns number.https://ieeexplore.ieee.org/document/10064303/Coverage path planningpredator-prey modelreinforcement learningQ-learning algorithmmobile robot |
| spellingShingle | Meiyan Zhang Wenyu Cai Lingfeng Pang Predator-Prey Reward Based Q-Learning Coverage Path Planning for Mobile Robot IEEE Access Coverage path planning predator-prey model reinforcement learning Q-learning algorithm mobile robot |
| title | Predator-Prey Reward Based Q-Learning Coverage Path Planning for Mobile Robot |
| title_full | Predator-Prey Reward Based Q-Learning Coverage Path Planning for Mobile Robot |
| title_fullStr | Predator-Prey Reward Based Q-Learning Coverage Path Planning for Mobile Robot |
| title_full_unstemmed | Predator-Prey Reward Based Q-Learning Coverage Path Planning for Mobile Robot |
| title_short | Predator-Prey Reward Based Q-Learning Coverage Path Planning for Mobile Robot |
| title_sort | predator prey reward based q learning coverage path planning for mobile robot |
| topic | Coverage path planning predator-prey model reinforcement learning Q-learning algorithm mobile robot |
| url | https://ieeexplore.ieee.org/document/10064303/ |
| work_keys_str_mv | AT meiyanzhang predatorpreyrewardbasedqlearningcoveragepathplanningformobilerobot AT wenyucai predatorpreyrewardbasedqlearningcoveragepathplanningformobilerobot AT lingfengpang predatorpreyrewardbasedqlearningcoveragepathplanningformobilerobot |