Distributed Non-Communicating Multi-Robot Collision Avoidance via Map-Based Deep Reinforcement Learning
It is challenging to avoid obstacles safely and efficiently for multiple robots of different shapes in distributed and communication-free scenarios, where robots do not communicate with each other and only sense other robots’ positions and obstacles around them. Most existing multi-robot collision a...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/17/4836 |
| _version_ | 1827707482251722752 |
|---|---|
| author | Guangda Chen Shunyi Yao Jun Ma Lifan Pan Yu’an Chen Pei Xu Jianmin Ji Xiaoping Chen |
| author_facet | Guangda Chen Shunyi Yao Jun Ma Lifan Pan Yu’an Chen Pei Xu Jianmin Ji Xiaoping Chen |
| author_sort | Guangda Chen |
| collection | DOAJ |
| description | It is challenging to avoid obstacles safely and efficiently for multiple robots of different shapes in distributed and communication-free scenarios, where robots do not communicate with each other and only sense other robots’ positions and obstacles around them. Most existing multi-robot collision avoidance systems either require communication between robots or require expensive movement data of other robots, like velocities, accelerations and paths. In this paper, we propose a map-based deep reinforcement learning approach for multi-robot collision avoidance in a distributed and communication-free environment. We use the egocentric local grid map of a robot to represent the environmental information around it including its shape and observable appearances of other robots and obstacles, which can be easily generated by using multiple sensors or sensor fusion. Then we apply the distributed proximal policy optimization (DPPO) algorithm to train a convolutional neural network that directly maps three frames of egocentric local grid maps and the robot’s relative local goal positions into low-level robot control commands. Compared to other methods, the map-based approach is more robust to noisy sensor data, does not require robots’ movement data and considers sizes and shapes of related robots, which make it to be more efficient and easier to be deployed to real robots. We first train the neural network in a specified simulator of multiple mobile robots using DPPO, where a multi-stage curriculum learning strategy for multiple scenarios is used to improve the performance. Then we deploy the trained model to real robots to perform collision avoidance in their navigation without tedious parameter tuning. We evaluate the approach with multiple scenarios both in the simulator and on four differential-drive mobile robots in the real world. Both qualitative and quantitative experiments show that our approach is efficient and outperforms existing DRL-based approaches in many indicators. We also conduct ablation studies showing the positive effects of using egocentric grid maps and multi-stage curriculum learning. |
| first_indexed | 2024-03-10T16:47:12Z |
| format | Article |
| id | doaj.art-caa9d03dab554d70a925f6804eda92c9 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T16:47:12Z |
| publishDate | 2020-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-caa9d03dab554d70a925f6804eda92c92023-11-20T11:31:59ZengMDPI AGSensors1424-82202020-08-012017483610.3390/s20174836Distributed Non-Communicating Multi-Robot Collision Avoidance via Map-Based Deep Reinforcement LearningGuangda Chen0Shunyi Yao1Jun Ma2Lifan Pan3Yu’an Chen4Pei Xu5Jianmin Ji6Xiaoping Chen7School of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, ChinaSchool of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, ChinaSchool of Data Science, University of Science and Technology of China, Hefei 230026, ChinaSchool of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, ChinaSchool of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, ChinaSchool of Data Science, University of Science and Technology of China, Hefei 230026, ChinaSchool of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, ChinaSchool of Computer Science and Technology, University of Science and Technology of China, Hefei 230026, ChinaIt is challenging to avoid obstacles safely and efficiently for multiple robots of different shapes in distributed and communication-free scenarios, where robots do not communicate with each other and only sense other robots’ positions and obstacles around them. Most existing multi-robot collision avoidance systems either require communication between robots or require expensive movement data of other robots, like velocities, accelerations and paths. In this paper, we propose a map-based deep reinforcement learning approach for multi-robot collision avoidance in a distributed and communication-free environment. We use the egocentric local grid map of a robot to represent the environmental information around it including its shape and observable appearances of other robots and obstacles, which can be easily generated by using multiple sensors or sensor fusion. Then we apply the distributed proximal policy optimization (DPPO) algorithm to train a convolutional neural network that directly maps three frames of egocentric local grid maps and the robot’s relative local goal positions into low-level robot control commands. Compared to other methods, the map-based approach is more robust to noisy sensor data, does not require robots’ movement data and considers sizes and shapes of related robots, which make it to be more efficient and easier to be deployed to real robots. We first train the neural network in a specified simulator of multiple mobile robots using DPPO, where a multi-stage curriculum learning strategy for multiple scenarios is used to improve the performance. Then we deploy the trained model to real robots to perform collision avoidance in their navigation without tedious parameter tuning. We evaluate the approach with multiple scenarios both in the simulator and on four differential-drive mobile robots in the real world. Both qualitative and quantitative experiments show that our approach is efficient and outperforms existing DRL-based approaches in many indicators. We also conduct ablation studies showing the positive effects of using egocentric grid maps and multi-stage curriculum learning.https://www.mdpi.com/1424-8220/20/17/4836multi-robot navigationdistributed collision avoidancedeep reinforcement learning |
| spellingShingle | Guangda Chen Shunyi Yao Jun Ma Lifan Pan Yu’an Chen Pei Xu Jianmin Ji Xiaoping Chen Distributed Non-Communicating Multi-Robot Collision Avoidance via Map-Based Deep Reinforcement Learning Sensors multi-robot navigation distributed collision avoidance deep reinforcement learning |
| title | Distributed Non-Communicating Multi-Robot Collision Avoidance via Map-Based Deep Reinforcement Learning |
| title_full | Distributed Non-Communicating Multi-Robot Collision Avoidance via Map-Based Deep Reinforcement Learning |
| title_fullStr | Distributed Non-Communicating Multi-Robot Collision Avoidance via Map-Based Deep Reinforcement Learning |
| title_full_unstemmed | Distributed Non-Communicating Multi-Robot Collision Avoidance via Map-Based Deep Reinforcement Learning |
| title_short | Distributed Non-Communicating Multi-Robot Collision Avoidance via Map-Based Deep Reinforcement Learning |
| title_sort | distributed non communicating multi robot collision avoidance via map based deep reinforcement learning |
| topic | multi-robot navigation distributed collision avoidance deep reinforcement learning |
| url | https://www.mdpi.com/1424-8220/20/17/4836 |
| work_keys_str_mv | AT guangdachen distributednoncommunicatingmultirobotcollisionavoidanceviamapbaseddeepreinforcementlearning AT shunyiyao distributednoncommunicatingmultirobotcollisionavoidanceviamapbaseddeepreinforcementlearning AT junma distributednoncommunicatingmultirobotcollisionavoidanceviamapbaseddeepreinforcementlearning AT lifanpan distributednoncommunicatingmultirobotcollisionavoidanceviamapbaseddeepreinforcementlearning AT yuanchen distributednoncommunicatingmultirobotcollisionavoidanceviamapbaseddeepreinforcementlearning AT peixu distributednoncommunicatingmultirobotcollisionavoidanceviamapbaseddeepreinforcementlearning AT jianminji distributednoncommunicatingmultirobotcollisionavoidanceviamapbaseddeepreinforcementlearning AT xiaopingchen distributednoncommunicatingmultirobotcollisionavoidanceviamapbaseddeepreinforcementlearning |