Adaptive Environment Modeling Based Reinforcement Learning for Collision Avoidance in Complex Scenes

Shuaijun Wang1,2; Rui Gao2; Ruihua Han2,3; Shengduo Chen2; Chengyang Li2; Qi Hao2,4

10.1109/IROS47612.2022.9982107

2022

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

2153-0858

978-1-6654-7928-8

2022 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS)

9011-9018

OCT 23-27, 2022

null,Kyoto,JAPAN

345 E 47TH ST, NEW YORK, NY 10017 USA

IEEE

The major challenges of collision avoidance for robot navigation in crowded scenes lie in accurate environment modeling, fast perceptions, and trustworthy motion planning policies. This paper presents a novel adaptive environment model based collision avoidance reinforcement learning (i.e., AEMCARL) framework for an unmanned robot to achieve collision-free motions in challenging navigation scenarios. The novelty of this work is threefold: (1) developing a hierarchical network of gated-recurrent-unit (GRU) for environment modeling; (2) developing an adaptive perception mechanism with an attention module; (3) developing an adaptive reward function for the reinforcement learning (RL) framework to jointly train the environment model, perception function and motion planning policy. The proposed method is tested with the Gym-Gazebo simulator and a group of robots (Husky and Turtlebot) under various crowded scenes. Both simulation and experimental results have demonstrated the superior performance of the proposed method over baseline methods.

Training Adaptation models Adaptive systems Navigation Computational modeling Dynamics Reinforcement learning

通讯

英语

CPCI-S

Science and Technology Innovation Committee of Shenzhen City[JCYJ20200109141622964]

Automation & Control Systems ; Computer Science ; Engineering ; Robotics

Automation & Control Systems ; Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic ; Robotics

WOS:000909405301088

Web of Science

1.Harbin Inst Technol, Harbin, Heilongjiang, Peoples R China
2.Southern Univ Sci & Technol, Dept Comp Sci & Engn, Shenzhen 518055, Guangdong, Peoples R China
3.Univ Hongkong, Dept Comp Sci, Hong Kong 999077, Peoples R China
4.Southern Univ Sci & Technol, Rsearch Inst Trustworthy Autonomous Syst, Shenzhen, Guangdong, Peoples R China

Shuaijun Wang,Rui Gao,Ruihua Han,et al. Adaptive Environment Modeling Based Reinforcement Learning for Collision Avoidance in Complex Scenes[C]. 345 E 47TH ST, NEW YORK, NY 10017 USA:IEEE,2022:9011-9018.