Multirobot coordination with deep reinforcement learning in complex environments

作者：

Highlights：

• In dynamic environment, starting tabula rasa, only image data are used as input.

• Our approach can solve multirobot task assignment and path planning problems.

• Results indicate that our method effectively applied to multi-robot coordination.

摘要

•In dynamic environment, starting tabula rasa, only image data are used as input.•Our approach can solve multirobot task assignment and path planning problems.•Results indicate that our method effectively applied to multi-robot coordination.

论文关键词：Multirobot coordination,Reinforcement learning,Deep learning,Visual perception

论文评审过程：Received 26 February 2020, Revised 29 January 2021, Accepted 24 April 2021, Available online 29 April 2021, Version of Record 11 May 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115128