Distributed safe reinforcement learning for multi-robot motion planning
Distributed safe reinforcement learning for multi-robot motion planning
复制标题
DOI：
10.1109/med51440.2021.9480176
复制
发表时间：
2021-06
期刊：
影响因子：
--
通讯作者：
Yang Lu;Yaohua Guo;Guoxiang Zhao;Minghui Zhu
中科院分区：
文献类型：
--
作者：
Yang Lu;Yaohua Guo;Guoxiang Zhao;Minghui Zhu
本文研究了具有避碰功能的多个移动机器人的最优运动规划。我们开发了一种分布式强化学习算法，可确保同时实现次优目标和随时避免碰撞。建立了神经网络权值收敛性、闭环系统系统状态一致和最终有界性、任意碰撞避免等方面的理论结果。单个积分器和独轮机器人的数值模拟说明了我们理论结果的有效性。
This paper studies optimal motion planning of multiple mobile robots with collision avoidance. We develop a distributed reinforcement learning algorithm which ensures suboptimal goal reaching and anytime collision avoidance simultaneously. Theoretical results on the convergence of neural network weights, the uniform and ultimate boundedness of system states of the closed-loop system, and anytime collision avoidance are established. Numerical simulations for single integrator and unicycle robots illustrate the effectiveness of our theoretical results.