Intelligent guidance of autonomous mobile robots based on adaptive dynamic programming
Xuejing Lan, Chengxuan Qin, Yiwen Liu, Haibin Ouyang, Guiyun Liu · 2019
This paper discussed the guidance problem of autonomous mobile robots with multiple constraints: target tracking, synchronization, and obstacle avoidance. The guidance strategy is proposed according to the structure of adaptive dynamic programming to enable the online learning and optimization. An action neural network and a critic neural network are designed to estimate the guidance strategy and cost function, respectively. Then, an optimal intelligent guidance law is obtained according to the designed weight updating rules of neural networks. Finally, the validity of the intelligent guidance scheme is demonstrated with a simulation of five autonomous mobile robots tracking a dynamic target in the obstacle environment.