AUTHOR=Zhou Yuxiang , Shu Jiansheng , Zheng Xiaolong , Hao Hui , Song Huan 

TITLE=Real-time route planning of unmanned aerial vehicles based on improved soft actor-critic algorithm

JOURNAL=Frontiers in Neurorobotics

VOLUME=Volume 16 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/neurorobotics/articles/10.3389/fnbot.2022.1025817

DOI=10.3389/fnbot.2022.1025817

ISSN=1662-5218

ABSTRACT=With the application and development of UAV technology and navigation and positioning technology, higher requirements are put forward for UAV maneuvering obstacle avoidance ability and real-time trajectory planning. In this paper, aiming at the UAV online path planning problem in unknown environment, the superior generalization of deep reinforcement learning is used to improve the state observation and reward function combined with the idea of artificial potential field algorithm. The problem of sparse reward in reinforcement learning algorithm is solved, and the convergence speed of the algorithm is improved. Based on the idea of curriculum learning and migration learning, the algorithm is trained step by step according to the difficulty of the task, and the generalization of the algorithm is improved. The simulation results show that the improved SAC algorithm has fast convergence speed, good timeliness and strong generalization, and can better complete the UAV path planning task.