A multi-agent path planning algorithm based on hierarchical reinforcement learning and artificial potential field | IEEE Conference Publication | IEEE Xplore