查看文章

Pursuit-evasion is the problem of capturing mobile targets with one or more pursuers. We use deep reinforcement learning for pursuing an omnidirectional target with multiple, homogeneous agents that are subject to unicycle kinematic constraints. We use shared experience to train a policy for a given number of pursuers, executed independently by each agent at run-time. The training uses curriculum learning, a sweeping-angle ordering to locally represent neighboring agents, and a reward structure that encourages a good formation and combines individual and group rewards. Simulated experiments with a reactive evader and up to eight pursuers show that our learning-based approach outperforms recent reinforcement learning techniques as well as non-holonomic adaptations of classical algorithms. The learned policy is successfully transferred to the real-world in a proof-of-concept demonstration with three …