Multi-agent Real-time Route Planning Based on Fuzzy Adaptive DWA

Aiming at the real-time route planning problem of intelligent agents, an improved dynamic window algorithm (DWA) with fuzzy adaptive parameter adjustment is proposed to complete the real-time route planning of multi-agent. The motion model of agent and the discretized motion representation are established, and a fuzzy controller is introduced to give the fuzzy adaptive adjustment of the parameters. The weight parameters of each part in the real-time evaluation function of the waypoints are evaluated according to the distances between the agent and the global target point, the distance of agent to nearest obstacle, and the distance between the neighboring agent, then the best the optimal control instructions is selected, which solves the real-time route planning of the cooperative movement of multi-agent. Simulation experiments verify that the algorithm proposed in this paper called Fuzzy Adaptive DWA is an effective method to get a reasonable route for the multi-agents.