Weighted Multi-tree RRT Algorithm for Efficient Path-planning of Mobile Robots

Path-planning of autonomous mobile robots is very challenging for complex large-scene environments. Existing path-planning strategies often encounter problems such as excessive searching in local areas and producing a large number of useless nodes. To address this problem, this paper proposes a new iterative path planning algorithm for mobile robots via weighted multi-tree rapidly-exploring random trees (WMRRT) algorithm. Aiming at achieving progressive optimization and accelerating convergence, the node list of the first random tree is utilized to heuristically search for subsequent random trees. Combining with artificial potential field, the planning efficiency and collision-free success rate can be strengthened. The related path parameters can be flexibly configurated to meet different requirements. Several experiments are performed to verify the effectiveness under complex scenarios.