MI-RRT-Connect Algorithm for Quadruped Robotics Navigation with Efficiently Path Planning

Abstract

Autonomous navigation is playing an increasingly important role in quadruped robotic systems. However, providing safe and reliable path planning for robots is still an open problem. In this paper, we propose a sampling-based path planning algorithm fused with a dual-tree structure, here called Multiple Informed RRT-Connect (MI-RRT-Connect). The proposed MI-RRT-Connect can overcome the disadvantage that the Informed RRT* algorithm takes a long time in the initial search path, by using the RRT-Connect algorithm and the target deviation method, and the initial search path can be obtained quickly. Then, for the optimization of the initial search path, the multi-level parent node selection strategy and the method of calculating the path cost for the parent node are used. The simulation and quadruped robot experimental results show that our method can find an optimal path in a shorter time, and it has been well applied in a real quadruped robot.