Graph-Based Path/Foothold Planning and Quantitative Map Evaluation for Multi-Limbed Climbing Robots

This paper presents a graph-based algorithm for path and foothold planning for multi-limbed climbing robots. For multi-limbed climbing robots locomoting in unstructured environments, path planning and foothold planning are essential because the robots need to move while grasping a limited number of discrete graspable points. In this paper, we present a new graph-based method for obtaining the robot pose path that connects the start and destination points in a minimum number of steps and footholds while following the path when the graspable points distribution and robot configuration are given. The proposed method quantitatively evaluates a given map and planned path in terms of importance level for graspable points as well as path robustness against the loss of graspable points. The proposed algorithm is implemented in a multi-legged robot simulation.