Safe-radius based motion planning of hexapod using RRT-connect

Abstract

Rapidly exploring Random Trees (RRT), a sampling based algorithm, efficiently computes a path between a start and a goal configuration. RRT-Connect, is a variant of RRT that works by incrementally building two RRTs rooted at the start and the goal configurations. Significant amount of research has been done on the motion planning of six-legged robots. We improve upon a certain technique which employs a human-like approach. Initially, a rough path is calculated using a motion planning algorithm. This path is then further divided into start and goal configurations and another motion planning algorithm is used to compute these paths which may force the robot to take detours. Our proposed technique brings an improvement to this approach by making use of another human behaviour in the algorithm.