Variable-polygon-based planning method for reconfigurable topology truss robot

Abstract

This paper proposes a variable-polygon-based random tree search algorithm with a variable polygon shape (VPRT) for Variable Topology Truss (VTT). This algorithm is a modified version of the previously proposed polygon-based random tree search expanded to the use of variable polygons other than triangles. The algorithm defines a new foot placement for polygons and subsequently determines the lowest cost required to complete the locomotion path of a VTT with non-triangular faces. Using this algorithm, the path planning of a VTT with any type of polygon can be performed. The performance of the algorithm was validated using simulated results with reference polyhedrons.