Simulation-Based Climbing Capability Analysis for Quadrupedal Robots

Abstract

The design of legged robotic systems targeted for climbing on steep terrain is critical for expanding robotic exploration in challenging terrain. While the most advanced quadruped robots show encouraging performance in level locomotion, there is little knowledge on the optimal design for climbing legged robots. In this paper, we investigate the climbing performance of quadrupedal systems with different joint topologies. To this end, we present a quantitative comparison performed in simulations of two robots in different configurations concerning locomotion stability, energy efficiency, and control versatility. Based on the results, optimized nominal stances are selected for each robot, and their climbing locomotion is demonstrated and compared in a virtual deployment.