Towards More Possibilities: Motion Planning and Control for Hybrid Locomotion of Wheeled-Legged Robots

Abstract

This letter proposed a control framework to tackle the hybrid locomotion problem of wheeled-legged robots. It comes as a hierarchical structure with three layers: hybrid foot placement planning, Centre of Mass (CoM) trajectory optimization and whole-body control. General mathematical representation of foot movement is developed to analyze different motion modes and decide hybrid foot placements. Gait graph widely used in legged locomotion is extended to better describe the hybrid movement by adding extra foot velocity information. Thereafter, model predictive control is introduced to optimize the CoM trajectory based on the planned foot placements considering terrain height changing. The desired trajectories together with other kinematic and dynamic constraints are fed into a whole-body controller to produce joint commands. In the end, the feasibility of the proposed approach is demonstrated by the simulation and experiments of hybrid locomotion running on our wheeled quadrupedal robot Pholus.