Robust obstacle crossing of a wheel-legged mobile robot using minimax force distribution and self-reconfiguration

This paper focuses on the problem of robust obstacles crossing for a high mobility wheel-legged robot. To improve the obstacle clearance capability, a method dealing with the contact stability optimization is developed. A specific stability criterion taking the friction into account is proposed. The optimization algorithm uses both the kinematic redundancy in order to modify the position of the Center of Mass (CoM), modifying the resulting distribution of contact forces, and the actuation redundancy to improve the stability of frictional contacts by adapting the internal forces. We show that the choice of this particular criterion allows us to maximize the robustness of contacts stability relatively to the modeling errors affecting force control (friction in mechanical transmission). Performances of this algorithm are evaluated in simulation and the necessity for a CoM trajectory planning is highlighted by an analysis of obstacle crossing using this criterion.