An Optimized Linear Model Predictive Control Solver for Online Walking Motion Generation

Dimitar Dimitrov (1) , Pierre-Brice Wieber (2, 3) , Olivier Stasse (2) , Joachim Ferreau (4) , Holger Diedam (5)

1 Örebro University
2 CNRS-AIST JRL - Joint Robotics Laboratory
3 BIPOP - Modelling, Simulation, Control and Optimization of Non-Smooth Dynamical Systems
4 ESAT - Department of Electrical Engineering [Leuven]
5 IWR - Interdisciplinary Center for Scientific Computing

This article addresses the fast solution of a Quadratic Program underlying a Linear Model Predictive Control scheme that generates walking motions. We introduce an algorithm which is tailored to the particular requirements of this problem, and therefore able to solve it efficiently. Different aspects of the algorithm are examined, its computational complexity is presented, and a numerical comparison with an existing state of the art solver is made. The approach presented here, extends to other general problems in a straightforward way.

Domaines: Robotique [cs.RO]; Optimisation et contrôle [math.OC]

Dimitar Dimitrov, Pierre-Brice Wieber, Olivier Stasse, Joachim Ferreau, Holger Diedam. An Optimized Linear Model Predictive Control Solver for Online Walking Motion Generation. ICRA 2009 - IEEE International Conference on Robotics & Automation, May 2009, Kobe, Japan. pp.1171-1176, ⟨10.1109/ROBOT.2009.5152380⟩. ⟨inria-00390593⟩