Hierarchical Control of the Over-Actuated ROSPO Platform via Static Input Allocation

C Nainer (1) , Michele Furci (2) , Alexandre Seuret (3) , Luca Zaccarian (3) , Antonio Franchi (2)

1 UNITN - Università degli Studi di Trento = University of Trento
2 LAAS-RIS - Équipe Robotique et InteractionS
3 LAAS-MAC - Équipe Méthodes et Algorithmes en Commande

This paper addresses the problem of control allocation applied to an over-actuated hovercraft-type vehicle. A hierarchical control architecture, consisting of a high level controller for trajectory tracking, and a control allocation algorithm, is developed and proved to be effective in tracking a desired trajectory while optimizing some cost related to actuator constraints. The control allocation algorithm exploits the redundancy of the system in order to keep the actuator states inside their saturation limits and tries to minimize the power consumption of the propellers. Unlike other papers on control allocation, actuator dynamics is taken into account. The control architecture is tested through simulations that well illustrate the capabilities of the proposed control design.

Mots clés: Robot Control; Mobile Robots; Over-actuated Robots; Allocation; Hierarchical Control

Domaines: Automatique / Robotique

C Nainer, Michele Furci, Alexandre Seuret, Luca Zaccarian, Antonio Franchi. Hierarchical Control of the Over-Actuated ROSPO Platform via Static Input Allocation. IFAC World Congress 2017, Jul 2017, Toulouse, France. 6p., ⟨10.1016/j.ifacol.2017.08.2260⟩. ⟨hal-01496283v2⟩