Path following with collision avoidance and velocity constraints for multi-agent group formations

Ionela Prodan (1) , Sorin Olaru (1) , Cristina Stoica (1) , Silviu-Iulian Niculescu (2)

1 E3S - Supélec Sciences des Systèmes
2 L2S - Laboratoire des signaux et systèmes

This paper deals with avoidance constraints while following an optimal trajectory for a group of agents operating in open space. The basic idea is to use the Model Predictive Control (MPC) technique to solve a real time optimization problem over a finite time horizon. Following a specified trajectory, the agents move in the same direction and eventually they end up in a particular formation. Combining the optimization-based control study with the ability of MPC to handle convex and non-convex constraints allows a thorough analysis of the motion control of a group of agents with linear dynamics subject to state constraints. Avoidance constraints are also added to the optimization problem when the agents are operating in an environment with obstacles. A flat trajectory is planned in the physical open space allowing the agents to maneuver successfully in such a dynamic environment and to reach a common objective.

Mots clés: Multi-agent systems; cooperative control; Model Predictive Control; non-convex constrains

Domaines: Automatique / Robotique

Ionela Prodan, Sorin Olaru, Cristina Stoica, Silviu-Iulian Niculescu. Path following with collision avoidance and velocity constraints for multi-agent group formations. Annals of the University of Craiova Physics AUC , 2010, 7(34) (2), pp.33-38. ⟨hal-00593091⟩