A mixed integer linear programming approach for strategic issues in RoboFlag
Authors: | Foirien Nicolas, Ecole Nationale Supérieure des Mines de Paris, France Murray Richard, California Institute of Technology, United States |
---|
Topic: | 4.3 Robotics |
---|
Session: | Unmanned Aerial, Ground and Marine Vehicles |
---|
Keywords: | co-operative control, autonomous vehicles, decision making, linear programming, dicrete event systems, real-time |
---|
Abstract
Many robotics applications involve teams composed of agents that have to act in a cooperative manner, often corresponding to several tasks to accomplish. For dynamic vehicles, with an adverse and thus strategic context, the nature of the optimization problem is hybrid and gathers allocation and optimal trajectories problems. In this paper a mixed-integer linear programming approach is developed to bring a solution to these types strategic issues, facing the challenge of finding a real-time solution. The problem is presented with the context of RoboFlag, motivated by the hardware implementation at Caltech.