Cooperative Sensing and Path Planning in a Multi-vehicle Environment∗∗This work was in part (Melin, J., Ritala, R.) funded by the Academy of Finland, project “Optimization of observation subsystems in autonomous mobile machines”, O3-SAM
J. Melin, Mikko Lauri, Antti Kolu, J. Koljonen, Risto K. Ritala · IFAC-PapersOnLine · 2015
We study the cooperation of an unmanned ground vehicle (UGV) and micro air vehicle (MAV) in a path planning task. The UGV requests the MAV to execute observation missions which provide stereo image data on areas that the UGVs sensors cannot observe. The problem is formulated as a partially observable decision making problem. The solution is a decision policy that determines conditional on current information whether the UGV should move along a certain route or request an observation mission. A case example for a real-world demonstration of the cooperative path planning task is described. We discuss the features of single-vehicle control systems and propose a multi-vehicle communication framework that is tolerant against communication breaks.