Application of a probabilistic market-based approach in UAV sensor & perception management

The integration of unmanned airborne sensor platforms into distributed sensor networks is considered beneficial in terms of extending the surveillance area and enhancing multi-perspective sensor data fusion. In this context, exploiting the respective platform capabilities, but hiding the complex airborne system internals e.g. platform guidance and sensor control is one of the challenges. We propose a task based deployment of such UAVs (Unmanned Aerial Vehicles) combined with an onboard Sensor- and Perception Management System (SPMS) capable of coping with dynamic environments and resource limitation. Herein environmental dependencies and sensor constraints are modeled with expert knowledge using Bayesian networks and evaluated within a market based coordination approach during flight.