Collaborative UAV-UGV Tracking of Elusive Dismounts

Unmanned Systems (UMS) provide increased battlespace awareness and act as a force multiplier in operations today and in the future. However, there are several enabling technologies that need to be further developed in order to provide the forces with effective UAV – UGV collaborative operation to get and keep “cross hairs on target.” Currently, UAVs are performing autonomous tracking and surveillance of ground threats with some success in relatively open terrain where the onboard sensors can easily and continually view ground objects of interest. However, this tracking capability is critically needed in urban environments, where many of today’s conflicts take place. Automated trackers applied in urban environments have shown poor performance due to a number of factors: Occlusions, Confusors, and Tracking reliability. In order for UMSs to effectively track elusive dismounts operating in an urban environment, fundamental changes have to be made to the way that threats are tracked autonomously. We are proposing an active vision approach that provides Geometric Modeling, Geometric reasoning, Collaborative Threat Modeling, Track fusion, Vision Based Control to improve tracking performance in complex environments