Real-Time Resource Optimization of a UAV Network for Continuous Video Tracking

Networks of UAV’s equipped with video sensors promise an affordable capability to keep track of moving ground targets of military interest in an urban environment in the presence of target occlusion, clutter, and other background traffic. But maintaining continuous track on multiple targets that move in and out of the field of view of any single UAV’s sensor poses a significant problem of control, coordination, and fusion of the multiple sensor resources. In addition, limited processing onboard for video tracking and limited communication bandwidth available to transmit video data to ground stations or other UAV’s, make it crucial to allocate these resources in concert to achieve the maximum system benefit. Parietal Systems, Inc. (PSI) and Boston University (BU) propose to develop a unified real-time resource optimization algorithm to control sensors, processing, and communications in a distributed tracking system that fuses data from a network of UAV’s in order to continuously track multiple targets in an urban environment. In Phase I we will develop and demonstrate the algorithm using analytic and simulation models of sensor, processing, and communication subsystems. We will also validate key components of the video processing and tracking model with controlled video experiments.