Multi-Sensor Resource Management for Layered Sensing

Layered sensing concepts will significantly increase the number and types of sensors available for ATR, but at the same time the ISR problem is becoming become more difficult (e.g., thousands of targets, dismounts, urban terrain, extended operating conditions). To exploit fully these new sensing capabilities, it is necessary to develop multi-sensor resource management algorithms capable of computing near-optimal tasking decisions in real time for thousands of targets and dozens of sensors, and be responsive to changes in sensor performance capabilities versus different target types and operating conditions. To develop such a multi-sensor resource manager Parietal Systems, Inc. (PSI) has developed innovative multi-armed bandit algorithms in Phase I to compute efficient far-sighted solutions of the sensor allocation and scheduling problem for target search and classification. In Phase II PSI will extend these algorithms to integrate tracking with search and classification and to operate in a distributed decision-making environment with multiple users sharing common sensor assets. The effort will evaluate algorithm performance tradeoffs for different operating conditions, and demonstrate the algorithms both in PSIs virtual reality ISR simulation and in an AFRL evaluation system. BENEFIT: The research will produce real-time, distributed multi-sensor resource management algorithms that can be used for the next generation of sensor fusion and ATR systems, which will actively manage sensor resources across multiple platforms and multiple users. The sensor management algorithms will thus benefit the development new multi-sensor systems for military and security applications, but also the development of commercial multi-sensor fusion systems for detection and classification (biosurveillance monitoring, medical diagnosis, cargo inspection, inspection of infrastructure, etc.).