Sensor Tipping and Cueing using Artificial Intelligence Techniques

ABSTRACT:The ultimate goal of this proposed effort is to improve OPIR follow-on ISR tipping and cueing; dramatically shorten the associated response time to new OPIR events; and improve ISR collection quality, resulting in: improved analyst performance; better-quality schedules; and higher, and better utilization of sensors. The goals of the Phase I research are to understand the OPIR tipping and cueing domain; elicit required planning knowledge related to follow-on sensor tasking from expert satellite image analysts; understand the resources, tasks, and constraints relating to the ISR scheduling problem; investigate integration requirements; define the appropriate performance metrics; elaborate the heuristics, algorithms, and techniques for improved OPIR tipping/cueing as well as to analyze them as to their feasibility in several dimensions and against the metrics; develop the Phase II system design; and further prove the feasibility of the techniques through prototype development and demonstrations in realistic scenarios. Stottler Henke has developed significant technologies and techniques for both the planning and scheduling parts of the problem AND we have teamed with actual analysts from BOTH the defense and intelligence communities who have extensive experience with the tipping and cueing process.BENEFIT:The most direct target for the results of this effort is, of course, this countrys ISR satellite system itself. The work is also applicable to other ISR assets such as UAVs. By both demonstrating significantly improved performance during Phase II and ensuring that the ultimate results can be integrated into the operational systems, we will provide the DoD and intelligence communities with a strong incentive to operationalize the Phase II development of the proposed PROACT. Beyond these there are several commercialization possibilities available in the Government and private sector. The algorithms developed here could be marketed to our existing Air Force, Navy, and MDA sensor customers interested in sensor management. As mentioned in this proposal, commercial companies now offer satellite sensor services and these are potential targets as well. There are other commercial applications for the technology developed for the Scheduler. We are already marketing Aurora, our general intelligent planning and scheduling framework, and customizing it for a variety of domains. The additional capabilities developed for OPIR tipping and cueing can be synergistically marketed to the same clients. For example, the bottleneck avoidance algorithm proposed as fully developed here would be applicable to many Aurora customers. We anticipate that this effort will result in additional scheduling algorithms that we will be able to incorporate into our existing scheduling products, thus increasing the benefits they provide and their value.