Improving Trust in Autonomy Through Modeling Performance

As the Air Force pushes toward reliance on autonomous systems for navigation, situational awareness, threat analysis and target engagement there are several requisite technologies that must be developed.Key among these is the concept of trust in the autonomous system to perform its task.This term, trust has many application specific definitions.In the context of this proposal we shall focus on target engagement by a sensing munition, and development of a tool suite that shall provide target recognition confidence given signature observations and estimates of the sensing environment and target library. Under this proposed effort we will provide a toolkit to foster trust in target declaration for autonomous and semi-autonomous munition systems.This toolkit will construct performance models that enable ATR performance estimation, theoretical classification bounds, and sensor fusion performance estimation as a function of operating conditions.