Test and Certification Techniques for Autonomous Guidance and Navigation Algorithms for Navy Air Vehicle Missions

The robustness and flexibility of UAS will be enhanced through the integration of advanced outer-loop algorithms that modify system behavior based on mission requirements and environmental factors. The greatest obstacle to the operational use of such algorithms is test methodologies to assess algorithm performance and safety during the airworthiness certification process. Under the proposed Phase II program, the research team will develop the INSPECT tool to enable certification of outer-loop algorithms through the use of intelligent simulation-based testing. The INSPECT tool will: (1) efficiently determine both worst-case and expected performance over a high-dimensional test space, (2) provide effective methods for modeling performance in the challenging “black box” case, and (3) implement theoretically sound approaches for assessing performance in outer-loop algorithms that exhibit nondeterministic and/or discontinuous behavior. INSPECT will provide a high level of confidence in next-generation outer-loop algorithms with a substantially reduced computational burden relative to brute-force methods. During the Phase II base effort, an airspace integration scenario focusing on autonomous collision avoidance will be used to explore the capabilities and address any limitations of the approach. During the Phase II option effort, multiple classes of algorithms will be evaluated to demonstrate the broad applicability of the approach.