A Cognitive Architecture for Naval Mine Countermeasures (MCM)

Current Mine Countermeasure (MCM) operations attempt to balance risk and time via processes that are staff and time intensive. The range of experience in the MCM staff further complicates these processes. Existing tools such as MEDAL provide some support to address these issues, but those tools lack a comprehensive coverage of the process; in particular, there is no direct replanning support. Additionally, the tools do not help the inexperienced staff understand and deal with various situations. Existing tools are mostly passive, representing data, but not interpreting it, leaving issues of issue detection, option generation, and option evaluation entirely up to the staff. The Navy is interested in exploring whether cognitive architectures can help address these issues. Recent years have seen increased development of cognitive architectures, which are ideal platforms for encoding expert knowledge and interacting with users. TRUFAST (Transparent Reasoning under Uncertainty for Antimine Systems in real-Time) leverages mixed-initiative interaction, human-behavior models, explanation, and adaptation using episodic memory in order to perform its issue detection, option generation, and option evaluation functions. We call this combination of features Cognitive Episodic replan Assessment (CErA). CErA is a general approach that could be applied to many domains; TRUFAST performs CErA for MCM.