Description:
OBJECTIVE: Develop innovative frameworks, tools, and human-machine interfaces that provide improved trust, transparency in the autonomous system or provide more flexible, cognitively matched human-machine interaction and cooperation. DESCRIPTION: Human-autonomous system interaction is frequently limited by lack of confidence and trust among the (combined) team. In order to have humans collaborate effectively with autonomous systems, improved interfaces and interaction techniques, and frameworks of interaction must be developed that allow for a common perception of the goals, constraints, resources, and other variables relevant to the team"s overarching objective. Tools that allow for improved transparency into the machine"s reasoning, human-machine interfaces that allow for more natural and flexible interaction and shared decision making, and information/decision frameworks that provide cognitively matched human-machine situational awareness would all support more effective and trusted human-machine teaming. Development of better measures of human trust or measures of the accuracy of shared human-machine perception would also facilitate improved teaming. PHASE I: The first phase consists of investigating and developing novel interaction techniques and frameworks, tools, or human-autonomous system interfaces that would provide one or more of the following advantages: effective teaming through improved transparency, a natural shared human-machine perception of the problem space or operational environment, flexible and trusted shared decision making (or task allocation), or improved measures of human intent/trust/vigilance. The Phase I deliverables should include a final Phase I report that will include the algorithms and hardware needed to implement the framework, tool, or human-machine interface. Feasibility of the proposed approach should be demonstrated through simulation or implementation. PHASE II: Phase II shall produce and deliver fully functional, prototype software that demonstrates in a more refined and robust manner the functionality and capabilities developed in Phase I. Analysis of the performance of the combined team must be included, for example: A comparison of the transparency of the automated system to the human operators relative to current (i.e. baseline) systems, or a comparison of the level of human trust in the automation. Some reasonable measure of the improved performance afforded by the framework, tool, or interface relative to a current (baseline) system. Discussion of the trade-offs inherent in the novel software interface or tool with regards to human trust, efficiency of the automation, and total system performance. The Phase II prototype software should show proof-of-concept by applying the framework, tool, or interface to a particular, relevant DoD use case.. Of particular interest would be use cases that can be generalized across domains (e.g., Underwater vehicles) or include heterogenous autonomous systems (e.g., air/ground/sea coordination). PHASE III: DUAL USE COMMERCIALIZATION: Transition the work of phase II to a DoD development effort. Improved human-machine teaming in autonomous systems should reduce training time, increase ease of use, and improve total system performance. Autonomous systems are also in limited use in manufacturing fabrication, logistics scheduling, and remote vehicle management/operation. Potential commercial applications of this technology include designing human-autonomous systems interfaces and interaction tools for management of complex manufacturing systems, scheduling of complex or rapid logistics processes, or trusted multi-aircraft or multi-truck vehicle system control.
