An interactive decision support tool for real-time in-flight replanning of autonomous vehicles
There is significant interest in the Department of Defense to design and build networks of unmanned vehicles that will have the ability of operating autonomously, which will in effect, take the human out of the loop at certain levels of tasking. While unmanned systems do not necessarily have a human in direct, manual control, humans will be necessarily involved in planning, higher-level operation, and contingency interventions (including system failures). Thus a system of command and control unmanned vehicles is still a manned system. The implementation of networks of autonomous airborne vehicles means that not only will battlefield commanders have more flexibility and options; it also means that a layer of human command and control will be needed where none previously existed. In this futuristic command and control network, the nature of human interaction will change from that of manual control to supervisory control, with different layers of automation, which will require more abstract knowledge-based reasoning. Requiring an operator to manage high value assets in a time-pressure environment through constant replanning requires substantial cognitive effort, and affects both individuals and teams in the command and control environment. This paper will discuss a prototype created to address this dynamic replanning problem through a U.S. Navy future naval capabilities research effort.