Simulation of a Soar-Based Autonomous Mission Management System for Unmanned Aircraft.

State-of-the-art unmanned aerial vehicles are typically able to autonomously execute a preplanned mission. However, unmanned aerial vehicles usually fly in a very dynamic environment that requires dynamic changes to the flight plan; this mission management activity is usually tasked to human supervision. Through the use of a set of theoretical concepts that allow the description of a flight plan, a software system that autonomously accomplishes the mission management task for an unmanned aerial vehicle was developed. The system was implemented using a combination of Soar intelligent agents and traditional control techniques, and it is thoroughly described in the first part of the paper. An extensive testing campaign, based on the use of a realistic simulation environment, was performed on the system; the second part of this paper presents results obtained during this campaign. The system was demonstrated to be capable of automatically generating and executing an entire flight plan after being assigned a set of objectives. In conclusion, possible future developments are discussed, focusing in particular on prospective hardware implementation for the system.