Multiple UAV Dynamic Task Allocation Using Mixed Integer Linear Programming in a SEAD Mission

*† ‡ This paper extends previous work using Mixed Integer Linear Programming (MILP) to efficiently assign vehicles in a Suppression of Enemy Air Defense (SEAD) mission by adding important new assumptions that increase the complexity of the task allocation problem. The new assumptions include survivable vehicles modeled after unmanned combat aerial vehicles, heterogeneous teams, and dynamic target discovery. With the addition of a new vehicle and mission type, constraints formulated in the previous MILP research are modified or replaced. A dynamic constraint builder is constructed in MATLAB and integrated into the MultiUAV research tool. This allows the new tasking and timing constraints to be formulated based upon the number of mission specified vehicles and the number of targets that have either been detected or partially prosecuted. These constraints can easily be altered based upon rules designated throughout the MATLAB software. Integration into MultiUAV allows the testing of this approach in a dynamic environment.