Quasi-Decentralized Task Assignment for Multiple UAV Coordination
Jarurat Ousingsawat · AIAA Guidance, Navigation, and Control Conference and Exhibit · 2006
Task Assignment is one of fundamental problems in multiple UAV systems. Decision of one UAV has effect on the others in the same team. Because the performance no longer measures from an individual vehicle, the tasks have to be distributed to maximize the team performance. An underlying problem is computational burden that grows considerably as the system gets larger. Many approaches cannot extend to larger systems because of this problem. The objective of this paper is to develop a quasi-decentralized methodology to solve the task assignment problem using less computational time. The assignment is decomposed into sub-problems that can be solved quickly. Then the results are combined in order to determine the best solution for the team performance. This approach is also more adaptable for dynamic environment. The coordination of multiple UAVs is utilized as a platform to demonstrate the methodology. The performance is compared to a centralized approach. For a system with five UAVs and ten goals, the cost is 5% higher than the optimal cost, while the run time is cut down by 55% on average.