Finite Stage MDP Method and Algorithm for Task Allocation in UCAVS

Abstract

Multi-UCAV (Unmanned Combat Air Vehicle) UAV operations task allocation is an important part of decision-making. It is also for solving a class of difficult combinatorial optimization problem for which there is no mature and high effective solving schemes. The problem of task allocation can be treated as grading sequence decision programming by Markov decision process (MDP), and the single integer coded method is used for current stage. The solution algorithm based on finite phase MDP is also proposed in the paper. The simulation results show that the computing rate and precision achieve well in solving multi-UCAV task allocation problem using this algorithm.