基于DDQN的孔径分割多功能雷达任务调度方法OA

In order to enhance the task execution efficiency of multifunction radar with aperture partition,a double deep Q-networks(DDQN)based dynamic task scheduling method is proposed in this paper.Firstly,the basic task model and task scheduling model of multifunction radar are established according to the task characteristics of multifunction radar with aperture partition.Then,by analyzing the radar multi-task state information and its change process,the task scheduling process of multifunction radar is con-structed as a Markov decision process,in which the state space,action space and reward function are designed.Finally,consider-ing the influence of current task scheduling decision on subsequent task scheduling decisions,a DDQN-based dynamic task schedu-ling method is formulated to realize efficient dynamic task scheduling for multifunction radar with aperture partition.Simulation re-sults show that compared with the traditional task scheduling method for multifunction radar with aperture partition,the proposed method significantly reduces the task drop ratio of multifunction radar in a high load environment.