UAV Path Planning For Persistent Monitoring Based On Improved Value Iteration

Persistent monitoring path planning for unmanned aerial vehicles is a hot research topic.With the rapid development of the existing UAV hardware and intelligent optimization algorithms,the level of intelligence of single UAV has been greatly improved,and the leap from single UAV intelligence to multi-UAV cooperative intelligence has been realized.Aiming at the information security of monitoring path,this paper studies the continuous monitoring path planning problem of single UAV and multi-UAV under time and security constraints.Firstly,this article improves the modeling method of existing algorithms.In order to improve the timeliness and safety of persistent monitoring tasks,this paper proposes a new value function iteration method suitable for unmanned aerial vehicle path planning.The traditional value function iteration algorithm associates the reward return function with the action,without considering the timeliness of the target point to be monitored.However,in practical applications,it is required that the target point be accessed by drones within the maximum allowable monitoring period.This article applies the εprinciple improves the randomness of path point selection,and uses the information entropy of the monitoring path to evaluate the randomness.Finally,a persistent monitoring path with good safety of the monitoring path is obtained.Secondly,aiming at the problem of persistent monitoring of multiple UAVs,this paper establishes a path planning model for persistent monitoring of multiple UAVs.In order to ensure the synchronous update of the UAV status,the discretization preprocessing method is used to divide the road network into multiple segments of the same length.In order to improve the randomness of the persistent monitoring path,the roulette method is used to randomly select the extended virtual nodes,and the feasibility of the algorithm is verified in the incomplete road network environment.Finally,the verification of the designed algorithm is carried out by using the physical experiment platform,and the results show that the deviation between the flight path of the UAV and the simulation experimental path is small,and a good experimental effect is obtained.The results of simulation and physical experiments jointly verify that the proposed method can effectively improve the efficiency and security of UAV persistent monitoring.