Optimal Search Strategy For UAVs Based On Markov Mobile Targets

Nowadays,the power of technology is pervasive and permeated in all walks of life.As a new scientific and technological force,the application of UAVs in various industries has attracted much attention.The search and rescue using UAVs in the military target search system have drawn increasing attention of all countries in the world.Route planning is an important problem to be considered in the application of UAVs to the target search system.The route planning of the UAV target search system consists of two major categories,that is,the route planning of the static target search system and the path planning of the moving target search system.At present,the research results of the route planning of the static target search system are relatively mature and have been widely used in practical problems.But the research on route planning of the moving target search system is relatively less.Up to now,the research direction of moving target searching system is mainly focused on the design and application of the search algorithms,but there is no strict mathematical definition for the motion law of the target in search process.In view of this phenomenon,firstly,this paper studies the motion law of the search object---limited position Markov moving target---from the perspective of probability theory,and builds the probability model of the target.Secondly,on the basis of the existing research results of the UAV search system route planning,the optimal search problem for the UAV for a Markov moving target is studied.This thesis uses the theory of Markov decision process(MDP)to establish the search model of a single UAV and multiple-UAVs on a single Markov moving target.And the policy iteration algorithm of traditional MDP model was improved and optimized,so that the UAV has higher search efficiency,and better overall search track.For a single UAV search problem,a multi-stage heuristic strategy iteration algorithm(MDP-MSHPI)based on MDP is proposed,and the optimal search path is planned by this method.To solve the problem of multi-UAV search for a Markov target,a cooperative path planning mechanism based on MDP model for multiple UAVs is proposed in this paper.In addition,five cooperative task allocation search strategies are studied,and path conflict between UAVs in the multi-UAV cooperative search system is studied,and a track coordination strategy based on information interaction is proposed.Based on the above research,this paper builds a simulation platform for the single UAV and multi-UAV search system using MATLAB tools.In the simulation platform of the single UAV target search system,scanning search algorithm,random search algorithm and greedy search algorithm are compared with the MDP-MSHPI search algorithm,which highlights the advantages of the MDP-MSHPI search algorithm for the search performance of a Markov moving target.In the multi-UAV target search system platform,the advantages and disadvantages of the five different task allocation strategies for multi-UAV aiming at Markov moving target search are compared,and the influence of the number of cooperative UAVs on the search efficiency is studied under the same task allocation strategy.Finally,the search algorithm based on MDP multi-UAV cooperative task allocation algorithm,the multi-UAV random search algorithm and the multi-UAV greedy search algorithm for the Markov moving target search track and search step are simulated,and the effectiveness of the collaborative search algorithm is verified.