Research on UAV Collision Avoidance Path Planning Based on MDP
Target search path planning of UAVon the premise of collision avoidanceis to find the target in the faster and more effi-cient form by reasonable flight path planning in the complex and numerous environmental obstacles.Discussed the law of finite posi-tion Markov movement under barrier-free environment and constructed the corresponding Markov movement distribution model.Based on the cutting-edge research results of search system trajectory planning,combined with the MDP theory,the negative reward mechanism was innovatively introduced to iterate the Q-Learning strategy algorithm,and the single UAV target search model was constructed.By analogy with the"risk well"visualization method,the negative reward effect of the obstacle threat area on the UAV was intuitively presented,and the single UAV target search path planning model under complex obstacle constraint environment was constructed.and the simulation experiment showed that the algorithm is feasible,Which has certain reference significance for the de-sign of the route planning algorithm.
UAVpath planningcollision avoidancestatic target searchMDP(Markov decision process)risk well
NETL
NSTL
万方数据
