Research On UAV Path Planning Algorithm And Its Application

Because of Unmanned Air Vehicle(UAV) zero casualty, with low using cost, strong survival ability and strong mobility, it has broad application prospect in military field and civil field especially. UAV is a very popular field in recent years.Intelligent degree is an important measurement index for UAV's performance. The UAV path planning plays an important role in UAV mission planning system. It's very essential to UAV's autonomous flight implementation and finish the UAV's flight mission. For a given area of reconnaissance(threat deployment and target distribution, ground-to-air defense),advanced planning and scheduling help the UAV complished the flight mission, it can minimize the possibility being discovered by eneymies and reduce the flight distance, enhancing UAV's survivability and improve reconnaissance efficiency.This dissertation concentrated on the research works listed below and achieved some creative results:(1) Put forward an new optimization algorithm called Quantum-behaved Particle Swarm Optimization(QPSO) and used it in UAV's path planning. It used to solve the problem that Particle Swarm Optimization(PSO) easily falls into a local extremum. Simulation results indicate that QPSO method has the advantages of fast convergence, also on the premise of threat avoidance, it obtains excellent solution of the UAV mission planning. Compared on PSO algorithm, experimental results verify the validity of our method.(2) Actually, the threat being met on the period of UAV flying in practice are very different, the threats are irregular. We know the previous studies considered all threat as identity, this be divorced from reality. So we puts forward the solution of the threat avoidance of irregular objects.(3) The objective function is constituted based on the factors of the flight distance and the antiaircraft threat. We take the function of flight distance combined with penalty function, on the premise of threat avoidance, the cost of flightway are least. Also according to the low altitude detection features, we create space flight environment and establish flight modeling. All experiment are in two-dimensional space, its easy to simulation.