3-Dimensional UAV Path Planning Based On Improved Particle Swarm Optimization Algorithm

Unmanned aerial vehicle path planning is an important part of the UAV mission planning system, it is also the key technology that achieving the penetration investigation, threat circumvent, terrain following and other tasks of the UAV. The key points of the path planning are the speed and exactness, to a certain extent, the quality of the path planning depends on the positive effects and negative effects of the algorithm and the solving strategies of the specific problem.In response to this situation, this thesis proposes a Route Planning stratification concept which improves the security of the track effectively and reduces the information processing and computation of the onboard computer significantly to overcome the obstacles of the path planning which is caused by the airborne equipment performance. In addition, this thesis analyzes the parameters and the convergence of the PSO, which proposes a new improved algorithm. This algorithm presents a weighting factor in the location update formula to limit the speed of particle which is in order to avoid the defect of the standard PSO, for instance falling into local optimum easily, low accuracy and searching blindly in the solution space. This thesis also does some work on the equivalent treatment of the sudden threat in the field digital map, which is to simulate the sudden threat to a mountain terrain and do pretreatment for the equivalent digital map by the smooth handing. The fitness function was built through the terrain threat, performance limitations of the UAV and range restriction. Finally, using MATLAB7.1to establish the equivalent of digital maps, and simulated the static path planning and dynamic path planning respectively, when there is a sudden threat, and changed the weight values in the fitness function to simulate the different strategies of the decision-maker, who responds to the sudden threat, to get the specific path.Through using the different initializations of the static path planning and dynamic path planning, and changing the parameters of the fitness function, the simulation results show that this method can effectively improve the quality of the flight path and make the rapid and effective reflection for sudden threat. In addition, in the aspect of the selection of the strategy, this method provides the decision-makers a broader range of options. To a large extent, compared to the traditional route planning method, this method is able to guarantee the UAV flight safety, improve the quality of the path, reduce the requirements of the performance of the airborne equipment effectively, and reduce the cost of the flight. In the future, the field of UAV flight path planning has a broad development prospect.