Research On UAV Path Planning

Due to its high performance and low cost,Unmanned Aerial Vehicle(UAV)has been widely used in military and civilian fields,and are used to perform various complex flight tasks.The environment and tasks faced by UAV are becoming more and more complex,and it is destined that UAV will inevitably develop towards autonomy and clustering.Path planning is the basis for UAVs to achieve autonomous flight,and to avoid obstacles and threats.And when single UAV is becoming more and more difficult to cope with increasingly complex tasks,the application of multi-UAV gradually begins to rely on the environment where multi-UAV perform tasks together.Therefore,path planning for a single UAV and collaborative path planning with multi-UAV have also become research hotspots on the road to autonomous clustering of UAV.For the path planning of single UAV,this thesis proposes a single UAV path planning scheme based on the firefly algorithm for the problem of unstable track and low convergence efficiency under dynamic threats that often occurs when meta-heuristic algorithms such as the firefly algorithm are applied to this problem.In this scheme,this thesis models and analyzes the environment and threats of single UAV path planning,and on this basis,proposes a trajectory planning algorithm based on adaptive adjustment strategy(AFF-TPA).The algorithm introduces a chaotic strategy to adjust the firefly’s absorption coefficient adaptively,and adjusts the control step update formula through the time-varying inertia weight to enhance its global search ability;and then introduces a disturbance factor based on the Boltzmann selection strategy to the iterative solution is perturbed to expand the search space of the track and enhance its convergence efficiency.Finally,the simulation proves that the AFF-TPA has a better track and more efficient convergence efficiency than the traditional firefly algorithm and other improved firefly algorithms.For the NP-hard problem of multi-UAV cooperative path planning,meta-heuristic algorithms such as ant colony algorithm are commonly used for optimization.This thesis analyzes the constraints and objective functions of the current ant colony algorithm and its improved algorithm,and proposes a multi-UAV collaborative path planning scheme based on hybrid optimization algorithm.This scheme generates a set of initial path sets by improving the state transition formula and pheromone update of the ant colony algorithm,and then uses it as the initial population to generate the global optimal track through the mixed optimization of the improved ant colony algorithm and genetic algorithm The curve is smoothed by the Bezier curve mechanism to obtain the final path suitable for UAV.The final simulation proves that the trajectory generated by the scheme proposed in this paper has higher stability,and has better performance in coordination and planning time.