Research And Optimization Of Area Target Exploration Path Planning Algorithms Based On Unmanned Aerial Vehicles

The rapid development of technology has led to increasingly sophisticated functionalities for unmanned intelligent machines,resulting in their widespread adoption across various industries for social production and everyday use.Unmanned aerial vehicles(UAVs),with their high mobility advantages,have found extensive applications in military,industrial,and civilian sectors.For instance,in large-scale inspection tasks such as forest monitoring,UAVs can significantly save manpower,material resources,and time compared to traditional manual inspections.However,UAVs are constrained by energy limitations,resulting in restricted flight ranges and lower efficiency.Therefore,finding the optimal flight path while balancing task completion time and coverage is of significant research importance.In this thesis,we focus on the UAV swarm coverage path planning problem for wide-area inspection tasks,aiming to optimize the task completion time while ensuring coverage.We conducted the following research:(1)To address the lack of a map risk model in UAV coverage path planning,we designed a two-dimensional heat map environmental model based on UAV structural characteristics and map modeling principles,accurately describing the risk information of the task area.We established a UAV mobility constraint model,including maximum endurance time and flight speed,defined the wide-area inspection task as a Minimum Time Maximum Coverage(MTMC)problem,and constructed a path cost model involving balance factors.This model takes into account constraints such as flight time and task point weight,providing a theoretical basis for algorithm design.(2)For the MTMC problem,we proposed a UAV Path Coverage Algorithm Based on the Greedy Strategy and Ant Colony Optimization(PCBGA)using a two-dimensional heat map.Firstly,we designed a sub-optimal point terminal judgment mechanism for the improved greedy algorithm,optimizing task point allocation and initial UAV path.Secondly,we applied ACO to further optimize the task point visitation order,shortening the flight path length and improving energy utilization and task safety.Finally,through simulation experiments,our algorithm demonstrated a 2.8% reduction in task completion time and a 4.4% increase in coverage compared to the WTSC and G-MSCR algorithms.(3)To address the slow convergence speed and susceptibility to local optima traps of ant colony optimization(ACO)when optimizing the MTMC problem,we proposed a cyclicity adaptive parameter ACO(CAPACO)algorithm.By combining CAPACO with the PCBGA algorithm,we further proposed a more efficient PCBGA-NEW algorithm.We improved ACO performance from three aspects: firstly,introducing a pioneer ant method to optimize the initial pheromone distribution;secondly,optimizing the state transition function based on improved pseudo-random strategy and adaptive parameters;and lastly,designing a periodic update strategy for global or local optimal path pheromone concentration.Through multiple comparative simulation experiments with ACO and MMAS algorithms on the TSPLIB standard test library,the CAPACO algorithm generated the lowest deviation rate of only 0.005% for the optimal path length.Moreover,the PCBGA-NEW algorithm maintained excellent coverage while reducing task completion time by 3.28% compared to the WTSC and G-MSCR algorithms,further improving the performance in solving the MTMC problem.