Research On Path Planning Of Multiple Uavs In Environmental Monitoring

With the accelerated pace of social industrialization,while the economy continues to develop,natural and man-made environmental damage is also frequent,so the country has increased the emphasis on environmental protection.Environmental monitoring is an important means to detect environmental pollution in time and to prevent and control disasters.What kind of effective monitoring method is gradually becoming a hot research topic in the industry.The traditional environmental monitoring method mainly relies on the manual carrying of detection equipment for on-site detection,which is not only inefficient,and the safety of the staff can not be guaranteed.UAV is widely used because of its small size,easy control and other advantages.The combination of UAV technology and environmental monitoring can effectively improve the efficiency of environmental monitoring and effectively guarantee the safety of staff.Considering that the energy consumption of UAV is limited when it is used to complete environmental monitoring tasks,the shorter the path is,the less energy UAV will consume.Therefore,reasonable path planning plays a crucial role in successfully completing environmental monitoring tasks.In summary,aiming at the problem of multi-UAV collaborative path planning in environmental monitoring,this dissertation carries out the following research work:(1)This dissertation summarizes the research background,significance and research status at home and abroad of the application of multi-UAV path planning in the field of environmental monitoring,and expounds the UAV environment modeling method,the algorithm to be used in the single UAV path planning and the relevant theoretical knowledge of multi-UAV task assignment.The above introduction provides relevant theoretical knowledge for the follow-up work of single UAV three-dimensional path planning and multi-aircraft multi-target point task assignment.(2)An improved particle swarm optimization(PSO)hybrid algorithm was proposed to solve the three-dimensional path planning problem of single UAV.Based on the cost of path length,height,collision,turning Angle and climbing Angle,the fitness function is constructed.The particle swarm optimization algorithm is combined with genetic algorithm and harmonic search algorithm.At the same time,the acceleration idea of gravity search algorithm is introduced,and dynamic updating learning factor and dynamic inertia weight of concave function model are set.By comparing the improved algorithm with some other algorithms,the simulation results show that the improved particle swarm optimization algorithm has significant advantages in path length,convergence speed,average consumption time and stability in both simple and complex environments.(3)Aiming at the task assignment problem of multi-UAV in environmental monitoring,an improved k-means clustering algorithm is proposed,which improved the initial clustering center by introducing fuzzy c-means clustering algorithm and particle swarm optimization algorithm to improve the clustering quality.In order to verify the effectiveness of the improved clustering algorithm,Iris and Wine data sets in machine learning were used to test it,and comparison experiments are set to compare it with the traditional k-means clustering algorithm and kmeans++ clustering algorithm.The experimental results show that,the improved clustering algorithm achieves better results than k-means clustering algorithm and k-means ++ clustering algorithm in accuracy and error squared sum.(4)In order to solve the problem of path planning after multi-machine task assignment,the simulated annealing algorithm is used to determine the sequence of task points,and the improved particle swarm optimization algorithm in single-machine path planning is adopted to avoid obstacles.Finally,the multi-unmanned units can effectively monitor the environment of each task point on the basis of obstacle avoidance and mutual collision avoidance.The improved kmeans clustering algorithm in this dissertation is applied to the actual UAV simulation experiment,and it is compared with the traditional k-means clustering algorithm.The simulation results show that,no matter the total path length of UAV flight,or the time taken to complete the environmental monitoring task,the improved clustering algorithm in this dissertation has more advantages than the traditional k-means clustering algorithm.