Research On High-efficient 3-D Position Deployment And Path Planning Of UAV Array

With the continuous iterative upgrading of mobile networks,people’s demand has shown an explosive growth.These services have higher requirements for quality of service and service delay,which is a challenge to the traditional communication infrastructure.Therefore,UAV has attracted wide attention due to its high flexibility,low cost and easy operation.However,the communication channel between the UAV and the ground,the interference between the UAV and the limited energy of the UAV all have an impact on the deployment effect and movement path of the UAV,thus affecting the performance of the UAV.Therefore,the location deployment and path research of low energy consumption is of great significance for UAV auxiliary network.In order to achieve better coverage performance,multiple UAVs need to find the best location under the limitations of limited energy and cooperative coverage,so as to achieve low energy consumption and cover as many ground users as possible.However,there is a trade-off between coverage utility and energy consumption.UAV has good performance in alleviating the instability of data link and has been applied to relevant networks to speed up data collection and transmission.However,when multiple UAVs cover one sensor at the same time,it indicates that the coverage area of UAVs is overlapped.Multiple UAVs will not be selected at the same time,which will lead to low energy efficiency.This thesis studies the high-efficient deployment and path planning of UAVs.The main research contents are as follows:(1)The three-dimensional deployment and power consumption of multiple UAVs are studied.Based on the channel characteristics and energy consumption model of UAV communication,considering the influence of the number,three-dimensional position and speed of UAVs,a multiobjective optimization problem of coverage utility and energy consumption is formulated,which aims to maximize the coverage utility and minimize the energy consumption of UAVs at the same time.Due to the complexity and NP-hard of the problem,an improved multi-objective grey wolf optimization algorithm is proposed.In this algorithm,the role determination algorithm is first proposed to determine the upper limit of the number of UAVs.Then,a mixed initial solution scheme is proposed to improve the diversity of the initial solution,and the idea of Levy flight and Sine-Cosine methods are used to update the solution to obtain multiple UAV deployment strategy.The simulation results show that the proposed algorithm can achieve high efficient and low-power deployment than other benchmark methods.(2)The problem of energy saving cooperative path planning for multiple UAVs is studied.Based on the constructed multi-UAV auxiliary sensor data collection system,the optimization objective of maximizing the data collected by UAV under the constraints of energy and coverage area is proposed.Because the correlation between the UAV and the environment is unknown,the deep reinforcement learning algorithm is improved to obtain trajectory.Considering the coverage and energy cost of UAV comprehensively,the moving direction of UAV is determined and the path is represented.In order to solve the problem of UAV cooperation,the hexagonal region search algorithm is used to obtain the location information of the global UAV.The reward function is determined by the coverage constraints and energy constraints of the UAV.Finally,adaptive learning rate is proposed to improve the convergence performance of the algorithm.The experiment illustrates that the proposed algorithm can effectively achieve the data collection of multiple UAVs in unknown environment at low energy cost.