Path Planning Of Multi-UAV Based On Secure Communication In Three-dimensional Space

Multi-UAV are widely used in modern combat missions because of their good coordination and high fault tolerance.To achieve path planning while maintaining secure communication among multi-UAV is an urgent problem to be solved.In this paper,we study the above problems.The main contents are as follows:(1)In the cooperative control,it is considered from two aspects: the clustering of multi-UAV and the obstacle avoidance of multi-UAV.Aiming at the slow convergence rate of some nodes during the flocking process,the clustering algorithm is proposed to make these nodes quickly converge by adding flight control factors.It is shown that clustering algorithm can quickly,smoothly and steadily complete the flocking process of UAV group.Then,aiming at the problem of fewer communication links between UAVs during obstacle avoidance,the clustering obstacle avoidance algorithm is proposed to increase the communication links by finding out the nodes that have no communication link with the UAV group and making them move closer to the virtual leader.Experimental results show that the given algorithm increases the average number of communication links of UAV group by 40.(2)In the secure communication,the hierarchical virtual communication ring strategy is proposed to reduce insecure range.In the process of UAV group communication,there will exist the insecure area.The hierarchical algorithm is designed to find UAV group boundary nodes,and the range of insecure area can be decreased by reducing the communication range of boundary nodes.The movement algorithm is designed to make the boundary nodes move closer to the group,so as to effectively communicate.The effectiveness of this strategy is proved by a large number of experiments.(3)In the path planning,Q-learning algorithm is combined with Dijkstra algorithm to study the Multi-UAV path planning.Firstly,the grid method is used to build a map,and the reward matrix with map information is generated.Then the Q-learning algorithm is used to calculate the reward matrix,thus obtaining the weight matrix.Finally,according to the actual situation,the flight direction of the UAV is limited.And the Dijkstra algorithm is used to find the shortest path.The presented algorithm is compared with the A* algorithm,and it is verified that the given algorithm can finish the path planning in a shorter time.This paper studies the path planning of multi-UAV based on secure communication in three-dimensional space and provides an effective solution to the multi-UAV's flight mode in battlefield environment.It makes a valuable attempt for future research.