Research On Multi-AUV Cooperative Detection Path Planning Technology Based On Marine Environment

In recent years,with the development of autonomous underwater vehicles,underwater acoustic target detection has become one of its main applications.Compared with single AUV detection,a cluster composed of multiple AUVs not only has strong fault tolerance,but also improves the effectiveness of mission operations.Considering that the marine environment is complex and changeable,sonar detection performance depends on the impact of the marine environment.The effective application of marine environmental information to AUV path planning and sonar processing to improve detection efficiency/AUV’s own concealment has become one of the key and difficult problems in underwater acoustic engineering research in recent years.In this paper,based on the problem of multi-AUV coordinated detection based on the marine environment,the research on path planning technology is carried out.The specific research content is as follows.First,according to the different types and requirements of target detection tasks,the problems are divided into two types: area coverage problems and target search problems.This article first studies the area maximum coverage problem,uses environmental information to analyze the detection capabilities of nodes at different locations in the area,constructs the sound field in each direction of the node,and converts the propagation loss into the detection probability through the sonar equation,and then converted into detection radius according to the detection probability threshold.Using the detection radius information,design a multi-AUV path planning method based on the ant colony algorithm,in which the A* algorithm is introduced to prevent the ants from falling into local extreme points during the pathfinding process,so as to maximize the area coverage in the limited operation time Demand.Through the simulation of the environmental field assimilation data in the South China Sea,the simulation results show that the algorithm can adjust the path according to the planning time,the number of AUVs,and the starting point position to achieve optimal path planning based on node detection capabilities,and verify the performance of the algorithm in multiple scenarios.Secondly,for the search problem of moving targets with unknown target states,a hidden Markov model is used to model the problem,and the influence of marine environment information on sonar detection is added to the model.A genetic algorithm-based multi-AUV collaborative target search planning method is designed.The genetic algorithm searches from the set of solutions and has a good effect in solving the planning problems of multi-AUV operations and time-varying environmental information.Through the simulation of the environmental field assimilation data in the South China Sea,and the use of scenarios that are prone to fall into the local optimal solution to test the algorithm,it is verified that the algorithm has a strong global search ability in solving the multi-AUV target search problem.Finally,aiming at the problem of AUV maneuvering strategy after target discovery,an AUV optimized detection/concealment path planning method based on the target source sound field distribution is proposed.After finding the target,this path planning strategy is tracked by one AUV to provide target location information,and other AUVs perform hidden path planning.By constructing the sound field of the target sound source,analyzing the detection probability and the probability of being detected at each position in the space where the AUV and the target are located,an optimized detection/concealment path planning method based on the ant colony algorithm is designed,according to the sound field The characteristics of complex and uneven distribution,the introduction of reward and punishment factors in the pheromone update rules of the ant colony algorithm to improve the global convergence ability of the algorithm.In the simulation,the effectiveness of the algorithm is first verified in two-dimensional and three-dimensional scenes,and finally the changes in the time dimension are added on the basis of three-dimensional,and the four-dimensional path planning simulation verification is performed.