Research On Optimization Method Of AUV Path Planning Under The Influence Of Dynamic Environment

Nowadays,in order to strengthen the development and utilization of marine resources,a novel marine exploration equipment autonomous underwater vehicle(AUV)has become a research hotspot because of its high sensitivity and strong safety.AUV typically performs tasks in complex marine environment,which are often affected by dynamic marine environment during navigation,where dynamic obstacle environmental information can affect the safety and stability of AUV navigation.Therefore,the AUV route planning under the influence of dynamic environment is a valuable research problem.In this paper,a two-layer algorithm based on improved level set algorithm to deal with global planning and local planning algorithm of using rolling window fuzzy control is proposed.The algorithm can realize path time optimization and real-time collision avoidance.The main work of research of this paper include:Firstly,the AUV route planning problem is described,and the problems which need to be solved in the research process are proposed.Then,the environmental modeling methods applicable to different environmental models and the evaluation criteria for planning the navigation path are given.In addition,the mathematical theory basis of the level set algorithm is introduced in detail,the data structure of the algorithm is analyzed,and the implicit interface and symbol distance function are proposed to describe the evolution of the level set equation and the level set curve.Secondly,the level set algorithm is modeled and applied to the route planning.The level set function is used to describe the level set curve.The evolution of the zero level set represents the navigation of the AUV.It will keep evolving until it converges to the end point for the first time,then backtrack from the end point to the iteration,and then obtain the time optimal global path.In order to solve the problem of low efficiency of traditional algorithms,this paper improves the level set algorithm and improves the scanning method of fast scanning method.By combining the reconstructed narrow band and the improved fast scanning method,the efficiency and accuracy of the algorithm are improved.At the same time,in order to solve the time lag problem in the system,the Hopf formula is used to solve the viscosity problem and the online trajectory optimization with delay system.Thirdly,in order to solve the influence of dynamic obstacles existing in the AUV navigation process,a rolling window fuzzy control algorithm is proposed.A novel fuzzy controller is designed to process the information detected by the rolling window,and the fuzzy controller design scheme such as input and output variable selection,fuzzy inference,anti-fuzzification,and improvement of the fuzzy controller are added.The S-plane control algorithm is designed with a variable universe fuzzy controller.The controller is simple and robust to model,which greatly improves the efficiency of the algorithm.The local collision prediction and collision avoidance strategy under the influence of dynamic environment are given.Finally,in order to solve the time optimal path and real-time problem of AUV path planning under the influence of dynamic environment,this paper uses two-layer algorithm combination to realize path planning.The improved level set algorithm is used for global road strength planning,and then the local path planning in dynamic environment is performed based on rolling window fuzzy control.The design simulation experiment verifies the feasibility and optimization effect of various improved algorithms.The experimental results show that the improved algorithm is feasible and improves the efficiency of the algorithm.