Path Planning And Risk Assessment Of Unmanned Surface Vehicle

The Unmanned Surface Vehicle(USV)has the advantages of modularization,high speed and intelligence.It is widely used in the military field by carrying various complex tasks with different sensors or equipment.Path planning and risk assessment are the important premises for safeguarding USV navigation safety and mission completion,having important research significance.USV path planning technology has always been the core technology of USV research.USV path planning mainly includes global path planning and local path planning.The research of collision risk index has always been research hotspot in the navigation field.Accurate collision risk index can provide effective auxiliary information for USV to ensure the safe navigation.Firstly,the theoretical basis related to USV path planning and risk assessment have been introduced.The path planning section of USV describes its basic concepts and general steps,detailing several common methods of environmental modeling.The risk assessment section of USV details the process of risk assessment and common assessment methods.Secondly,aiming at the shortcomings of the basic genetic algorithm to solve the USV global path planning,a USV global path planning method based on improved genetic algorithm is proposed.In order to overcome the shortcomings of slow convergence of basic genetic algorithms,this method proposes a new crossover operator.Compared with traditional crossover operators,it compares two non-bearing nodes instead of each single node to improve the convergence speed of the algorithm.To solve the problem of long path distance and low smoothness,an improved method based on genetic recombination path optimization algorithm is proposed.This method uses three kinds of operators of genetic recombination algorithm to optimize the path,shorten the path length and improve path smoothness.The simulation results show that the proposed method can be applied to USV global path planning,reducing the energy consumption of USV and ensuring the safety of USV navigation.Thirdly,aiming at the shortcomings of the basic Rapidly-exploring Random Tree(RRT)algorithm in the USV local path planning,a USV local path planning method based on the improved RRT algorithm is proposed.In order to overcome the randomness and uncertainty of the basic RRT algorithm,a method for optimizing the growth point and the extension point is proposed.This method optimizes the selection of growth points by adding the suppression factor of the failure number of the measurement nodes,and limits the expansion point by increasing the constraint conditions.The choice of the algorithm improves the search efficiency of the algorithm;in order to solve the problem of redundant navigation point in the path,the method uses the Dijkstra algorithm to optimize the path.The simulation results show that the proposed method can be applied to the local path planning of USV,reduce the rotation angle during USV navigation and improve the operability.Finally,for the USV risk assessment study,research from USV collision risk index describing the relationship between collision risk index and path planning decisions.The fuzzy comprehensive evaluation method and BP neural network algorithm are analyzed,and the two methods are used to evaluate the collision risk index.The simulation results show that the fuzzy comprehensive evaluation method and BP neural network algorithm have higher accuracy,which corresponds the actual situation and the ship’s demand for collision risk index accuracy.