Path Planning of USV Based on Improved Hybrid Genetic Algorithm

Abstract

In recent years, with the continuous development of economic globalization, the protection of maritime rights and interests of countries around the world has become increasingly strong, and the development of marine resources has become a focus of international attention. Due to the rise of unmanned aerial vehicle(UAV) and other intelligent devices, unmanned surface vehicle(USV) is receiving unprecedented attention. The path planning is the key to ensure the safe navigation of USV in the complex and changeable marine environment. In order to solve the problem of traditional genetic algorithm, such as the lack of searching ability and the large amount of calculation, a method based on genetic algorithm and simulated annealing algorithm is proposed to plan the optimal path of USV. In this paper, the electronic river map is used to select the water area of the Yangtze river near honghu city, hubei province, China. The grid method is used to establish the model of the environment, and the insertion operator and deletion operator are added to improve the efficiency of population evolution and optimize the generation path. Matlab simulation results show that in complex marine environment, this algorithm can quickly and effectively obtain smooth optimal path, and compared with the traditional genetic algorithm, the convergence speed and search quality has been significantly improved.