Global path planning and multi-objective path control for unmanned surface vehicle based on modified particle swarm optimization (PSO) algorithm

This paper investigates the novel path navigation method for an unmanned surface vehicle (USV), which is divided into two-stage: global path planning and path control. In the first stage, combined with the travelling salesman problem (TSP), a global path is obtained by maximizing the profit per unit time in multiple task locations. In the second stage, a nonlinear multi-objective optimization model is formulated for the path control between two task locations. In addition, fixed and time-varying currents are also considered for USV motion, which aims to avoid collision and take full advantage of the direction of currents. To solve the problem quickly and accurately, a chaotic and sharing-learning particle swarm optimization (CSPSO) algorithm is developed to solve the extended TSP and the nonlinear multi-objective model. Simulation experiments validate that the proposed hierarchical navigation method, CSPSO algorithm, and collision avoidance rules are effective and justifiable.