Path Planning of USV Based on the Improved Differential Evolution Algorithm

Planning a reasonable path and avoiding collisions with surrounding obstacles are among the most critical aspects of Unmanned Surface Vehicle (USV) navigation, which has drawn considerable attention from researchers in recent years, with various heuristic and intelligent optimization algorithms being applied to path planning. However, most existing algorithms have not sufficiently integrated safety and economy, leading to the planned paths that may not align with maritime practice. Therefore, to tackle the aforementioned issues, this paper introduces a differential evolution algorithm (DE) with an adaptive crossover factor for path planning and collision avoidance in USV. The collision risk index (CRI) is integrated with the DE, and the CRI is improved by introducing a restriction factor. The experimental results demonstrate that, compared with the other three algorithms, the improved DE exhibits greater advantages in terms of closest distance to the encountered ship, closest distance to obstacles, and total yaw distance, thereby validating the effectiveness of the algorithm.