This paper presents a practical and efficient path planning algorithm based on the architecture of conventional particle swarm optimization (PSO) for a self-developed unmanned surface vehicle to complete tasks of water quality detection and sampling. A preprocessing orientation angle-based grouping strategy appropriately enhances computing efficiency and adaptability to the distribution feature of planned points. Meanwhile, mutation is introduced in the initialization phase to improve the global search ability of the PSO. Multiparticle competition is added to maintain population diversity during late iterations. After stringing all the subdomain paths together, a simplified 4-opt is employed for further improvement and convergence acceleration. Computational simulations show that the proposed algorithm is superior to three of our previous algorithms and comparable with several existing algorithms from other reports. The path-planning algorithm is also integrated into the vehicle's navigation, guidance, and control system with satisfactory feasibility.

本文提出了一种基于常规粒子群优化（PSO）架构的实用高效的路径规划算法，用于自行开发的无人水面车辆来完成水质检测和采样任务。基于预处理方位角的分组策略可以适当地提高计算效率和对计划点分布特征的适应性。同时，在初始化阶段引入了变异，以提高PSO的全局搜索能力。增加了多粒子竞争，以在后期迭代期间维持种群多样性。在将所有子域路径串在一起之后，采用简化的4-opt进行进一步的改进和加速收敛。计算仿真表明，所提出的算法优于我们之前的三种算法，并且可以与其他报告中的几种现有算法相媲美。路径规划算法也被集成到车辆的导航，制导和控制系统中，具有令人满意的可行性。