针对基本人工鱼群算法全局最优解精度不高,后期收敛效率降低等问题,笔者设计了人工鱼的视野与步长分段自适应策略用以改进基鱼群算法。根据煤矿救援环境的特点,在二维平面内建立环境模型,采用威胁区域距离检测的方式处理优化约束条件,并将改进后的鱼群算法应用于煤矿救援机器人路径规划。仿真结果表明,分段自适应鱼群算法性能明显优于基本鱼群算法。 In order to solve the problem that the global optimal solution of the basic artificial fish swarm algorithm is not accurate and the efficiency of convergence is low at the later stage, the author designed the artificial fish’s field of view and the step-by-step adaptive strategy to improve the basic fish swarm algorithm. According to the characteristics of coal mine rescue environment, an environment model is established in two-dimensional plane, the optimal constraint condition is dealt with by threat zone distance detection, and the improved fish-swarm algorithm is applied to path planning of rescue robot in coal mine. Simulation results show that the PSO algorithm performance is obviously better than the basic fish algorithm.