In order to improve the optimization function of artificial fish swarm algorithm in path planning, the improved visual range and crowding factor function are used to improve the optimization work of fish swarm algorithm in robot path planning. In the traditional fish swarm algorithm, the visual range is constant. The visual range determines the global and local work of the optimization, and the crowding factor has an effect on the convergence of the algorithm. At the same time, in the traditional fish school algorithm, the optimal solution is chosen to execute every time, which often leads to the global optimal and local optimal mutual interference in the grid environment, which leads to the unreasonable path planning. Therefore, by using the improved visual range crowding factor, the feasible solution is recorded at the same time. When the fish school finds the target point, the track of the fish school that finds the target point is recorded to form the path planning In the feasible solution, the shortest path is the best to ensure the rationality of path planning. Compared with the traditional fish school algorithm, it is proved that the research algorithm has better optimization work in path planning. Through MATLAB simulation experiment, the effectiveness and stability of the algorithm are verified.