Abstract: To address the drawbacks of slow convergence speed and poor finding accuracy of sparrow search algorithm (SSA) in AGV path planning, a chaotic sparrow algorithm based on the falling mechanism(SSA-CD) is proposed to solve the AGV path planning algorithm. Firstly, Sinusoidal chaos mapping and variable scale chaos strategy are introduced to initialize the population and improve the population diversity so that the algorithm has the ability to jump out of the local optimal solution. Finally, the optimal solution is further optimized by hermite interpolation to obtain shorter and smoother path paths. The effectiveness, feasibility and robustness of the improved algorithm are demonstrated by simulation experiments with raster maps.