A method of fractional sliding mode variable structure control method based on neural network optimized by particle swarm optimization (PSO) is studied and applied to the path tracking of industrial robots. Firstly, the neural network optimized by particle swarm optimization is used to identify the system model of industrial robots, and the model most relevant to the resolution of system control parameters is trained. Then, the fractional order sliding mode variable structure controller is designed based on fractional order theory and sliding mode variable structure theory, which is applied to the path tracking of industrial robots as the main controller.Simulation and experimental results show that this method has good tracking performance, fast and robust.