Abstract: For dealing with grinding, the robot kinematics model is designed and established, the robot trajectory is planned, and the trajectory curve is optimized by an improving particle swarm optimization algorithm to reduce the running time and improve the working efficiency of the robot. Firstly, this paper establishes the spatial kinematics model of the industrial robot for polishing, calculates the joint space inverse solution position corresponding to the target pose points in Cartesian space, uses the 3-5-3 three segment polynomial curve to plan the trajectory of the inverse solution points in joint space, and takes the shortest trajectory motion time as the optimization goal. By using the improved particle swarm optimization algorithm with immune operation to optimize the trajectory curve, the optimization results of the improved algorithm are compared with the traditional particle swarm optimization algorithm, and it is found that the improved new algorithm can better avoid falling into local optimization, the fitness result is better, and the algorithm effect is better.