The palletizing robot was analyzed through its constructure, and constructed its kinematics model and expressed the function of the model. On this basis, used the cubic spline function to carry out reasonable trajectory planning according to the working conditions of the robot, and established the time and jitter as the most optimization model of optimal target. And then used particle swarm optimization to optimize the target model. Through further experimental verification, the results show that the trajectory planning method proposed in this paper can improve the efficiency of the palletizing robot while ensuring stability.