In order to solve the repetitive motion path control problem of lens sorting robot under imprecise kinetic model, simplified Lagrange dynamics model of the robot was built by ignoring weight of follower arm, moving platform and loads. Dynamics control coupling of active arms was released and control was exerted to each single active arm. The iterative learning control algorithm was applied to each single active arm trajectory control. Active arm trajectory tracking control system of lens sorting robot based on ILC was built in Matlab. Trajectory tracking accuracy of the system was verified by simulating , the results show that precision of the active arm trajectory tracking control system is high under finite number of iterations.