A method for programming a repeating motion of a redundant robotic arm on the basis of a variable parameter convergence differential neural network, comprising the following steps: (1) establishing an inverse kinematics equation of the redundant robotic arm on a velocity layer via a track of an end of the redundant robotic arm; (2) designing the inverse kinematics issue in step (1) as a time-varying convex quadratic programming issue constrained by an equation; (3) introducing a repeating motion indicator in the time-varying convex quadratic programming issue; (4) converting the time-varying convex quadratic programming issue, into which the repeating motion indicator is introduced, into a time-varying matrix equation by using Lagrangian function; (5) solving the time-varying matrix equation by using a variable parameter convergence differential neural network; and (6) integrating an optimal solution, obtained in step (5), of the redundant robotic arm on the velocity layer to obtain an optimal solution of a joint angle. Using the variable parameter convergence differential neural network to solve repeating redundant mechanical motion results in high computational efficiency, high real-time performance, and robustness.