A One-Parameter Structure for Adaptive Iterative Learning Control of Robot Manipulators

In this paper, a new one-parameter structure is presented for the design of adaptive iterative learning controller for repetitive control of robot manipulators. Only one control parameter is needed for compensation of the unknown plant information so that the size of memory for storage of previous control information can be greatly reduced. In addition to the new control structure, an adaptive algorithm combining time-domain and iteration-domain adaptation is also proposed. The combined adaptive law can relax the requirement of certain unknown upper bound usually used in current existing adaptive iterative learning control schemes. The technical analysis shows that the boundedness of all the internal signals and convergence of the learning error are guaranteed without using projection mechanism in adaptive laws. Finally, a simulation result is provided to illustrate the effectiveness of the learning controller.