Position Control of Robotic Manipulator Using Repetitive Control Basedon Inverse Frequency Response Design

Abstract

It is very common nowadays to see robotic manipulators performing repetitious work during the assembly process. One of their general tasks is to assemble electronic components which requires high position accuracy along the desired path. Conventional implementation with classical control schemes have proved capable of only limited performance. This paper demonstrates the efficiency achieved by using a repetitive control (RC) structure with the ability to learn from experience to reduce positioning errors from robotic manipulators in performing repetitive tasks. Firstly, the performance of the classical control method is illustrated to show the occurrence of nearly repeated patterns in tracking errors from each period. A simple RC design was then developed, with the aim of eliminating the tracking error in each repetition. However, some tracking errors were still noticeable and could even increase when using an inappropriate gain. An inverse frequency response from the RC design is therefore proposed by taking into consideration stability in the design criteria. The stability and robustness of the RC approach is verified by the experiments. The results show that the proposed method can reduce the tracking error by 99:22% for joint T and 95:03% for joint Z compared to the classical control method. In addition, the tracking error was reduced by 80% for both joints compared to the simple RC design.