Deterministic Learning-based Generalizable Trajectory Tracking Control for Permanent-Magnet Synchronous Motors Driven Two-Axis X-Y Table

Zixian Zhao, Yiming Fei, Chengyu Han, Jiangang Li

Abstract: In this paper, based on the deterministic learning theory and the model of permanent-magnet synchronous motors (PMSMs) driven two-axis X-Y table, a radial basis function neural network (RBFNN) learning control generalization rule for the non-repetitive trajectory tracking control of the system is proposed. Because of its excellent approximation capability, adaptability and learning capability for uncertain systems, RBFNN is used to approximate the model of X-Y table. Aiming to improve the generalization capability of deterministic learning theory, a generalization rule of deterministic learning theory for X-Y table is proposed. Based on the proposed generalizable deterministic learning control scheme, the tracking accuracy of the X-Y table is improved and the effectiveness of the generalization rules are verified through corresponding experiments.