Performance Improvement of Pneumatic Isolation Table Control with a Non-linear Quantizer

Some pneumatic isolation tables used to suppress vibrations from the floor in the manufacturing and measuring fields of precision components can be controlled to suppress vibrations using input from an on-off valve. However, the quantizers used to acquire the inputs of these pneumatic isolation tables are generally non-linear, and response deterioration appears when feedback modulators or optimal dynamic quantizers are used. To solve this problem, Koike et al.'s previous research proposed a method that selects an input with a low response that deteriorates based on the Lyapunov function considering the quantization error. In this method, the input is applied only when the difference of the Lyapunov function is negative; otherwise, zero-input is applied. In this paper, we propose a method to select a more suitable input by using a switching function that determines the necessity of input by comparing the values when the difference of the Lyapunov function is non-negative. Furthermore, the improvement of the control performance of the pneumatic isolation table by the proposed method is confirmed by simulation and experiments.