System Identification and Vision-Based Motion Control for a Motor-Toggle Mechanism

• Main Authors: Kun-Yung Chen, 陳崑永
• Other Authors: Rong-Fong Fung
• Format: Others
• Language: en_US
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/42443001816125214752

碩士 === 國立高雄第一科技大學 === 機械與自動化工程所 === 93 === In this paper, the punching machine is made up by a toggle mechanism driven by a permanent magnet (PM) synchronous servomotor. First, Hamilton’s principle, Lagrange multiplier, geometric constraints and partitioning method are employed to derive its dynamic equations. The system parameters are difficult to obtain if the mechanism’s components can not be taken apart. The recursive least-squares (RLS) method is implemented to identify these parameters. This paper presents the comparisons of the visual servoing feedback motion control with the fuzzy logic control (FLC) and adaptive controller by using the stability analysis with inertia-related Lyapunov function to the punching machine. The main purpose of the punching machine is to transport the work pieces to a fixed position for manufacture. To satisfy the demand of the machine performance, three controllers including the FLC and adaptive controller are designed to control the slider responses. Distinguish from the previous studies, the vision servo system of an non-contact measurement of a charge coupled device (CCD) camera is employed in this paper to control the color pattern output state instead of using the expensive linear scale or encoder of the motor-mechanism coupled system. Finally, from the well agreements between numerical simulations and experimental results, it is convinced that the proposed controller by using of machine vision is robust to external disturbances for a punching machine system.