High-speed machining technology is becoming increasingly important in mod-ern manufacturing industry. With the requirement of high speed to yield high productivity, the precision and accuracy requirement becomes more and more stringent. In this thesis, research is focused on high-precision trajectory tracking control for high performance motion systems....[ Read more ]

High-speed machining technology is becoming increasingly important in mod-ern manufacturing industry. With the requirement of high speed to yield high productivity, the precision and accuracy requirement becomes more and more stringent. In this thesis, research is focused on high-precision trajectory tracking control for high performance motion systems.

The dimensional accuracy of the finished part is mainly determined by the contouring error which is different from the conventional tracking error. However, it is not easy to compute for arbitrary curves. In this work, a simplified contouring error model is presented to approximate the actual contouring error. By defining the Frenet-Serret frame attached to the desired trajectory, the task coordinate frame is obtained. The tracking error can be projected to the task coordinate frame where the tangential and normal errors are defined to approximate the contouring error.

Based on the contouring error model, two control schemes on a general class of motion control systems are developed in the task frame: Coordinated posi-tion control and adaptive control. Coordinated position control is a feedback linearization method derived in the task frame to achieve the desired contouring error dynamics. This control scheme requires the exact knowledge of system dy-namics. To enhance the robustness of the controller, adaptive contouring control is proposed in the task frame. This control algorithm is able to handle bounded external disturbances and system parameter variations while maintaining supe-rior contouring tracking performance. Stability analysis was presented together with remarks on selecting controller gains.

Experiments are implemented on an AC motor driven X-Y table to test the control algorithms. To provide the knowledge of systems dynamics, modelling and identification methodology for the experimental platform are presented in detail along with the identification experiments. Experimental results of the proposed control schemes indicate significant improvement of the contouring accuracy when compared with other conventional control algorithms under the same high speed. The adaptive controller exhibits better performance than the coordinated position controller even under certain disturbances.