Time-domain prefilter design for enhanced tracking and vibration suppression in machine motion control. (1st May 2018)

• Record Type: Journal Article
• Authors: Cole, Matthew O.T.; Shinonawanik, Praween; Wongratanaphisan, Theeraphong
• Is Part Of: Mechanical systems and signal processing. Volume 104(2018)
• Page Start: 106
• Page End: 119
• Publication Date: 2018-05-01
• Subjects: Motion control -- Command pre-filtering -- Vibration -- Tracking -- Servo-control -- Industrial robot -- Flexible structure; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2017.10.029
• Languages: English
• ISSNs: 0888-3270

Abstract

Highlights: New FIR prefilter design for motion control applications. Combines finite-time vibration cancellation with dynamic compensation properties. Applied experimentally to high-speed Delta robot with compliant end-effector. Achieves significant reductions in peak and RMS tracking error, as well as settling time, for complex motion patterns. Abstract: Structural flexibility can impact negatively on machine motion control systems by causing unmeasured positioning errors and vibration at locations where accurate motion is important for task execution. To compensate for these effects, command signal prefiltering may be applied. In this paper, a new FIR prefilter design method is described that combines finite-time vibration cancellation with dynamic compensation properties. The time-domain formulation exploits the relation between tracking error and the moment values of the prefilter impulse response function. Optimal design solutions for filters having minimum H2 norm are derived and evaluated. The control approach does not require additional actuation or sensing and can be effective even without complete and accurate models of the machine dynamics. Results from implementation and testing on an experimental high-speed manipulator having a Delta robot architecture with directionally compliant end-effector are presented. The results show the importance of prefilter moment values for tracking performance and confirm that the proposed method can achieve significant reductions in both peak and RMS tracking error, as well as settling time, for complex motion patterns.