Premium
Repetitive control of synchronized operations for process applications
Author(s) -
Ratcliffe James D.,
Hätönen Jari J.,
Lewin Paul L.,
Rogers Eric,
Owens David H.
Publication year - 2007
Publication title -
international journal of adaptive control and signal processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.73
H-Index - 66
eISSN - 1099-1115
pISSN - 0890-6327
DOI - 10.1002/acs.918
Subject(s) - iterative learning control , control theory (sociology) , repetitive control , pid controller , process (computing) , aliasing , feed forward , control system , trajectory , computer science , controller (irrigation) , robot , control engineering , filter (signal processing) , engineering , control (management) , artificial intelligence , temperature control , agronomy , physics , astronomy , computer vision , electrical engineering , biology , operating system
Abstract Repetitive control (RC) algorithms for a plant, which contain pairs of complex conjugate poles at low frequency, resulting in a resonant system, is the subject area of this paper where the experimental results given are for a gantry robot and conveyor system in which the gantry is required to transfer payloads to a constant velocity conveyor by performing a repeating ‘pick and place’ operation. Initially, the gantry robot is controlled by means of a PID feedback controller in parallel with a proportional (P‐type) repetitive feed‐forward loop, while the conveyor operates under proportional feedback control. It is found that the RC system is unable to achieve long‐term performance. The performance degrades within a relatively small number of repetitions due to the build up of resonant frequencies in the learning loop. To prevent this, a batch aliasing technique, originally developed for iterative learning control, is modified to work in the RC framework, and is implemented in real‐time. The superior performance potential of the aliasing system is demonstrated experimentally. In the second part of this paper, multi‐machine systems, are considered where the critical new factor is the relative error between the conveyor and the robot. Here a second supervisory learning loop is proposed for use to shift the reference trajectory of one machine so that the relative placement error is also reduced. Again, supporting experimental results are given. Copyright © 2006 John Wiley & Sons, Ltd.