Implementation of a robust dynamic control for SCARA robot

Citation
Jm. Lee et al., Implementation of a robust dynamic control for SCARA robot, KSME INT J, 12(6), 1998, pp. 1104-1115
Citations number
18
Categorie Soggetti
Mechanical Engineering
Journal title
KSME INTERNATIONAL JOURNAL
ISSN journal
12264865 → ACNP
Volume
12
Issue
6
Year of publication
1998
Pages
1104 - 1115
Database
ISI
SICI code
1226-4865(199812)12:6<1104:IOARDC>2.0.ZU;2-X
Abstract
A control system for SCARA robot is designed for implementing a robust dyna mic control algorithm. This study focuses on the use of DSPs in the design of joint controllers and interfaces in between the host controller and four joint controllers and in between the joint controllers and four servo driv es. The mechanical body of SCARA robot and the servo drives are selected fr om the commercially available products. The four joint controllers, assigne d to each joint separately, are combined into a common system through the m other board hardwarewise and through the global memory softwarewise. The mo ther board is designed to connect joint controllers onto the board through the slots adopting PC/104 bus structures. The global memory stores the comm on data which can be shared by joint controllers and used by the host compu ter directly, and it virtually combines the whole system into one. To demon strate the performance and efficiency of the system, a robust inverse dynam ic algorithm is proposed and implemented for a faster and more precise cont rol. The robust inverse dynamic algorithm is basically derived from an inve rse dynamic algorithm and a PID compensator. Based upon the derived dynamic equations of SCARA robot, the inverse dynamic algorithm is initially imple mented with 1 msec of control cycle-0.3 msec is actually used for the contr ol algorithm-in this system. The algorithm is found to be inadequate for th e high speed and precision tasks due to inherent modelling errors and time- varying factors. Therefore, a variable PID algorithm is combined with the i nverse dynamic algorithm to reinforce robustness of control. Experimental d ata using the proposed algorithm are presented and compared with the result s obtained from the PID and the inverse dynamic algorithms.