CMAC toward intelligent robot: motion and calibration

Author

Hwang, H. ; Choi, D.

Author_Institution

Dept. of Agric. Eng., Sungkyunkwan Univ., Seoul, South Korea

fYear

1990

fDate

3-6 Jul 1990

Firstpage

517

Abstract

The cerebellar model arithmetic controller (CMAC), a kind of neuro-net based adaptive control function generator, has been successfully applied to the problems of inverse kinematic motion of two DOF manipulator and direct inverse calibration of three and six DOF PUMA 560 robots. Since CMAC-net autonomously maps and generalizes a desired system function via learning, CMAC allows no knowledge in modeling a system and other error sources. The mapping structure, learning aspects with algorithms, and the convergence of CMAC net are also briefly stated

Keywords

adaptive control; artificial intelligence; calibration; learning systems; neural nets; position control; robots; PUMA 560; adaptive control; artificial intelligence; calibration; cerebellar model arithmetic controller; convergence; error sources; function generator; intelligent robot; inverse kinematic motion; learning; neural nets; structure mapping; Biological system modeling; Biological systems; Biology computing; Brain modeling; Calibration; Convergence; Intelligent robots; Kinematics; Neural networks; Organisms;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Robots and Systems '90. 'Towards a New Frontier of Applications', Proceedings. IROS '90. IEEE International Workshop on

Conference_Location

Ibaraki

Type

conf

DOI

10.1109/IROS.1990.262433

Filename

262433