Parallel robots pose accuracy compensation using artificial neural networks

Author

Yu, Da-yong ; Cong, Da-Cheng ; Han, Jun-wei

Author_Institution

Sch. of Mech. & Electron. Eng., Harbin Inst. of Technol., China

Volume

5

fYear

2005

fDate

18-21 Aug. 2005

Firstpage

3194

Abstract

Parallel robots pose accuracy compensation approach using artificial neural networks has been developed. In this method, an artificial neural network is used with conventional inverse kinematics computation module in parallel. A backpropagation neural network is designed and implemented to learn parallel robot kinematics model error. The trained neural network can be used to performed online pose accuracy compensation in task. Simulation and experimental results for a parallel robot are presented to show the effectiveness of the compensation method based on neural networks.

Keywords

backpropagation; control engineering computing; error compensation; motion compensation; neural nets; position control; robot kinematics; artificial neural network; backpropagation; inverse kinematics computation module; kinematics model error; online pose accuracy compensation; parallel robot pose accuracy compensation; Artificial neural networks; Biological neural networks; Calibration; Computer networks; Concurrent computing; Error correction; Orbital robotics; Parallel robots; Robot kinematics; Robotic assembly; Parallel robot; accuracy compensation; artificial neural networks; kinematics calibration; pose accuracy;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on

Conference_Location

Guangzhou, China

Print_ISBN

0-7803-9091-1

Type

conf

DOI

10.1109/ICMLC.2005.1527493

Filename

1527493