An influence based error identification for kinematics calibration of serial robotic manipulators

Files

hdl_120248.pdf (484.99 KB)
  (Accepted version)

Date

2019

Authors

Patel, D.
Lu, T.
Chen, L.

Editors

Yang, R.
Takeda, Y.
Zhang, C.
Fang, G.

Advisors

Journal Title

Journal ISSN

Volume Title

Type:

Conference paper

Citation

Mechanisms and Machine Science, 2019 / Yang, R., Takeda, Y., Zhang, C., Fang, G. (ed./s), vol.72, pp.145-154

Statement of Responsibility

D. A. Patel, T. F. Lu and L. Chen

Conference Name

IFToMM International Symposium on Robotics & Mechatronics (ISRM) (29 Nov 2017 - 1 Dec 2017 : Sydney, Australia)

Abstract

In serial robotic manipulators, due to the nature of the coupling of links, the influence of errors in joint parameters on pose accuracy varies with the configuration. Kinematics parameter’s error identification in the standard kinematics calibration has been configuration independent which does not consider the influence of kinematics parameter on robot tool pose accuracy for a given configuration. Mutually dependent joint parameter errors cannot be identified at the same time, and hence error of one parameter in each pair is identified. In a pair of mutually dependent joint parameters, the effect of error in one parameter on positional error can be more than the other one depending on the configuration. Therefore, the error detection may be incorrect if the influence of joint parameters is ignored during the error identification. This research analyses the configuration dependent influences of kinematics parameters error on pose accuracy of a robot. Based on the effect of kinematics parameters, the errors in the kinematics parameters are identified. Kinematics model of the robot is composed of the modified DH method and an improved DH method to avoid the limitations of the original DH method. First, the robot is calibrated to identify errors in 17 kinematics parameters conventionally, and then errors are detected based on the proposed method.

School/Discipline

Dissertation Note

Provenance

Description

Access Status

Rights

© Springer Nature Switzerland AG 2019

License

Grant ID

Call number

Persistent link to this record