Advertisement

RoManSy 6 pp 350-357 | Cite as

Dynamic Command Motion Tuning for Robots A Self Learning Algorithm

  • G. W. Vernon
  • J. Rees Jones
  • G. T. Rooney

Summary

Computer control offers a significant advantage in that the motion command can be readily up-dated. The robot’s trajectory can be stored as a set of discrete data rather than coefficients of an explicit function. A simple algorithm is derived for tuning this data subsequent to each run. Its use requires minimal knowledge of the dynamics and no additional transducers. Once the required tracking accuracy has been achieved, the up-dating process can be curtailed. Should the system be subject to parameter variations, retention of the process should cope with effects due to these variations. Experimental results are included, the peak dynamic tracking errors being reduced from 10% of the motion range to 0.5% in the best case and 4% to 0.6% in the worst.

Keywords

Tracking Accuracy Motion Range Feedback Control System Displacement Error Fundamental Natural Frequency 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. (1).
    Stoddart,D.A.,“Polydyne Cam Design”,Mach. Design,Nos.l–3,121,pp.146,1953.Google Scholar
  2. (2).
    Rees Jones,J., “A. Comparison of the Dynamic Performance of Tuned and Non-tuned multiharmonic cams”, Second IFToMM Int. Synp. on Linkages and Computer Aided Design Methods, Bucharest,Romania,Vol Ill-2,paper 47,pp. 553–564, 1977.Google Scholar
  3. (3).
    Atkeson,C.G., An,C.H. & Hollerbach,J.M., “Estimation of Inertial Parameters of Manipulator Loads and Links”, 3rd International Symposium on Robotics Research, Gouvieux ( Chantilly ), France, October 1985.Google Scholar
  4. (4).
    Craig,J.J., “Adaptive Control of Manipulators Through Repeated Trials”, paper FP2, pp.1566–1573, JACC 1984.Google Scholar
  5. (5).
    Arimoto, S.,Kawamura,S. & Miyazaki,F., “Bettering Operation of Dynamic Systems by Learning: A New Control Theory for Servomechanism or Mechatronics Systems”,TP4,pp.1064–1069,Proc. of the 23rd Conf. on Dec. and Control,IEEE,Las Vegas, NV, December 1984.Google Scholar
  6. (6).
    Arimoto,S., Kawamura,S. & Miyazaki,F., “Can Mechanical Robots Learn by Themselves ?”, Proceedings of end International Symposium of Robotics Research, Kyoto, Japan, August 1985Google Scholar

Copyright information

© Hermes, Paris 1987

Authors and Affiliations

  • G. W. Vernon
    • 1
  • J. Rees Jones
    • 1
  • G. T. Rooney
    • 1
  1. 1.Mechanisms and Machines Group Liverpool PolytechnicLiverpoolEngland

Personalised recommendations