Journal of Institute of Control, Robotics and Systems (제어로봇시스템학회논문지)

Institute of Control, Robotics and Systems (제어로봇시스템학회)
권호 표지
DOI QR코드

DOI QR Code

Perturbation/Correlation based Optimization

섭동/상관관계 기반 최적화 기법

  • 이수용 (홍익대학교 기계시스템디자인공학과)
  • Received : 2011.05.20
  • Accepted : 2011.06.20
  • Published : 2011.09.01
Download PDF
⟨ Previous Next ⟩

Abstract

This paper describes a new method of estimating the gradient of a function with perturbation and correlation. We impose a known periodic perturbation to the input variable and observe the output of the function in order to obtain much richer and more reliable information. By taking the correlation between the input perturbation and the resultant function outputs, we can determine the gradient of the function. The computation of the correlation does not require derivatives; therefore the gradient can be estimated reliably. Robust estimation of the gradient using perturbation/correlation, which is very effective when an analytical solution is not available, is described. To verify the effectiveness of perturbation/correlation based estimation, the results of gradient estimation are compared with the analytical solutions of an example function. The effects of amplitude of the perturbation and number of samplings in a period are investigated. A minimization of a function with the gradient estimation method is performed.

Keywords

References

  1. J. S. Lee, "Optimal design and simulation of SCARA robot arm," Journal of Institute of Control, Robotics and Systems (in Korean), vol. 15, no. 6, pp. 612-618, 2009. https://doi.org/10.5302/J.ICROS.2009.15.6.612
  2. C.-H. Chung, S.-H. Lee, K.-S. Kim, Y.-S. Cha, and Y.-S. Ryuh, "Optimization of input parameters by using DOE for dynamic analysis of bio-inspired robotic fish 'Ichthus'," Journal of Institute of Control, Robotics and Systems (in Korean), vol. 16, no. 8, pp. 799-803, 2010. https://doi.org/10.5302/J.ICROS.2010.16.8.799
  3. Y. Choi and H.-J. Yu, "Optimization-based real-time human elbow joint angle extraction method," Journal of Institute of Control, Robotics and Systems (in Korean), vol. 14, no. 12, pp. 1278-1285, 2008. https://doi.org/10.5302/J.ICROS.2008.14.12.1278
  4. P. E. Gill, W. Murray, and M. H. Wright, Practical Optimization, Academic Press, 1981.
  5. T. Goto, T. Inoyama, and K. Takeyasu, "Precise insert operation by tactile controlled robot," Industrial Robot: An International Journal, vol. 1, no. 5, pp. 225-228, 1974. https://doi.org/10.1108/eb004732
  6. S.-H. Li and H. Asada, "Automated robotic assembly using a vibratory work table: Optimal tuning of vibrators based on the Taguchi method," Thesis (Master of Science) - Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 1992.
  7. S. Lee and H. Harry Asada, "A perturbation/correlation method for force guided robot assembly," IEEE Transactions on Robotics and Automation, vol. 15, no. 4, pp. 764-773, 1999. https://doi.org/10.1109/70.782031
  8. B. Chung, P. Knuepfer, and S. Lee, "Obstacle avoidance with perturbation/correlation method," Proc. of the 2002 IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 2425-2430, 2002. https://doi.org/10.1109/IRDS.2002.1041631
  9. S. Lee, "Perturbation/correlation based optimal internal combustion engine tuning," International Journal of Automotive Technology, vol. 9, no. 4, pp. 429-436, 2008. https://doi.org/10.1007/s12239-008-0052-3
  10. J. Bae, G.-W. Kim, and S. Lee, "Application of perturbation/ correlation based gradient estimation for environment exploration," International Journal of Control, Automation, and Systems, vol. 7, no. 2, pp. 233-242, April 2009. https://doi.org/10.1007/s12555-009-0209-5
  11. M. T. Hagan, H. B. Demuth, and M. H. Beale, Neural Network Design, PWS Publishing company, 2002.
  12. R. Battiti, "First and second order methods for learning: between steepest descent and newton's method," Neural Computation, vol. 4, no. 2, pp. 141-166, 1992. https://doi.org/10.1162/neco.1992.4.2.141