Advertisement

Effects of Using Different Neural Network Structures and Cost Functions in Locomotion Control

  • Jih-Gau Juang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4221)

Abstract

Effects of using different neural network structures and cost functions in locomotion control are investigated. Simulations focus on refinement and a thorough understanding of an artificial intelligent learning scheme. This scheme uses a neural network controller with backpropagation through time learning rule. Through learning, the controller can generate locomotion trajectory along a pre-defined path. Different issues regarding the scheme have been examined. They include the effects of using different numbers of hidden units, the effects of using only angle parameters in the cost function, and the effects of including an energy criterion in the cost function.

Keywords

Cost Function Energy Criterion Hide Unit Reference Trajectory Angle Error 
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.
    Cybenko, G.: Approximation by Superpositions of a Sigmoidal Function. Mathematics of Control, Signals, and Systems 2, 303–314 (1987)CrossRefMathSciNetGoogle Scholar
  2. 2.
    Girosi, F., Poggio, T.: Networks and The Best Approximation Property. Biological Cybernetics 63, 169–179 (1990)MATHCrossRefMathSciNetGoogle Scholar
  3. 3.
    Park, J., Sanberg, I.W.: Universal Approximation Using Radial Basis Function Networks. Neur. Comp. 3, 246–257 (1991)CrossRefGoogle Scholar
  4. 4.
    Wang, Z., Tham, M.T., Morris, A.J.: Multilayer Neural Networks: Approximated Canonical Decomposition of Nonlinearity. International Journal of Control 56, 655–672 (1992)MATHCrossRefMathSciNetGoogle Scholar
  5. 5.
    Zhao, M., Liu, L., Wang, J., Chen, K.: Gaits Compensation Algorithm for Humanoid Robot Walking Control. In: Proc. of CLAWAR, pp. 171–176 (2002)Google Scholar
  6. 6.
    Fujimoto, Y., Kawamura, A.: Simulation of an Autonomous Biped Walking Robot Including Environmental Force Interaction. IEEE Robotics and Automation Magazine 5, 33–42 (1998)CrossRefGoogle Scholar
  7. 7.
    Collins, S.H., Wisse, M., Ruina, A.: A Three-Dimensional Passive-Dynamic Walking Robot with Two Legs and Knees. International Journal of Robotics Research 20, 607–615 (2001)CrossRefGoogle Scholar
  8. 8.
    Miller, W.T., Kun, A.L.: Dynamic Balance of a Biped Walking Robot. In: Omidvar, O., Smagt, P. (eds.) Neural Systems for Robotics, pp. 17–35. Academic Press, London (1997)Google Scholar
  9. 9.
    Park, S.H., Kim, D.S., Lee, Y.J.: Discontinuous Spinning Gait of a Quadruped Walking Robot with Waist-joint. In: Proc. of IEEE International Conference on Intelligent Robots and Systems, pp. 2744–2749 (2005)Google Scholar
  10. 10.
    Noh, K.K., Kim, J.G., Huh, U.Y.: Stability Experiment of a Biped Walking Robot with Inverted Pendulum. 3, 2475–2479 (2004)Google Scholar
  11. 11.
    Igarashi, H., Kakikura, M.: Path and Posture Planning for Walking Robots by Artificial Potential Field Method. In: Proceedings of IEEE International Conference on Robotics and Automation, vol. 3, pp. 2165–2170 (2004)Google Scholar
  12. 12.
    Juang, J.G.: Intelligent Trajectory Control Using Recurrent Averaging Learning. Applied Artificial Intelligence 15, 277–297 (2001)CrossRefGoogle Scholar
  13. 13.
    Juang, J.G.: Intelligent Locomotion Control on Sloping Surfaces. Information Sciences 147, 229–243 (2002)MATHCrossRefMathSciNetGoogle Scholar
  14. 14.
    Haykin, S.: Neural Networks: A Comprehensive Foundation, 2nd edn. Prentice-Hall, New Jersey (1999)MATHGoogle Scholar
  15. 15.
    Rajapakse, J.C., Wang, L.P. (eds.): Neural Information Processing: Research and Development. Springer, Berlin (2004)MATHGoogle Scholar
  16. 16.
    Werbos, P.: Backpropagation Through Time: What It Does and How to Do It. Proceedings of the IEEE 78, 1550–1560 (1990)CrossRefGoogle Scholar
  17. 17.
    Furusho, J., Masubuchi, M.: Control of a Dynamical Biped Locomotion System for Steady Walking. Journal of Dynamic Systems, Measurement, and Control 108, 111–118 (1986)MATHCrossRefGoogle Scholar
  18. 18.
    Kurematsu, Y., Kitamura, S., Kondo, Y.: Trajectory Planning and Control of a Biped Locomotive Robot -Simulation and Experiment. In: Jamshidi, M. (ed.) Robotics and Manufacturing. Recent Trends in Research, Education and Applications, pp. 65–72. ASME Press (1988)Google Scholar
  19. 19.
    Zheng, Y.F., Hemami, H.: Mathematical Modeling of a Robot Collision with its Environment. Journal of Robotic Systems 2, 289–307 (1985)CrossRefGoogle Scholar
  20. 20.
    Raibert, M., Tzafestas, S., Tzafestas, C.: Comparative Simulation Study of Three Control Techniques Applied to a Biped Robot. In: Proc. IEEE Int. Conf. on Systems, Man and Cybernetics, vol. 1, pp. 494–502 (1993)Google Scholar
  21. 21.
    Pars, L.A.: A Treatise on Analytical Dynamics, ch. 14. Heinemann Educational Books LTD, London (1965)Google Scholar
  22. 22.
    Hornik, K.: Some New Results on Neural Network Approximation. Neural Networks 6, 1069–1072 (1993)CrossRefGoogle Scholar
  23. 23.
    Funahashi, K.: On the Approximate Realization of Continuous Mapping by Neural Networks. Neural Networks 2, 183–192 (1989)CrossRefGoogle Scholar
  24. 24.
    Hornik, K., Stinchcombe, M., White, H.: Universal Approximation of an Unknown Mapping and Its Derivatives Using Multilayer Feedforward Networks. Neural Networks 3, 551–560 (1990)CrossRefGoogle Scholar
  25. 25.
    Hornik, K., Stinchcombe, M., White, H.: Multilayer Feedforward Networks are Universal Approximators. Neural Networks 2, 359–366 (1989)CrossRefGoogle Scholar
  26. 26.
    Hornik, K.: Approximation Capabilities of Multilayer Perceptrons. Neural Networks 4, 251–257 (1991)CrossRefGoogle Scholar
  27. 27.
    Kreinovich, V.: Abitrary Nonlinearity is Sufficient to Represent all Functions by Neural Networks: A Theorem. Neural Networks 4, 381–384 (1991)CrossRefGoogle Scholar
  28. 28.
    Ito, Y.: Approximation of Function on a Compact Set by Finite Sums of a Sigmoid Function without Scaling. Neural Networks 4, 385–394 (1991)CrossRefGoogle Scholar
  29. 29.
    Ito, Y.: Approximation of Continuous Functions on Rd by Linear Combinations of Shifted Rotations of a Sigmoid Function with and without Scaling. Neural Networks 5, 105–115 (1992)CrossRefGoogle Scholar
  30. 30.
    Murata, N., Yoshizawa, S., Amari, S.I.: Network Information Criterion-Determining the Number of Hidden Units for an Artificial Neural Network Model. IEEE Trans. Neural Networks. 5, 865–872 (1994)CrossRefGoogle Scholar
  31. 31.
    Kung, S.Y., Hwang, J.N.: Algebraic Projection Analysis for Optimal Hidden Units Size and Learning Rates in Back-Propagation Learning. In: Proc. IEEE Int. Conf. Neural Networks, vol. 1, pp. 363–370 (1988)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Jih-Gau Juang
    • 1
  1. 1.Department of Communications and Guidance EngineeringNational Taiwan Ocean UniversityKeelungTaiwan

Personalised recommendations