CMAC Structure Optimization Based on Modified Q-Learning Approach and Its Applications

Part of the Studies in Computational Intelligence book series (SCI, volume 465)

Abstract

Comparing with other neural network based models, CMAC has been applied successfully in many nonlinear control systems because of its computational speed and learning ability. However, for high-dimensional input CMAC in real world applications such as robot, the useable memory is finite or pre-allocated, thus we often have to make our choice between learning accuracy and memory size. This paper discusses how both the number of layer and step quantization influence the approximation quality of CMAC. By experimental enquiry, it is shown that it is possible to decrease the memory size without losing the approximation quality by selecting the adaptive structural parameters. Based on modified Q-learning approach, the CMAC structural parameters can be optimized automatically without increasing the complexity of its structure. The choice of this optimized CMAC structure can achieve a tradeoff between the learning accuracy and finite memory size. At last, this Q-learning based CMAC structure optimization approach is applied on the walk pattern generating for biped robot and workpiece orientation estimation for robot arm assembly respectively.

Keywords

CMAC neural network Structural parameters Q-learning Structure optimization 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Albus, J.S.: Data storage in the cerebellar model articulation controller (CMAC). Transactions of the ASME: Journal of Dynamic Systems, Measurement, and Control, 228–233 (1975)Google Scholar
  2. 2.
    Lu, H.-C., Yeh, M.-F., Chang, J.-C.: CMAC Study with adaptive Quantization. In: IEEE Int. Conf. on Systems, Man, and Cybernetics, Taipei, Taiwan, pp. 2596–2601 (2006)Google Scholar
  3. 3.
    Teddy, S.D., Lai, E.M.-K., Quek, C.: Hierarchically clustered adaptive quantization CMAC and its learning convergence. IEEE Trans. on Neural Networks 18(6), 1658–1682 (2007)CrossRefGoogle Scholar
  4. 4.
    Menozzi, A., Chow, M.: On the training of a multi-resolution CMAC neural network. In: Proc. Int. Conf. Ind. Electron. Control Instrum., vol. 3, pp. 1130–1135 (1997)Google Scholar
  5. 5.
    Lin, C.-M., Chen, T.-Y.: Self-organizing CMAC control for a class of MIMO uncertain nonlinear systems. IEEE Trans. on Neural Networks 20(9), 1377–1384 (2009)CrossRefGoogle Scholar
  6. 6.
    Nguyen, M.N., Shi, D., Quek, C.: Self-organizing Gaussian fuzzy CMAC with truth value restriction. In: Proc. IEEE ICITA, Sydney, Australia, pp. 185–190 (2005)Google Scholar
  7. 7.
    Shi, D., Nguyen, M.N., Zhou, S., Yin, G.: Fuzzy CMAC with incremental Bayesian Ying-Yang learning and dynamic rule construction. IEEE Trans. on Systems, Man and Cybernetics 40(2), 548–552 (2010)CrossRefGoogle Scholar
  8. 8.
    Watkins, C., Dayan, P.: Q-learning. Machine Learning, 279–292 (1992)Google Scholar
  9. 9.
    Yu, W., Sabourin, C., Madani, K., Yan, J.: Design of footstep planning controller for humanoid robot in dynamic environment. In: IEEE Int. Symp. on Knowledge Acquisition and Modeling, China, Wuhan (2008)Google Scholar
  10. 10.
    Carusone, J., D’Eleuterio, G.M.T.: The “FeatureCMAC”: A Neural-Network-Based Vision System for Robotic Control. In: Proc.of the 1998 IEEE Int. Conf. on Robotics & Automation, Leuven, Belgium (1998)Google Scholar
  11. 11.
    Langley, C.S., D’Eleuterio, G.M.T.: Pose Estimation for Fixtureless Assembly Using a Feature CMAC Neural Network. In: 31st Int. Symp. on Robotics, Montreal, Canada (2000)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  1. 1.School of Mechatronic EngineeringNorthwestern Polytechnical UniversityXi’anP.R. China
  2. 2.Signals, Images, and Intelligent Systems Laboratory (LISSI / EA 3956)Paris Est University, Senart Institute of TechnologyLieusaintFrance

Personalised recommendations