Advertisement

Vague Neural Network Based Reinforcement Learning Control System for Inverted Pendulum

  • Yibiao Zhao
  • Siwei Luo
  • Liang Wang
  • Aidong Ma
  • Rui Fang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4234)

Abstract

Reinforcement learning is a class of model-free learning control method that can solve Markov decision problems. But it has some problems in applications, especially in MDPs of continuous state spaces. In this paper, based on the vague neural networks, we propose a Q-learning algorithm which is comprehensively considering the reward and punishment of the environment. Simulation results in cart-pole balancing problem illustrate the effectiveness of the proposed method.

Keywords

Reinforcement Learning Inverted Pendulum Fuzzy Neural Network Continuous State Space Optimal Control Policy 
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.
    Horikawa, S., Furuhashi, T., Uchikawa, Y.: On fuzzy modeling using fuzzy neural network with the back-propagation algorithm. IEEE Trans. NN-3(5), 801–806 (1992)Google Scholar
  2. 2.
    Gau, W.L., Buehrer, D.J.: Vague sets [J]. IEEE Transactions on Systems,Man,and Cybernetics 23(2), 610–614 (1993)CrossRefMATHGoogle Scholar
  3. 3.
    Bustince, H., Burillo, P.: Vague sets are intuitionistic fuzzy sets. Fuzzy Sets and Systems 79(1), 403–405 (1996)CrossRefMATHMathSciNetGoogle Scholar
  4. 4.
    Chen, S.M.: Fuzzy system reliability analysis based on vague set theory. In: 1997 IEEE International Conference on Computational Cybernetics and Simulation, vol. 2, pp. 1650–1655 (1997)Google Scholar
  5. 5.
    Zheng, Y., Luo, S., Lv, Z.: The negative effect on the control of inverted pendulum caused by the limit cycle in reinforcement learning. In: ICNN&B 2005, pp. 772–775 (2005)Google Scholar
  6. 6.
    Onat, A.: Q-learning with recurrent Neural Networks as a Controller for the Inverted Pendulum Problem. In: The Fifth International Conference on Neural Information Processing, October 21-23, pp. 837–840 (1998)Google Scholar
  7. 7.
    Anderson, C.W.: Learning to control an inverted pendulum using neural networks. IEEE Control System Magazine 9(3), 31–37 (1989)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Yibiao Zhao
    • 1
  • Siwei Luo
    • 2
  • Liang Wang
    • 3
  • Aidong Ma
    • 3
  • Rui Fang
    • 2
  1. 1.School of Traffic and TransportationBeijing Jiaotong UniversityChina
  2. 2.School of Computer and Information TechnologyBeijing Jiaotong UniversityChina
  3. 3.School of Electronics and Information EngineeringBeijing Jiaotong UniversityBeijingP.R. China

Personalised recommendations