Advertisement

Pathnet: A Neuronal Model for Robotic Motion Planning

  • V. M. Aparanji
  • Uday V. Wali
  • R. Aparna
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 801)

Abstract

This paper proposes a new type of Multi-layered Artificial Neural Network (ANN) suitable for motion control of multi-joint robotic mechanisms with arbitrary Degrees of Freedom (DoF). Input layer classifies the incoming data using Auto Resonance Network (ARN) while higher levels implement Pathnet, a connection oriented neural network with Hebbian reinforcement learning capability. ARN networks grow with training input. Perturbation of ARN nodes allows the network to classify and recognize events with no previous history, Multilayer pathnets can recognize and recall temporal sequences. The network can memorize low cost paths and use parts of such segments in establishing new paths. We have used the system to control a multi segmented robotic system in R. Results of simulation presented in this paper encourage further explorations.

Keywords

Artificial Neural Networks (ANNs) Auto Resonance Network (ARN) Pathnet 

References

  1. 1.
    Saha, S.K.: Introduction to Robotics. McGrawHill, New Delhi (2008). ISBN 978-0070140011Google Scholar
  2. 2.
    Reif, J.H.: Complexity of the mover’s problem and generalizations. In: Department of Computer Science, University of Rochester, Research report - TR58, pp. 421–427 (1979)Google Scholar
  3. 3.
    Jha, P.: Inverse kinematic analysis of robot manipulators. Dissertation, National Institute of Technology, Rourkela (2015)Google Scholar
  4. 4.
    Tai, L., Ming, L.: Deep-learning in mobile robotics - from perception to control systems: a survey on why and why not. Cornell University Archives. arXiv:1612.07139v3 (2017)
  5. 5.
    Veslin, E.Y., Dutra, M.S., Lengerke, O., Carreno, E.A., Tavera, M.J.M.: A hybrid solution for the inverse kinematic on a seven DOF robotic manipulator. IEEE Lat. Am. Trans. 12(2), 212–218 (2014)CrossRefGoogle Scholar
  6. 6.
    Zhou, L., Cook, G.: Path planning for robotic manipulators with redundant degrees of freedom. IEEE Trans. Ind. Electron. 38(6), 413–420 (1991)CrossRefGoogle Scholar
  7. 7.
    Kawasaki, H., Bito, T., Kanzaki, K.: An efficient algorithm for the model based adaptive control of robotic manipulators. IEEE Trans. Robot. Autom. 12(3), 496–501 (1996)CrossRefGoogle Scholar
  8. 8.
    Slotine, J.J.E., Li, W.: On adaptive control of robot manipulators. Int. J. Robot. Res. 6(3), 50–59 (1987)CrossRefGoogle Scholar
  9. 9.
    Duka, A.: Neural network based inverse kinematics solution for trajectory tracking of a robotic arm. Procedia Technol. 12, 20–27 (2014)CrossRefGoogle Scholar
  10. 10.
    Faigl, J.: An application of self-organizing map for multirobot multigoal path planning with minmax objective. Comput. Intell. Neurosci. 2016, Article ID 2720630 (2016).  https://doi.org/10.1155/2016/2720630
  11. 11.
    Martin, A.E., Gregg, R.D.: Incorporating human-like walking variability in HZD-based bipedal model. IEEE Trans. Robot. 32(4), 943–949 (2016)CrossRefGoogle Scholar
  12. 12.
    Potkonjak, V., Svetozarevic, B., Jovanovic, K., Holland, O.: The puller-follower control of compliant and noncompliant antagonistic tendon drives in robotic systems. Int. J. Adv. Robot. Syst. 8(5), 69 (2011)CrossRefGoogle Scholar
  13. 13.
    He, H., McGinnity, T.M., Coleman, S., Gardiner, B.: Linguistic decision making for robot route learning. IEEE Trans. Neural Netw. Learn. Syst. 25(1), 203–215 (2014)CrossRefGoogle Scholar
  14. 14.
    Zhang, J., Springenberg, J.T., Boedecker, J., Burgard, W.: Deep reinforcement learning with successor features for navigation across similar environments. arXiv:1612.05533 (2017)
  15. 15.
    Zhang, Q., Li, M., Wang, X., Zhang, Y.: Reinforcement learning in robot path optimization. J. Softw. 7(3), 657–662 (2012)Google Scholar
  16. 16.
    Carrio, A., Sampedro, C., Rodriguez-Ramos, A., Campoy, P.: A review of deep learning methods and applications for unmanned aerial vehicles. J. Sens. 2017(2), 1–13 (2017).  https://doi.org/10.1155/2017/3296874. Article ID 3296874CrossRefGoogle Scholar
  17. 17.
    Aparanji, V.M., Wali, U.V., Aparna, R.: Robotic motion control using machine learning techniques. In: International Conference on 6th IEEE Communication and Signal Processing (ICCSP) Melmaravattur, April 2017. IEEE Xplore (2017, in press)Google Scholar
  18. 18.
    Aparanji, V.M., Wali. U.V., Aparna, R.: A novel neural network structure for motion control in joints. In: ICEECCOT, Mysore, pp 227–232 (2016). Also available from IEEE XploreGoogle Scholar
  19. 19.
    Aparanji, V.M., Wali, U.V., Aparna, R.: Automated path search and optimization of robotic motion using hybrid ART-SOM neural networks. In: International Conference on Recent Advancement in Computer and Communication, Bhopal, May 2017, ICRAC-2017. LNNS. Springer, Heidelberg (2017, in press)Google Scholar
  20. 20.
    Stufflebeam, R.: Neurons, Synapses, Action Potentials, and Neurotransmission, Consortium on Congnitive Science Instruction (2008). http://www.mind.ilstu.edu/curriculum/neurons_intro/neurons_intro.php
  21. 21.
    Verbny, Y., Zhang, C.L., Chiu, S.Y.: Coupling of calcium homeostasis to axonal sodium in axons of mouse optic nerve. J. Neurophysiol. 88(2), 802–816 (2002)CrossRefGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2018

Authors and Affiliations

  1. 1.Siddaganga Institute of TechnologyTumakuruIndia
  2. 2.KLE Dr. MSS CETBelagaviIndia

Personalised recommendations