Towards Learning Inverse Kinematics with a Neural Network Based Tracking Controller

  • Tim Waegeman
  • Benjamin Schrauwen
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7064)


Learning an inverse kinematic model of a robot is a well studied subject. However, achieving this without information about the geometric characteristics of the robot is less investigated. In this work, a novel control approach is presented based on a recurrent neural network. Without any prior knowledge about the robot, this control strategy learns to control the iCub’s robot arm online by solving the inverse kinematic problem in its control region. Because of its exploration strategy the robot starts to learn by generating and observing random motor behavior. The modulation and generalization capabilities of this approach are investigated as well.


Adaptive control Feedback control Inverse kinematics Neural network (NN) Reservoir computing (RC) 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Metta, G., Sandini, G., Vernon, D., Natale, L., Nori, F.: The iCub humanoid robot: an open platform for research in embodied cognition. In: Proc. of the 8th Workshop on Performance Metrics for Intelligent Systems, pp. 50–56. ACM (2008)Google Scholar
  2. 2.
    Tolani, D., Goswami, A., Badler, N.: Real-time inverse kinematics techniques for anthropomorphic limbs. In: Graphical Models, vol. 62, pp. 353–388. Elsevier (2000)Google Scholar
  3. 3.
    Grochow, K., Martin, S., Hertzmann, A., Popović, Z.: Style-based inverse kinematics. Transactions on Graphics (TOG) 23, 522–531 (2004)CrossRefGoogle Scholar
  4. 4.
    D’Souza, A., Vijayakumar, S., Schaal, S.: Learning inverse kinematics. In: Intelligent Robots and Systems, vol. 1, pp. 298–303. IEEE (2001)Google Scholar
  5. 5.
    Peters, J., Schaal, S.: Learning operational space control. In: Proceedings of Robotics: Science and Systems (RSS). MIT Press, Philadelphia (2006)Google Scholar
  6. 6.
    Guez, A., Ahmad, Z.: Solution to the inverse kinematics problem in robotics by neural networks. In: International Joint Conference on Neural Networks, pp. 617–624. IEEE (1988)Google Scholar
  7. 7.
    Reinhart, R., Steil, J.: Reaching movement generation with a recurrent neural network based on learning inverse kinematics for the humanoid robot iCub. In: 9th IEEE-RAS International Conference on Humanoids, pp. 323–330. IEEE (2010)Google Scholar
  8. 8.
    Neumann, K., Rolf, M., Steil, J., Gienger, M.: Learning Inverse Kinematics for Pose-Constraint Bi-Manual Movements. From Animals to Animats 11, 478–488 (2010)CrossRefGoogle Scholar
  9. 9.
    Jaeger, H., Lukosevicius, M., Popovici, D., Siewert, U.: Optimization and applications of echo state networks with leaky-integrator neurons. Neural Networks 20, 335–352 (2007)CrossRefzbMATHGoogle Scholar
  10. 10.
    Schrauwen, B., Verstraeten, D., Van Campenhout, J.: An overview of reservoir computing: theory, applications and implementations. In: Proc. of the 15th European Symposium on Artificial Neural Networks, D-side, pp. 471–482 (2007)Google Scholar
  11. 11.
    Sussillo, D., Abbott, L.: Generating coherent patterns of activity from chaotic neural networks. Neuron 63, 544–557 (2009)CrossRefGoogle Scholar
  12. 12.
    Jaeger, H.: A method for supervised teaching of a recurrent artificial neural network. Patent Application, WO 2002/031764 A2 (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Tim Waegeman
    • 1
  • Benjamin Schrauwen
    • 1
  1. 1.Department of Electronics and Information SystemsGhent UniversityGhentBelgium

Personalised recommendations