Learning to Exploit Proximal Force Sensing: A Comparison Approach

  • Matteo Fumagalli
  • Arjan Gijsberts
  • Serena Ivaldi
  • Lorenzo Jamone
  • Giorgio Metta
  • Lorenzo Natale
  • Francesco Nori
  • Giulio Sandini

Abstract

We present an evaluation of different techniques for the estimation of forces and torques measured by a single six-axis force/torque sensor placed along the kinematic chain of a humanoid robot arm. In order to retrieve the external forces and detect possible contact situations, the internal forces must be estimated. The prediction performance of an analytically derived dynamic model as well as two supervised machine learning techniques, namely Least Squares Support Vector Machines and Neural Networks, are investigated on this problem. The performance are evaluated on the normalized mean square error (NMSE) and the comparison is made with respect to the dimension of the training set, the information contained in the input space and, finally, using a Euclidean subsampling strategy.

Keywords

Force sensing machine learning humanoid robotics 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Haddadin, S., De Luca, A., Albu-Schaffer, A., Hirzinger, G.: Collision detection and safe reaction with the dlr-iii lightweight manipulator arm. In: IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, pp. 1623–1630 (2006)Google Scholar
  2. 2.
    Cawley, G.C.: Leave-one-out cross-validation based model selection criteria for weighted ls-svms. In: IJCNN 2006: Proceedings of the International Joint Conference on Neural Networks, Vancouver, BC, Canada, July 2006, pp. 1661–1668 (2006)Google Scholar
  3. 3.
    Denavit, J., Hartenberg, R.S.: A kinematic notation for lower-pair mechanisms based on matrices. Journal of Applied Mechanics 23, 215–221 (1955)MathSciNetGoogle Scholar
  4. 4.
    Duda, R.O., Hart, P.E., Stork, D.G.: Pattern Classification. Wiley-Interscience, Hoboken (2001)MATHGoogle Scholar
  5. 5.
    Metta, G., Cannata, G., Maggiali, M., Sandini: An embedded artificial skin for humanoid robots. In: Proc. of IEEE Int. Conf. on Multisensor Fusion and Integration for Intelligent Systems, pp. 434–438 (2008)Google Scholar
  6. 6.
    Dubowsky, S., Liu, G., Iagnemma, K., Morel, G.: A base force/torque sensor approach to robot manipulator inertial parameter estimation. In: Proceedings of the 1998 IEEE International Conference on Robotics and Automation, 1998. ICRA 1998 (1998)Google Scholar
  7. 7.
    Hagan, M.T., Menhaj, M.B.: Training feedforward networks with the marquardt algorithm. IEEE Transactions on Neural Networks 5, 989–993 (1994)CrossRefGoogle Scholar
  8. 8.
    Jamone, L., Nori, F., Metta, G., Sandini, G.: James: A humanoid robot acting over an unstructured world. In: International Conference on Humanoid Robots, Genova, Italy (2006)Google Scholar
  9. 9.
    Stinchombe, M., Hornik, K., White, H.: Multilayer feedforward networks are universal approximators. Neural Networks 2, 359–366 (1989)CrossRefGoogle Scholar
  10. 10.
    Kozlowski, K.: Modelling and Identification in Robotics. Springer, Secaucus (1998)Google Scholar
  11. 11.
    Lagarde, M., Andry, P., Gaussier, P., Boucenna, S., Hafemeister, L.: Proprioception and imitation: on the road to agent individuation. In: Sigaud, O., Peters, J. (eds.) From Motor Learning to Interaction Learning in Robots. SCI, vol. 264, pp. 43–63. Springer, Heidelberg (2010)Google Scholar
  12. 12.
    Levenberg, K.: A method for the solution of certain problems in least squares. Quarterly of Applied Mathematics 2, 164–168 (1944)MATHMathSciNetGoogle Scholar
  13. 13.
    Marquardt, D.: An algorithm for least-squares estimation of nonlinear parameters. SIAM Journal on Applied Mathematics 11, 431–441 (1963)MATHCrossRefMathSciNetGoogle Scholar
  14. 14.
    Ivaldi, S., Fumagalli, M., Jamone, L., Nori, F., Natale, L., Metta, G., Baglietto, M.: Estimation of forces and torques in a humanoid arm: comparison of model based and offline/online learning techniques. Submitted to the 48th IEEE Conference on Decision and Control (2009)Google Scholar
  15. 15.
  16. 16.
  17. 17.
    Nguyen, D., Widrow, B.: Improving the learning speed of 2-layer neural networks by choosing initial values of the adaptive weights. In: Proc. of the Int. Joint Conference on Neural Networks, June 1990, vol. 3, pp. 21–26 (1990)Google Scholar
  18. 18.
    Nguyen-Tuong, D., Seeger, M., Peters, J.: Real-time local gp model learning. In: Sigaud, O., Peters, J. (eds.) From Motor Learning to Interaction Learning in Robots. SCI, vol. 264, pp. 193–207. Springer, Heidelberg (2010)Google Scholar
  19. 19.
    Chung, J.H., Lu, S., Velinsky, S.A.: Human-robot collision detection and identification based on wrist and base force/torque sensors. In: Proceedings of the 2005 IEEE International Conference on Robotics and Automation, 2005. ICRA 2005, April 2005, pp. 3796–3801 (2005)Google Scholar
  20. 20.
    Sciavicco, L., Siciliano, B.: Modeling and control of robot manipulators. MacGraw-Hill, New York (1996)Google Scholar
  21. 21.
    Shinya, M., Kazuhiro, K.: Collision detection system for manipulator based on adaptive impedance control law. In: Proc. IEEE Int. Conf. on Robotics and Automation, pp. 1080–1085 (2003)Google Scholar
  22. 22.
    Smola, A.J., Schölkopf, B.: A tutorial on support vector regression. Statistics and Computing 14(3), 199–222 (2004)CrossRefMathSciNetGoogle Scholar
  23. 23.
    Suykens, J.A.K., Van Gestel, T., De Brabanter, J., De Moor, B., Vandewalle, J.: Least Squares Support Vector Machines. World Scientific Publishing Co. Pte. Ltd., Singapore (2002)MATHGoogle Scholar
  24. 24.
    Swevers, J., Ganseman, C., Tukel, D.B., De Schutter, J., Van Brussel, H.: Optimal robot excitation and identification. IEEE Trans. on Robotics and Automation 3(5), 730–740 (1997)CrossRefGoogle Scholar
  25. 25.
    Ting, J., Mistry, M., Peters, J., Schaal, S., Nakanishi, J.: A bayesian approach to nonlinear parameter identification for rigid body dynamics. In: Robotics: Science and Systems, RSS (2006)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Matteo Fumagalli
    • 1
  • Arjan Gijsberts
    • 1
  • Serena Ivaldi
    • 1
  • Lorenzo Jamone
    • 1
  • Giorgio Metta
    • 1
  • Lorenzo Natale
    • 1
  • Francesco Nori
    • 1
  • Giulio Sandini
    • 1
  1. 1.Robotics, Brain and Cognitive Science DepartmentItalian Institute of TechnologyGenoa

Personalised recommendations