Learning and Fast Object Recognition in Robot Skill Acquisition: A New Method

  • I. Lopez-Juarez
  • R. Rios-Cabrera
  • M. Peña-Cabrera
  • R. Osorio-Comparan
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6256)


Invariant object recognition aims at recognising an object independently of its position, scale and orientation. This is important in robot skill acquisition during grasping operations especially when working in unstructured environments. In this paper we present an approach to aid the learning of manipulative skills on-line. We introduce and approach based on an ANN for object learning and recognition using a descriptive vector built on recurrent patterns. Experimental learning results using a fast camera are presented. Some simple parts (i.e. circular, squared and radiused-square) were used for comparing different connectionist models (Backpropagation, Perceptron and FuzzyARTMAP) and to select the appropriate model. Later during experiments, complex figures were learned using the chosen FuzzyARTMAP algorithm showing a 93.8% overall efficiency and 100% recognition rate with not so complex parts. Recognition times were lower than 1 ms, which clearly indicates the suitability of the approach to be implemented in robotic real-world operations.


ART Theory Artificial Neural Networks Invariant Object Recognition Machine Vision Robotics 


  1. 1.
    Hoska, D.: Fixturless assembly manufacturing. Manufacturing Eng. 100, 49–54 (1988)Google Scholar
  2. 2.
    Ngyuen, W., Mills, J.K.: Multirobot control for flexible fixturless assembly of flexible sheet metal autobody parts. In: IEEE Int. Conf. Robotics & Automation, pp. 2340–2345 (1996)Google Scholar
  3. 3.
    Langley, C.S., Eleuterio, D.: GMT, A memory efficient neural network for robotic pose estimation. In: Proc. of IEEE Int. Symp.on Computational Intelligence in Robotics and Automation, vol. 1, pp. 418–423. IEEE CIRA (2003)Google Scholar
  4. 4.
    Alberto, S., Aguado, E., Montiel, M.: Invariant characterization of the Hough Transform for pose estimation of arbitrary shapes. Pattern Recognition 35, 1083–1097 (2002)CrossRefzbMATHGoogle Scholar
  5. 5.
    Wu, C.-H., et al.: A new computation of shape moments via quadtree decomposition. Pattern Recognition 34, 1319–1330 (2001)CrossRefzbMATHGoogle Scholar
  6. 6.
    Best, P.J., McKay, N.D.: A Method for Registration of 3-D Shapes. IEEE Trans. on Pattern Analysis and Machine Intelligence 14(2) (1992)Google Scholar
  7. 7.
    Bribiesca, E.: A new Chain Code. Pattern Recognition 32, 235–251 (1999)CrossRefGoogle Scholar
  8. 8.
    Gonzalez-Galvan, E.J., et al.: Application of Precision-Enhancing Measure in 3D Rigid-Body Positioning using Camera-Space Manipulation. The International Journal of Robotics Research 16(2), 240–257 (1997)CrossRefGoogle Scholar
  9. 9.
    Dickmanns, E.: Vehicles capable of dynamic vision: a new breed of technical beings? Artifical Intelligence 103, 49–76 (1998)CrossRefzbMATHGoogle Scholar
  10. 10.
    Kollnig, H., Nagel, H.: 3d pose estimation by directly matching polyhedral models to gray value gradients. Int. Journal of Comp.Vision 23(3), 282–302 (1997)CrossRefGoogle Scholar
  11. 11.
    Bone, G.M., Capson, D.: Vision-guided fixturless assembly of automotive components. Robotics and Computer Integrated Manufacturing 19, 79–87 (2003)CrossRefGoogle Scholar
  12. 12.
    Peña-Cabrera, M., Lopez-Juarez, I., Rios-Cabrera, R., Corona-Castuera, J.: Machine vision approach for robotic assembly. Journal of Assembly Automation 25(3), 204–216 (2005)CrossRefGoogle Scholar
  13. 13.
    Corona-Castuera, J., Lopez-Juarez, I.: Behaviour-based approach for skill acquisition during assembly operations, starting from scratch. Robotica 24(6), 657–671 (2006)CrossRefGoogle Scholar
  14. 14.
    Chen, K.: Efficient parallel algorithms for computation of two-dimensional image moments. Pattern Recognition 23, 109–119 (1990)CrossRefGoogle Scholar
  15. 15.
    Cybenko, G.: Approximation by superposition of a Sigmoidal Function. Mathematics of Control, Signals, and Systems 2, 303–314 (1989)MathSciNetCrossRefzbMATHGoogle Scholar
  16. 16.
    Carpenter Gail, A., et al.: Fuzzy ARTMAP: A Neural Network Architecture for Incremental Supervised Learning of Analog Multidimensional Maps. IEEE Transactions on Neural Networks 3(5) (1992)Google Scholar
  17. 17.
    Carpenter, G.A., Grossberg, S.: Adaptive Resonance Theory. In: The Handbook of Brain theory and neural networks, 2nd edn. MIT Press, Cambridge (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • I. Lopez-Juarez
    • 1
  • R. Rios-Cabrera
    • 1
  • M. Peña-Cabrera
    • 2
  • R. Osorio-Comparan
    • 2
  1. 1.Centro de Investigación y de Estudios Avanzados del I.P.N. (CINVESTAV)Ramos ArizpeMéxico
  2. 2.Instituto de Investigaciones en Matemáticas Aplicadas y en SistemasUniversidad Nacional Autónoma de México (UNAM)México

Personalised recommendations