Neural Networks for Improved Target Differentiation with Sonar

  • Naima Ait Oufroukh
  • Etienne Colle
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3138)


This study investigates the processing of sonar signals with neural networks for robust recognition of indoor robot environment composed of simple objects (plane, corner, edge and cylinder). The neural networks can differentiate more targets with higher accuracy. It achieves this by exploiting the identifying features extracted from sonar signals. In this paper we compare two different architectures of neural networks (global and specialized structure) in term of classification rates, the best classifier obtained is used to recognize a robot environment. The results strengthen our claims that sonar can be used as a viable system for object recognition in robotics and other application domains.


Mobile Robot Layer Neuron Ultrasonic Sensor Hide Layer Neuron Quadratic Discriminant Analysis 
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.


  1. 1.
    Borenstein, J., Everett, H., Feng, L.: Navigating mobile robots. A.K. Peters, Wellesley (1996)zbMATHGoogle Scholar
  2. 2.
    Leonard, J.J., Durrant-Whyte, H.F.: Directed sonar sensing for mobile robot navigation. The Kluwer international series in engineering and computer science. Kluwer Academic Publishers, Boston (1992)zbMATHGoogle Scholar
  3. 3.
    Barshan, B., Kuc, R.: Differentiating sonar reflections from corners and planes by employing an intelligent sensor. IEEE Trans. Pattern Anal (1990)Google Scholar
  4. 4.
    Barshan, B., Ayrulu, B., Utete, S.W.: Neural Network-Based Target Differentiation Using Sonar for Robotics Applications. IEEE Trans, Robotics and Automation, 435-442 (2000)Google Scholar
  5. 5.
    Bozma, Ö., Kuc, R.: Building a sonar map in a specular environment using a single mobile sensor. IEEE Trans. Pattern Anal and Machine Intell. (1991)Google Scholar
  6. 6.
    Kuc, R.: Three-dimensional tracking using qualitative bionic sonar. Robotics Autonomous Systems, 213–219 (1993)Google Scholar
  7. 7.
    Borenstein, J., Koren, Y.: Obstacle avoidance with ultrasonic sensors. IEEE Journal of Robotics and Automation 4(2), 213–218 (1988)CrossRefGoogle Scholar
  8. 8.
    Roiblat, H.L.: Au. W. W. L., Nachtigall. P. E., Shizumuru. R, and Moons. G., Sonar recognition of targets embedded in sediment, Neural Network, G. B, 1263-1273 (1995)Google Scholar
  9. 9.
    Carpenter, et al.: Fuzzy ARTMAP: A neural network architecture for incremental supervised learning of analog multidimensional maps. IEEE Transactions on Neural Networks 3(5), 698–712 (1992)CrossRefGoogle Scholar
  10. 10.
    Willson: Radar classification using a neural network. In: Proc. SPIE, optical Engineering and photonics in aerospace sensing, pp. 200–210 (1990)Google Scholar
  11. 11.
    Peremans, H., Audenaert, K., Van Campenhout, J.M.: A high resolution sensor based en tri-aural perception. IEEE Trans on Robotic and Automation, CA (1993)Google Scholar
  12. 12.
    Jain, K., Duin, P.W., Mao, J.: Statistical pattern recognition: A review. IEEE Trans. Pattern Anal and Machine Intell. (2000)Google Scholar
  13. 13.
    Ait Oufroukh, N., Colle, E.: Pattern Recognition with Ultrasonic Sensor using Classification Methods. In: 1st Hybrid intelligent system (HIS), Chili, pp. 673–680 (2002)Google Scholar
  14. 14.
    Bishop, C.: Neural Networks for Pattern Recognition. Clarendon Press, Oxford (1995)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Naima Ait Oufroukh
    • 1
  • Etienne Colle
    • 1
  1. 1.Laboratoire Systèmes ComplexesEvry CedexFrance

Personalised recommendations