Towards acquiring spatio-temporal knowledge from sensor data

  • Kazuo Hiraki
  • Yuichiro Anzai
Technical Papers Section VI
Part of the Lecture Notes in Computer Science book series (LNCS, volume 639)


This paper presents an architecture for acquiring spatio-temporal knowledge. This architecture uses two different algorithms — generalization to interval (GTI) method and feature construction method — for learning from sensory/perceptual information. These methods generalize over positive/negative examples of target knowledge, and output a constraint program that can be used declaratively as a learned concept about spatio-temporal patterns, and procedurally as a method for reasoning about spatio-temporal relations. Thus our methods transform numeric spatio-temporal patterns to symbolic declarative/procedural representations. We have implemented these two algorithms with acorn, a system that acquires spatio-temporal knowledge by observing examples. In this paper, we give two examples from different domains -layout problems and robot-commands learning — to demonstrate the ability of the system and the flexibility of constraint programs for knowledge representation.


Spatial Relation Constraint Program Symbolic Representation Layout Problem Target Concept 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Aha, D.W., Kibler, D. and Albert, M.K.: Instance-based learning algorithms, Machine Learning, 6, 37–66, 1991.Google Scholar
  2. Borning, A.: The programming language aspects of ThingLab: A constraint-oriented simulation laboratory. ACM Transactions on Programming Languages and Systems, 3, 353–387, 1981.Google Scholar
  3. Breiman, L., Friedman, J. H., Olshen, R. A. and Stone, C. J.: Classification and Regression Trees, Wadsworth International Group, 1984.Google Scholar
  4. Connell, J. H. and Brady, M.: Generating and generalizing models of visual objects, Artificial Intelligence, 31, 159–183, 1987.Google Scholar
  5. Dincbas, M., Van Hentenryck, P., Simmonis, H., Aggoum, A., Graf, T. and Berthier, F.: The constraint logic programming language CHIP, In Proc. of the International Conference on Fifth Generation Computer Systems, 693–702, 1988.Google Scholar
  6. Fikes, R. E. and Nilson, N. J.: STRIPS: A new approach to the application of theorem proving to problem solving, Artificial Intelligence, 2, 1971.Google Scholar
  7. Haar, R. L.: Sketching: Estimating object position from relational descriptions, Computer Graphics and Image Processing, 19, 227–247, 1982.Google Scholar
  8. Hiraki, K., Gennari, J., Yamamoto, Y., and Anzai, Y.: Learning Spatial Relations bom Images, In Proc. of Eighth International Machine Learning Workshop ML'91, 407–411, 1991.Google Scholar
  9. Hiraki, K., Gennari, J., Yamamoto, Y., and Anzai, Y.: Encoding Images into Constraint Expressions, In Proc. of Thirteenth Annual Conference of Cognitive Science Society, 31–36, 1991Google Scholar
  10. Jaffar, J. and Lassez, J.: From unification to constraints, In Furukawa, K., Tanaka, H. and Fujisaki, T. (eds.), Logic Programming '87, Lecture Notes in Computer Science, Springer, 1987.Google Scholar
  11. Lassez, C.: Constraint logic programming and option trading, IBM Technical Report, 1987.Google Scholar
  12. Leler, W.: Constraint Programming Languages: Their Specification and Generation, Addison-Wesley, 1988.Google Scholar
  13. Matheus, C.J.; The Need for Constructive Induction, In Proc. of Eighth International Machine Learning Workshop ML'91, 173–177, 1991.Google Scholar
  14. Michalski, R.: A theory and methodology of inductive learning, In Michalski, R., Carbonell, J. and Mitchell, T. (eds.), Machine Leaning: An Artificial Intelligence Approach. Los Altos, CA: Morgan Kaufmann, 1983.Google Scholar
  15. Newell, A. and Simon, H. A.: GPS, A program that simulates human thought, In Feigenbaum, E.A and Feldman, J. (eds.), Computers and Thought, McGraw-Hill, New York, 1963.Google Scholar
  16. Nishizawa, T., Hiraki, K. and Anzai, Y.: EPOCH: A constraint logic programming language with goal constraint description, Journal of Japanese Society for Artificial Intelligence, 7, No. 3, 105–113, May, 1989.Google Scholar
  17. Quinlan, J. R.: Simplifing decision trees, Int. Journal Man-Machine Studies, 27, 221–234, 1987.Google Scholar
  18. Schilimmer, J.C.: Incremental adjustment of representations in learning, In Proc. of the 4th International Workshop on Machine Learning, 79–90,1987.Google Scholar
  19. Sussman, G. J. and Steel, G. L.: CONSTRAINTS — A language for expressing almost-hierarchical descriptions, Artificial Intelligence, 14, 1–39, 1980.Google Scholar
  20. Winston, P.H.: Learning structural descriptions from examples, In Winston, P.H. (ed.), The Psychology of Computer Vision, McGraw-Hill, 1975.Google Scholar
  21. Yamada, A., Amitani, K., Hoshino, T., Nishida, T., Doshita, S.,: The analysis of the spatial description in natural language and the reconstruction of the scene, Journal of Information Processing Society of Japan, 31, No.5, 660–672, 1990.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1992

Authors and Affiliations

  • Kazuo Hiraki
    • 1
  • Yuichiro Anzai
    • 1
  1. 1.Department of Computer ScienceKeio UniversityYokohamaJapan

Personalised recommendations