Advertisement

Incremental Knowledge Acquisition Using Generalised RDR for Soccer Simulation

  • Angela Finlayson
  • Paul Compton
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6232)

Abstract

This paper describes a system that allows soccer coaches to specify the behaviour of agents for the Robocup 2D soccer simulation domain [1]. The work we present is based on Generalised Ripple Down Rules [7,2]and allows the coach to interact directly with the system to incrementally model behaviours along with intermediate features during the knowledge acquisition process. The system was evaluated over a period of 6 months to measure the level of performance of the multi-agent teams created with the system and to gather feedback about the usability of the system. During this period the system was successfully used by four soccer coaches with differing levels of soccer and computer expertise. All coaches were able to use the system to develop teams that could play at a world class level against the finalists from the Robocup 2007 2D simulation tournament. The approach we present is general enough to be applied to any complex planning problem, with the requirement that a rich feature language is developed to support the specific domain.

Keywords

Knowledge Acquisition High Level Attribute Output List Minimum Description Length Principle Knowledge Acquisition Process 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Noda, I., Matsubara, H., Hiraki, K., Frank, I.: Soccer server: A tool for research on mulitagent systems. Applied Artificial Intelligence 12, 233–250 (1998)CrossRefGoogle Scholar
  2. 2.
    Compton, P., Cao, T., Kerr, J.: Generalising Incremental Knowledge Acquisition. In: Proceedings of the Pacific Knowledge Acquisition Workshop (2004)Google Scholar
  3. 3.
    Stone, P.: Layered Learning in Multi-agent Systems. PhD Thesis, Carnegie Mellon University (1998)Google Scholar
  4. 4.
    de Boer, R., Kok, J.: The Incremental Development of a Synthetic Mutli-Agent System: The UvA Trilean 2001 Robotic Soccer Simulation Team. Master’s thesis, University of Amsterdam (2002)Google Scholar
  5. 5.
    Finlayson, A., Compton, P.: Incremental Knowledge Acquisition using RDR for Soccer Simulation. In: Proceedings of the Pacific Knowledge Acquisition Workshop (2004)Google Scholar
  6. 6.
    Preston, P., Edwards, E., Compton, P.: A 2000 Rule Expert System Without Knowledge Engineers. In: Proceedings of the 8th AAAI-Sponsored Banff Knowledge Acquisition for Knowledge-Based Systems Workshop, Banff, Canada, pp. 17.1–17.10 (1994)Google Scholar
  7. 7.
    Compton, P., Richards, D.: Generalising Ripple-Down Rules. In: Dieng, R., Corby, O. (eds.) EKAW 2000. LNCS (LNAI), vol. 1937, pp. 380–386. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  8. 8.
    Suryanto, H., Compton, P.: Invented Knowledge Predicates to Reduce Knowledge Acquisition Effort. In: Tecuci, G., Aha, D., Boicu, M., Cox, M., Ferguson, G. (eds.) Proceedings of the IJCAI 2003 Workshop on Mixed-Initiative Intelligent Systems, Eighteenth International Joint Conference on Artificial Intelligence, Austin tate, Acapulco Mexico, August 9-15, pp. 107–114 (2003)Google Scholar
  9. 9.
    Kang, B., Compton, P., Preston, P.: Multiple Classification Ripple Down Rules: Evaluation and Possibilities. In: Proceedings of the 9th AAAI-Sponsored Banff Knowledge Acquisition for Knowledge-Based Systems Workshop, Banff, Canada, pp. 17.1–17.20. University of Calgary (1995)Google Scholar
  10. 10.
    Shiraz, G., Sammut, C.: Acquiring Control Knowledge from Examples Using Ripple-down Rules and Machine Learning. In: Gaines, B.R., Musen, M. (eds.) Proceedings of Eleventh Workshop on Knowledge Acquisition, Modelgin and Management (KAW 1998), Banff, Alberta Canada, April 18-23, pp. KAT-5.1–KAT-5.17. University of Calgary, Calgary (1990)Google Scholar
  11. 11.
    Compton, P., Ramadan, Z., Preston, P., Le-Gia, T., Chellen, V., Mullholland, M.: A trade-off between domain knowledge and problem-solving method power. In: Gaines, B., Musen, M. (eds.) 11th Banff Knowledge Acquisition for Knowledge-Based Systems Workshop, SHARE, Banff, vol. 17, pp. 1–19. SRDG Publications/University of Calgary (1999)Google Scholar
  12. 12.
    Richards, D., Compton, P.: Revisiting Sisyphus I - an Incremental Approach to Resource Allocation Using Ripple-Down Rules. In: Gaines, B., Kremer, R., Musen, M. (eds.) 12th Banff Knowledge Acquisition for Knowledge-Based Systems Workshop, Banff, pp. 7-7.1–7.20. SRDG Publications/University of Calgary (1999)Google Scholar
  13. 13.
    Beydoun, G., Hoffman, A.: Incremental Acquisition of Search Knowledge. International Journal of Human Computer Studies 52(3), 493–530 (2000)CrossRefGoogle Scholar
  14. 14.
    Compton, P., Horn, R., Quinlan, R., Lazarus, L.: Maintaining an expert system in J. R. Quinlan. In: Applications of Expert Systems, pp. 366–385. Addison Wesley, London (1989)Google Scholar
  15. 15.
    Kang, B., Yoshida, K., Motoda, H., Compton, P.: A help desk system with Intelligent Interface. Applied Artificial Intelligence 11(7-8), 611–631 (1997)CrossRefGoogle Scholar
  16. 16.
    Obst, O., Bodecker, J.: Flexible Coordination of Multiagent Team Behaviour Using HTN Planning. In: Bredenfeld, A., Jacoff, A., Noda, I., Takahashi, Y. (eds.) RoboCup 2005. LNCS (LNAI), vol. 4020, pp. 521–528. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  17. 17.
    CMU RoboCup Simulator Team Homepage: Carnegie Mellon University, Pittsburgh, PA (2003), http://www-2.cs.cmu.edu/~pstone/RoboCup/CMUnited-sim.html
  18. 18.
    The Official UvA Trilearn Website: University of Amsterdam (2001), http://www.science.uva.nl/~jellekok/robocup
  19. 19.
    Brainstormers base source code, http://www.ni.uos.de
  20. 20.
    Helios base source code, http://sourceforge.jp/projects/rctools
  21. 21.
    Riedmiller, M., Merke, A.: Using Machine Learning Techniques in Complex Multi-Agent Domains. In: Stamatescu, I., Menzel, W., Richter, M., Ratsch, U. (eds.) Perspectives on Adaptivity and Learning. LNCS. Springer, Heidelberg (2002)Google Scholar
  22. 22.
    Kok, J., Spaan, M., Vlassis, N.: Multi-robot decision making using coordination graphs. In: de Almeida, A.T., Nunes, U. (eds.) Proceedings of the 11th International Conference on Advanced Robotics, ICAR 2003, Coimbra, Portugal, June 30-July 3, pp. 1124–1129 (2003)Google Scholar
  23. 23.
    Yunpeng, C., Jiang, C., Jinyi, Y., Li, S.: Global Planning from Local Eyeshot: An Implementation of Observation -based Plan Coordination in RoboCup Simulation Games. In: Birk, A., Coradeschi, S., Tadokoro, S. (eds.) RoboCup 2001. LNCS (LNAI), vol. 2377, p. 12. Springer, Heidelberg (2002)Google Scholar
  24. 24.
    Buttinger, S., Diedrich, M., Hennig, L., Honemann, A., Hugelmeyer, P., Nie, A., Pegam, A., Rogowski, C., Rollinger, C., Steffens, T., Teiken, W.: The ORCA Project Report, http://www.cl-ki.uni-osnabrueck.de/~tsteffen/orcapub.html
  25. 25.
    Scerri, P., Coradeschi, S., Torne, A.: A user oriented system for developing behaviour based agents. In: Asada, M., Kitano, H. (eds.) RoboCup 1998. LNCS (LNAI), vol. 1604, pp. 173–186. Springer, Heidelberg (1999)CrossRefGoogle Scholar
  26. 26.
    Compton, P., Preston, P., Kang, B., Yip, T.: Local patching produces compact knowledge bases. In: Steels, L., Van de Velde, W., Schreiber, G. (eds.) EKAW 1994. LNCS, vol. 867, pp. 103–117. Springer, Heidelberg (1994)Google Scholar
  27. 27.
    Suryanto, H., Richards, D., Compton, P.: The Automatic Compression of Multiple Classification Ripple Down Rule Knowledge Base Systems: Preliminary Experiments. In: Proceedings of the Third International Conference on Knowledge-Based Intelligent Information Engineering Systems, L. Jain Adelaide, pp. 203–206 (1999)Google Scholar
  28. 28.
    Edwards, G., Compton, P., Malor, R., Srinivasan, A., Lazarus, L.: PEIRS: a pathologist maintained expert system for the interpretation of chemical pathology reports. Pathology 25, 27–34 (1993)CrossRefGoogle Scholar
  29. 29.
    Asakawa, H., Ueda, M., Yamazaki, Y., Takeuchi, I.: YowAI 2007 Team Description. In: RoboCup 2007. LNCS (LNAI), vol. 2377. Springer, Heidelberg (2007)Google Scholar
  30. 30.
    Berger, R., Burkhard, H.: AT Humboldt Team Description 2007. In: Robocup 2007. LNCS (LNAI), vol. 2377. Springer, Heidelberg (2007)Google Scholar
  31. 31.
    Norouzitallab, M.: Nemesis 2D - Team Description 2007. In: Robocup 2007. LNCS (LNAI), vol. 2377. Springer, Heidelberg (2007)Google Scholar
  32. 32.
    Akiyama, H.: HELIOS 2007 Team Description. In: RoboCup 2007. LNCS (LNAI), vol. 2377. Springer, Heidelberg (2007)Google Scholar
  33. 33.
    Riedmiller, M., Gabel, T.: Brainstormers 2D Team Description 2007. In: RoboCup 2007. LNCS (LNAI), vol. 2377. Springer, Heidelberg (2007)Google Scholar
  34. 34.
    Kitano, H., Kuniyoshi, M., Noda, Y., Osawa, E.: RoboCup: The Robot World Cup Initiative. In: The First International Conference on Autonomous Agents (1997)Google Scholar
  35. 35.
    Gspandl, S., Monichi, D., Reip, M., Steinbauer, G., Wolfram, M., Zehentnerm, C.: KickOffTug - Team Description Paper 2007. In: Robocup 2007. LNCS (LNAI), vol. 2377. Springer, Heidelberg (2007)Google Scholar
  36. 36.
    Bekmann, J., Hoffman, A.: Improved Knowledge Acquisition for High Performance Heuristic Search. In: Proceedings of the International Joint Conference on Artificial Intelligence, pp. 41–46 (2005)Google Scholar
  37. 37.
    Kerr, J., Compton, P.: Toward Generic Model-Based Object Recognition by Knowledge Acquisition and Machine Learning. In: Workshop on Mixed-Initiative Intelligent Systems, Int. Joint Conf. on AI (2003)Google Scholar
  38. 38.
    Singh, P., Compton, P.: Combining machine learning and heuristic rules using GRDR for detection of honeycombing in HRCT lung images. In: 9th International Conference on Knowledge-Based Intelligent Information and Engineering Systems,Google Scholar
  39. 39.
    Kwok, R.: Using Ripple Down Rules for Actions and Planning. In: Proceedings of the 7th Pacific Rim International Conference on Artificial Intelligence: Trends in Artificial Intelligence (2002)Google Scholar
  40. 40.
    Compton, P., Jansen, B.: A philosophical basis for knowledge acquisition. Knowledge Acquisition 2, 241–257 (1990)CrossRefGoogle Scholar
  41. 41.
    Robocup 2007 Soccer Simulation Competition (2007), http://wiki.cc.gatech.edu/robocup/index.php/Soccer_Simulation
  42. 42.
    Yoshida, T., Wada, T., Motoda, H., Washio, T.: Adaptive Ripple Down Rules method based on minimum description length principle. Intelligent Data Analysis 8(3) (2004)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Angela Finlayson
    • 1
  • Paul Compton
    • 1
  1. 1.Department of Artificial Intelligence, School of Computer Science and EngineeringThe University of New South WalesSydneyAustralia

Personalised recommendations