Skip to main content
SpringerLink
Log in

Embedding Connectionist Autonomous Agents in Time: The ‘Road Sign Problem’

  • Published:
Neural Processing Letters Aims and scope Submit manuscript

Abstract

This paper identifies a problem of significance for approaches to adaptive autonomous agent research seeking to go beyond reactive behaviour without resorting to hybrid solutions. The feasibility of recurrent neural network solutions are discussed and compared in the light of experiments designed to test ability to handle long-term temporal dependencies, in a more situated context than hitherto. It is concluded that a general-purpose recurrent network with some processing enhancements can begin to fulfil the requirements of this non-trivial problem.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Brooks, R. A.: A robust layered control system for a mobile robot, IEEE J. Robot. Anim. RA-2 (1986), 14–23

    Google Scholar 

  2. Brooks, R. A.: Intelligence without representation, Artif. Intel. 47 (1991), 139–159

    Google Scholar 

  3. Brooks, R. A.: Intelligence without reason, Proc. of the Twelfth Int. Conf. on Artificial Intelligence (IJCAI), Vol. 1, 1991, 569–595

    Google Scholar 

  4. Meeden, L.: Towards Planning, Incremental Investigations into Adaptive Robot Control, Ph.D. Thesis, Department of Computer Science, Indiana University, 1994.

  5. Rylatt, R. M., Czarnecki, C. A. and Routen, T. W.: A neural network based adaptive controller for mobile robots. Proc. Twelfth Int. Conf. on Systems Engineering, Coventry, U.K., Vol. 2, 1997, 587–592.

    Google Scholar 

  6. Ziemke, T.: Towards adaptive perception in autonomous robots using second-order recurrent networks, Proc. First Euromicro Workshop on Advanced Mobile Robots, Kaiseslautern, 1996, 89–98.

  7. Lin, L.-J. and Mitchell, T. M.: Reinforcement learning with hidden states. From Animals to Animats 2: Proc. Second Int. Conf. on the Simulation of Adaptive Behaviour, MIT Press, 1994, 271–280

  8. Meeden, L.: An incremental approach to developing intelligent neural network controllers for robots. IEEE Trans. Syst., Man and Cybernetics, 26, (Special Issue on Learning Autonomous Robots), 1996.

  9. Ziemke, T.: Towards adaptive behaviour system integration using connectionist infinite state automata. From Animals to Animats 4: Proc. Fourth Int. Conf. on the Simulation of Adaptive Behaviour, MIT Press/Bradford Books, Cambridge, MA., 1996.

    Google Scholar 

  10. Yamauchi, B. and Beer, R.: Integrating reactive, sequential and learning behaviour using dynamical neural networks. From Animals to Animats 3: Proc. Third Int. Conf. on the Simulation of Adaptive Behaviour, MIT Press, Cambridge, MA., 1994, 383–394.

    Google Scholar 

  11. Elman, J.: Finding structure in time, Cog. Sci. 14 (1990), 179–211.

    Google Scholar 

  12. Dorffner, G.: Neural networks and a new AI-questions and answers, In: Dorffner, G. (ed.) Neural Networks and a New Artificial Intelligence, International Thompson Computer Press, 1993.

  13. Ulbricht, C.: Handling time-warped sequences with neural networks. From Animals to Animats 4: Proc. Fourth Int. Conf. on Simulation of Adaptive Behaviour, 1996, 180–192.

  14. Verschure, P.M.: Connectionist explanation: taking positions in the mind-brain dilemma. In: Dorffner, G. (ed.) Neural Networks and a New Artificial Intelligence, International Thompson Computer Press, 1993, 134–187.

  15. Clarke, A.: Associative engines: connectionism, concepts, and representational change. Bradford Books, MIT Press, Cambridge, MA. 1993.

    Google Scholar 

  16. Stein, L. A.: Imagination and situated cognition, In: Ford, K. M., Glymour, C. and Hayes, P. J., (eds), Android Epistemology, AAAI Press/Press, Menlo Park, CA, 1995.

    Google Scholar 

  17. Mozer,M. C.: Neural net architectures for temporal sequence processing, In: Weigend, A. and Gerschenfeld, N., (eds), Predicting the Future and Understanding the Past, Addison Wesley, 1993.

  18. Ludik, J., Prins, W., Meert, K. and Catfolis, T.: A comparative study of fully and partially recurrent Networks. Proc. 1997 IEEE Int. Conf. on Neural Networks, (ICNN97), Houston, Texas, Vol. I, 292–297.

  19. Cottrell, G. W. and Tsung, F. S.: Learning simple arithmetic procedures, In: Barnden, J. A. and Pollack J. B., (eds.), High-Level Connectionist Models, 1991, 305–321.

  20. Lin, T., Horne, B. G., Tino, P. and Giles, L.: Learning long-term dependencies is not as difficult with NARX recurrent networks, IEEE Trans. Neural Networks, 1996.

  21. Kadaba, N., Nygard, K. E., Juell, P. L. and Kanga, L.: Modular backpropagation network for large domain pattern classification. Proc. IJCNN, Washington D.C., Vol. II, 1990, 551–554.

    Google Scholar 

  22. Tani, J. and Fukumura, N.: Learning goal directed sensory-based navigation of a mobile robot. Neural Networks 7 (1994), 550–564.

    Google Scholar 

  23. Rylatt, R. M., Czarnecki, C. A. and Routen, T. W.: A partially recurrent gating network approach to learning action selection by reinforcement. Proc. 1997 IEEE Int Conf. on Neural Networks, (ICNN97), Houston, TX, Vol. III, 1997, 1689–1698.

    Google Scholar 

  24. Dreyfus, H. and Dreyfus, S.: Mind over Machine: the Power of Human Intuition in the Era of the Computer. Free Press, 1986.

  25. Pal, P. K. and Kar, A.: Sonar-based mobile robot navigation through supervised learning in a neural net. Autonomous Robots 3 (1996), 355–374.

    Google Scholar 

  26. Rylatt, R. M., Czarnecki, C. A. and Routen, T. W.: Connectionist learning in behaviour-based mobile robots: a survey. Artif. Intel. Rev., Kluwer Academic Publishers (in press).

  27. Steels, L.: Emergent frame recognition and its use in artificial creatures. Proc. Twelfth Int. Joint Conf. on Artificial Intelligence, 1991, 1219–1224.

  28. Jordon, M. I.: Attractor dynamics and parallelism in a connectionist sequential machine. Proc. 8th. Conf. on Cognitive Science, 1986, 531–546.

  29. Peschl, M. F.: Autonomy vs. environmental dependency in neural knowledge representation. In: Brooks, R. A. and Maes, P., (eds), Artificial Life IV: Proc. Fourth International Workshop on the Synthesis and Simulation of Living Systems, 1993, 417–423.

  30. Rylatt, M, and Czarnecki, C.: Beyond physical grounding and naíve time: investigations into short-term memory for autonomous agents. In: From Animals to Animats 5, Bradford Books, MIT Press, 1998, 22–31.

Download references

Author information

Authors and Affiliations

  1. Department of Computer Science, De Montfort University, Leicester, U.K.

    R. M. Rylatt

  2. Department of Computer Science, De Montfort University, Leicester, U.K.

    C. A. Czarnecki

Authors
  1. R. M. Rylatt
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. C. A. Czarnecki
    View author publications

    You can also search for this author in PubMed Google Scholar

Rights and permissions

Reprints and permissions

About this article

Cite this article

Rylatt, R.M., Czarnecki, C.A. Embedding Connectionist Autonomous Agents in Time: The ‘Road Sign Problem’. Neural Processing Letters 12, 145–158 (2000). https://doi.org/10.1023/A:1009645229062

Download citation

  • Issue Date:

  • DOI: https://doi.org/10.1023/A:1009645229062

Search

Navigation