Abstract
Within the EMOBOT approach to adaptive behaviour, the task of learning to control the behaviour is one of the most interesting challenges. Learned action selection between classically implemented control mechanisms, with respect to internal values and sensor readings, provides a way to modulate a variety of behavioural capabilities. To demonstrate the potential of the learning emotional controller, we chose a 10-5-12 MLP to implement the σ, α controller of the EMOBOT. Since no teacher vector is available for the chosen task, the neural network is trained with a reinforcement strategy. The emotion-value-dependent reinforcement signal, together with the output of the network, is the basis with which to compute an artificial teacher vector. Then, the established gradient descent method (backpropagation of error) is applied to train the neural network. First results obtained by extensive simulations show that a still unrevealed richness in behaviour can be realised when using the neural-network-based learning emotional controller.
Similar content being viewed by others
References
Arkin RC (1998) Behaviour-based robotics. Bradford Books, MIT Press, Cambridge, Massachusetts
Fayek RE, Liscano R, Karam GM (1993) A system architecture for a mobile robot based on activities and a blackboard control unit. In: Proceedings of the IEEE international conference on robotics and automation, Los Alamitos, California, May 1993, pp 267–274
Tigli JY, Occello M, Thomas MC (1993) Towards a new intelligent reactive controller for autonomous mobile robots. In: Proceedings of the IEEE international conference on robotics and automation, Los Alamitos, California, May 1993, pp 249–254
Goerke N (2003) Systemic intelligence: methods for growing up artefacts that live. In: Proceedings of the international joint conference on neural networks (IJCNN 2003), Portland, Oregon, July 2003
Kintzler F, Schatten R (2002) Systemic architecture for growing up animats. In: Workshop on growing up artifacts that live—basic principles and future trends at the 7th international conference on the simulation of adaptive behavior (SAB 2002), Edinburgh, UK, August 2002
Elman et al (1999) Rethinking innateness: a connectionist perspective on development. MIT Press, Cambridge, Massachusetts
Skarmeta AG, Barberà HM, Alonso MS (1999) Learning behaviour fusion in autonomous mobile robots. In: Proceedings of 1999 EUSFLAT–ESTYLF joint conference, Mallorca, Spain, September 1999
Kintzler F (2003) Layout and first tests of a systemic architecture for growing up artefacts that live. In: Proceedings of the conference on the engineering application of neural networks (EANN 2003), Torremolinos, Malaga, Spain, September 2003, pp 72–79
Goerke N (2002) Perspectives for the next decade of neural computation. In: Proceedings of the conference on limitations and future trends in neural computation (LFTNC 2001), Siena, Italy, October 2001, pp 1–8
Gadanho SC, Hallam J (1998) Exploring the role of emotions in autonomous robot learning. In: Proceedings of the 1998 AAAI fall symposium—emotional and intelligent: the tangled knot of cognition. Also as DAI Research paper 909, University of Edinburgh, UK
Canamero D (1997) Modeling motivations and emotions as a basis for intelligent behavior. In: Proceedings of 1st international symposium on automation agents (AA’97), Marina del Rey, California, February 1997. ACM Press, New York
Damasio (1994) Descartes’ error: emotion, reason and the human brain. Picador, New York
Cos I, Hayes G (2002) Behaviour control using a functional and emotional model. In: Proceedings of the 7th international conference on the simulation of adaptive behavior (SAB 2002), from animals to animats 7, Edinburgh, UK, August 2002
Goerke N, Müller J (2003) Emotions are driving a robot: EMOBOT. In: Proceedings of the conference on the engineering application of neural networks (EANN 2003), Torremolinos, Malaga, Spain, September 2003, pp 56–63
McFarland D (1999) Animal behaviour. Addison Wesley, Longman, Harlow, England
McFarland D (1999) Biologie des Verhaltens. Spektrum Akademischer Verlag, Berlin, Germany
Braitenberg V (1984) Vehicles: experiments in synthetic psychology. MIT Press, Cambridge, Massachusetts
Kaelbling LP, Littman ML, Moore AW (1996) Reinforcement learning: a survey. J Art Int Res 4:237–285
Sutton RS, Barto AG (1998) Reinforcement learning: an introduction, MIT Press, Cambridge, Massachusetts
Acknowledgements
Part of this work is supported by the European Commission’s Information Society Technologies Programme, project SIGNAL, IST-2000-29225. Partners in this project are Napier University, National Research Council Genoa, Austrian Research Institute OFAI and the University of Bonn.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Goerke, N., Henne, T. & Müller, J. Neural networks for the EMOBOT robot control architecture. Neural Comput & Applic 13, 299–308 (2004). https://doi.org/10.1007/s00521-004-0424-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-004-0424-1