Abstract
This paper presents a novel method for real-time coordination control of multiagent systems in maximizing global benefits keeping a balance with individual benefits of agents. In this coordination mechanism a reinforcement-learning agent learns to select its action estimating global state value and immediate reward. The estimated global state value of the system makes an agent cooperative with others. This learning method is implemented in the test bed multiagent transportation service control for a city. The outstanding performance of the proposed method in different aspects compared to other heuristic methods indicates its effectiveness for multiagent cooperative systems.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Tan, M.: Multi-agent Reinforcement Learning: Independent vs. Cooperative Agents. In: Proceedings of the Tenth International Conference on Machine Learning, pp. 330–337 (1993)
Whitehead, S.D.: A Complexity Analysis of Cooperative Mechanisms in Reinforcement Learning. In: Proceedings of AAAI, pp. 607–613 (1991)
Sutton, R.S., Barto, A.G.: Reinforcement Learning: An Introduction. MIT Press, Cambridge (1998)
Kamal, M.A.S., Murata, J., Hirasawa, K.: Task Oriented Reinforcement Learning for Continuing Task in Dynamic Environment. Research Reports on Information Science and Electrical Engineering of Kyushu University 9(1), 7–12 (2004)
Kamal, M.A.S., Murata, J., Hirasawa, K.: Task-Oriented Reinforcement Learning for Continuous Tasks in Dynamic Environment. In: Proceeding of the SICE annual conference, pp. 932–935 (2002)
Kamal, M.A.S., Murata, J., Hirasawa, K.: Task-Oriented Multiagent Reinforcement Learning Control for a Real Time High-Dimensional Problem. In: Proceedings of the Eighth International Symposium on Artificial Life and Robotics (AROB), vol. 2, pp. 353–356 (2003)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Kamal, M.A.S., Murata, J. (2004). Coordination in Multiagent Reinforcement Learning Systems. In: Negoita, M.G., Howlett, R.J., Jain, L.C. (eds) Knowledge-Based Intelligent Information and Engineering Systems. KES 2004. Lecture Notes in Computer Science(), vol 3213. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30132-5_162
Download citation
DOI: https://doi.org/10.1007/978-3-540-30132-5_162
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-23318-3
Online ISBN: 978-3-540-30132-5
eBook Packages: Springer Book Archive