Abstract
In this paper, we realize an action learning of obstacle avoidance by Profit Sharing using Self-Organizing Map-based Probabilistic Associative Memory (SOMPAM). In this method, patterns corresponding to the pairs of observation and action are memorized to the SOMPAM, and the brief degree is set to value of the rule. In this research robot learns with the aim of acquiring an action rule that can reach the goal point from the start point with as few steps as possible while avoiding collision with the obstacle. We use the reduced image of the image taken with the small camera mounted on the robot as observation. In the simulation environment reproducing the experimental environment, we confirmed that the learning converged to a state where it can reach the goal while avoiding obstacles with the minimum steps. Moreover, even in the real environment, it was confirmed that the robot can reach the goal while avoiding obstacles.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Sutton, R.S., Barto, A.G.: Reinforcement Learning: An Introduction. The MIT Press, Cambridge (1998)
Shibata, K., Sakashita, Y.: Reinforcement learning with internal-dynamics-based exploration using a chaotic neural network. In: Proceedings of IEEE and INNS International Joint Conference on Neural Networks, Killarney (2015)
Koma, D., Osana, Y.: Profit sharing that can learn deterministic policy for POMDPs environments by Kohonen feature map associative memory. In: Proceedings of IEEE International Conference on System, Man and Cybernetics, Manchester (2013)
Katayama, T., Osana, Y.: Realization of profit sharing by self-organizing map-based probabilistic associative memory. In: Proceedings of International Conference on Artificial Neural Networks (2016)
Osana, Y.: Self-organizing map-based probabilistic associative memory. In: Proceedings of International Conference on Neural Information Processing, Kuching (2014)
Grefenstette, J.J.: Credit assignment in rule discovery systems based on genetic algorithms. Mach. Learn. 3, 225–245 (1988)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Temma, D., Osana, Y. (2017). Obstacle Avoidance by Profit Sharing Using Self-Organizing Map-Based Probabilistic Associative Memory. In: Lintas, A., Rovetta, S., Verschure, P., Villa, A. (eds) Artificial Neural Networks and Machine Learning – ICANN 2017. ICANN 2017. Lecture Notes in Computer Science(), vol 10613. Springer, Cham. https://doi.org/10.1007/978-3-319-68600-4_7
Download citation
DOI: https://doi.org/10.1007/978-3-319-68600-4_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-68599-1
Online ISBN: 978-3-319-68600-4
eBook Packages: Computer ScienceComputer Science (R0)