论文信息 - Reinforcement learning on an omnidirectional mobile robot

Reinforcement learning on an omnidirectional mobile robot

With this paper we describe a well suited, scalable problem for reinforcement learning approaches in the field of mobile robots. We show a suitable representation of the problem for a reinforcement approach and present our results with a model based standard algorithm. Two different approximators for the value function are used, a grid based approximator and a neural network based approximator.

Martin A. Riedmiller | Roland Hafner | Roland Hafner

[1] B. Achiriloaie,et al. VI REFERENCES , 1961 .

[2] Richard W. Prager,et al. Reinforcement learning methods for multi-linked manipulator obstacle avoidance and control , 1993, Proceedings 1993 Asia-Pacific Workshop on Advances in Motion Control.

[3] Jorge Angeles,et al. The design of kinematically isotropic rolling robots with omnidirectional wheels , 1995 .

[4] Martin Spott,et al. Fynesse: a hybrid architecture for self-learning control , 2000 .

[5] Bernhard Nebel,et al. Towards a Life-Long Learning Soccer Agent , 2002, RoboCup.

[6] Artur Merke,et al. A Necessary Condition of Convergence for Reinforcement Learning with Function Approximation , 2002, ICML.

[7] Martin A. Riedmiller,et al. Using Machine Learning Techniques in Complex Multi-Agent Domains , 2003 .