Policy Gradient Methods for Robotics
Top Cited Papers
- 1 October 2006
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21530858,p. 2219-2225
- https://doi.org/10.1109/iros.2006.282564
Abstract
The acquisition and improvement of motor skills and control policies for robotics from trial and error is of essential importance if robots should ever leave precisely pre-structured environments. However, to date only few existing reinforcement learning methods have been scaled into the domains of high-dimensional robots such as manipulator, legged or humanoid robots. Policy gradient methods remain one of the few exceptions and have found a variety of applications. Nevertheless, the application of such methods is not without peril if done in an uninformed manner. In this paper, we give an overview on learning with policy gradient methods for robotics with a strong focus on recent advances in the field. We outline previous applications to robotics and show how the most recently developed methods can significantly improve learning performance. Finally, we evaluate our most promising algorithm in the application of hitting a baseball with an anthropomorphic armKeywords
This publication has 17 references indexed in Scilit:
- Natural Policy Gradient Reinforcement Learning for a CPG Control of a Biped RobotPublished by Springer Nature ,2004
- Introduction to Stochastic Search and OptimizationPublished by Wiley ,2003
- Gradient estimation for ratiosPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- On Choosing and Bounding Probability MetricsInternational Statistical Review, 2002
- Practical Methods of OptimizationPublished by Wiley ,2000
- Modelling and Control of Robot ManipulatorsPublished by Springer Nature ,2000
- Natural Gradient Works Efficiently in LearningNeural Computation, 1998
- Statistical Inference, Occam's Razor, and Statistical Mechanics on the Space of Probability DistributionsNeural Computation, 1997
- A Kendama Learning Robot Based on Bi-directional TheoryNeural Networks, 1996
- Simple Statistical Gradient-Following Algorithms for Connectionist Reinforcement LearningPublished by Springer Nature ,1992