Reinforcement Learning in Robotics: Applications and Real-World Challenges†Reportar como inadecuado

Reinforcement Learning in Robotics: Applications and Real-World Challenges† - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

Department of Advanced Robotics, Istituto Italiano di Tecnologia, via Morego 30, 16163 Genova, Italy

Based on -Kormushev, P.; Calinon, S.; Caldwell, D.G.; Ugurlu, B. Challenges for the Policy Representation When Applying Reinforcement Learning in Robotics. In Proceedings of WCCI 2012 IEEE World Congress on Computational Intelligence, Brisbane, Australia, 10–15 June 2012-.


Author to whom correspondence should be addressed.

Abstract In robotics, the ultimate goal of reinforcement learning is to endow robots with the ability to learn, improve, adapt and reproduce tasks with dynamically changing constraints based on exploration and autonomous learning. We give a summary of the state-of-the-art of reinforcement learning in the context of robotics, in terms of both algorithms and policy representations. Numerous challenges faced by the policy representation in robotics are identified. Three recent examples for the application of reinforcement learning to real-world robots are described: a pancake flipping task, a bipedal walking energy minimization task and an archery-based aiming task. In all examples, a state-of-the-art expectation-maximization-based reinforcement learning is used, and different policy representations are proposed and evaluated for each task. The proposed policy representations offer viable solutions to six rarely-addressed challenges in policy representations: correlations, adaptability, multi-resolution, globality, multi-dimensionality and convergence. Both the successes and the practical difficulties encountered in these examples are discussed. Based on insights from these particular cases, conclusions are drawn about the state-of-the-art and the future perspective directions for reinforcement learning in robotics. View Full-Text

Keywords: reinforcement learning; robotics; learning and adaptive systems reinforcement learning; robotics; learning and adaptive systems

Autor: Petar Kormushev * , Sylvain Calinon and Darwin G. Caldwell



Documentos relacionados