Back to Search
Start Over
An extended policy gradient algorithm for robot task learning
- Source :
- IROS, Intelligent Robots and Systems, 2007. IROS 2007. IEEE/RSJ International Conference on, IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, IROS'07, IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, IROS'07, Oct 2007, San Diego, United States. ⟨10.1109/IROS.2007.4399219⟩
- Publication Year :
- 2007
- Publisher :
- IEEE, 2007.
-
Abstract
- International audience; In real-world robotic applications, many factors, both at low-level (e.g., vision and motion control parameters) and at high-level (e.g., the behaviors) determine the quality of the robot performance. Thus, for many tasks, robots require fine tuning of the parameters, in the implementation of behaviors and basic control actions, as well as in strategic deci-sional processes. In recent years, machine learning techniques have been used to find optimal parameter sets for different behaviors. However, a drawback of learning techniques is time consumption: in practical applications, methods designed for physical robots must be effective with small amounts of data. In this paper, we present a method for concurrent learning of best strategy and optimal parameters, by extending the policy gradient reinforcement learning algorithm. The results of our experimental work in a simulated environment and on a real robot show a very high convergence rate.
- Subjects :
- business.industry
Computer science
Active learning (machine learning)
Robot performance
Online machine learning
02 engineering and technology
010501 environmental sciences
Machine learning
computer.software_genre
Motion control
Optimal parameters
01 natural sciences
Robot learning
Rate of convergence
0202 electrical engineering, electronic engineering, information engineering
[INFO.INFO-RB]Computer Science [cs]/Robotics [cs.RO]
Robot
Reinforcement learning
020201 artificial intelligence & image processing
Artificial intelligence
business
International conferences
computer
0105 earth and related environmental sciences
Subjects
Details
- Database :
- OpenAIRE
- Journal :
- 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems
- Accession number :
- edsair.doi.dedup.....a41a9bdb6710c1ffea2708c2ea938ea1