Teaching robots to plan through Q-learning

Authors:
Dongbing Gu;Huosheng Hu
Affiliations:
Department of Computer Science, University of Essex, Wivenhoe Park, Colchester CO4 3SQ (UK) E-mail: dgu@essex.ac.uk, hhu@essex.ac.uk;Department of Computer Science, University of Essex, Wivenhoe Park, Colchester CO4 3SQ (UK) E-mail: dgu@essex.ac.uk, hhu@essex.ac.uk
Venue:
Robotica
Year:
2005

Citing 13
Cited 0

Today the earwig, tomorrow man?

Artificial Intelligence
Technical Note: \cal Q-Learning

Machine Learning
Self-Improving Reactive Agents Based on Reinforcement Learning, Planning and Teaching

Machine Learning
Three-layer architectures

Artificial intelligence and mobile robots
An Behavior-based Robotics

An Behavior-based Robotics
Introduction to Reinforcement Learning

Introduction to Reinforcement Learning
Reinforcement Learning in the Multi-Robot Domain

Autonomous Robots
Development of an Autonomous Quadruped Robot for Robot Entertainment

Autonomous Robots - Special issue on autonomous agents
Continuous-Action Q-Learning

Machine Learning
Evolving Fuzzy Logic Controllers for Sony Legged Robots

RoboCup 2001: Robot Soccer World Cup V
Integration of partially observable markov decision processes and reinforcement learning for simulated robot navigation

Integration of partially observable markov decision processes and reinforcement learning for simulated robot navigation
Evaluating the dynamics of agent-environment interaction

Evaluating the dynamics of agent-environment interaction
Reinforcement learning: a survey

Journal of Artificial Intelligence Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a Q-learning approach to state-based planning of behaviour-based walking robots. The learning process consists of a teaching stage and an autonomous learning stage. During the teaching stage, the robot is instructed to operate in some interesting areas of the solution space to accumulate some prior knowledge. Then, the learning is switched to the autonomous learning stage to let the robot explore the solution space based on its prior knowledge. Experiments are conducted in the RoboCup domain and results show a good performance of the proposed method.