Learning visual docking for non-holonomic autonomous vehicles

Martinez-Marin, Tomas and Duckett, Tom (2008) Learning visual docking for non-holonomic autonomous vehicles. In: The Intelligent Vehicles 2008 Symposium (IV08), June 4-6, 2008 , Eindhoven, Netherlands.

Documents
Learning Visual Docking for Non-Holonomic Autonomous Vehicles
[img]
[Download]
[img]
Preview
PDF
tmtd_iv08ver5.8.pdf

1MB

Abstract

This paper presents a new method of learning visual docking skills for non-holonomic vehicles by direct interaction with the environment. The method is based on a reinforcement algorithm, which speeds up Q-learning by applying memorybased sweeping and enforcing the “adjoining property”, a filtering mechanism to only allow transitions between states that satisfy a fixed distance. The method overcomes some limitations of reinforcement learning techniques when they are employed in applications with continuous non-linear systems, such as car-like vehicles. In particular, a good approximation to the optimal
behaviour is obtained by a small look-up table. The algorithm is tested within an image-based visual servoing framework on a docking task. The training time was less than 1 hour on the real vehicle. In experiments, we show the satisfactory performance of the algorithm.

Item Type:Conference or Workshop Item (Paper)
Additional Information:This paper presents a new method of learning visual docking skills for non-holonomic vehicles by direct interaction with the environment. The method is based on a reinforcement algorithm, which speeds up Q-learning by applying memorybased sweeping and enforcing the “adjoining property”, a filtering mechanism to only allow transitions between states that satisfy a fixed distance. The method overcomes some limitations of reinforcement learning techniques when they are employed in applications with continuous non-linear systems, such as car-like vehicles. In particular, a good approximation to the optimal behaviour is obtained by a small look-up table. The algorithm is tested within an image-based visual servoing framework on a docking task. The training time was less than 1 hour on the real vehicle. In experiments, we show the satisfactory performance of the algorithm.
Keywords:mobile robot, vision-based control, reinforcement learning
Subjects:G Mathematical and Computer Sciences > G700 Artificial Intelligence
G Mathematical and Computer Sciences > G760 Machine Learning
G Mathematical and Computer Sciences > G400 Computer Science
G Mathematical and Computer Sciences > G740 Computer Vision
Divisions:College of Science > School of Computer Science
ID Code:1683
Deposited By: Tom Duckett
Deposited On:20 Nov 2008 16:33
Last Modified:13 Mar 2013 08:30

Repository Staff Only: item control page