A Planning and Learning Hierarchy using Qualitative Reasoning for the On-Line Acquisition of Robotic Behaviors

Wiley, T, Sammut, C, Hengst, B and Bratko, I 2016, 'A Planning and Learning Hierarchy using Qualitative Reasoning for the On-Line Acquisition of Robotic Behaviors', Advances in Cognitive Systems, vol. 4, pp. 93-111.


Document type: Journal Article
Collection: Journal Articles

Title A Planning and Learning Hierarchy using Qualitative Reasoning for the On-Line Acquisition of Robotic Behaviors
Author(s) Wiley, T
Sammut, C
Hengst, B
Bratko, I
Year 2016
Journal name Advances in Cognitive Systems
Volume number 4
Start page 93
End page 111
Total pages 19
Publisher Cognitive Systems Foundation
Abstract Trial-end-error learning is often needed to acquire a new skill. Humans can use domain knowledge to minimize the number of trials required. However, existing reinforcement learning systems are either incapable of reasoning about domain knowledge or use hard-coded domain knowledge. Thus, these systems are insufficient for the online learning of robotic skills. We present a hierarchical architecture that learns the domain knowledge of a robotic system in the form of a qualitative model. The model is used by a symbolic planner that reduces the search space for trial-and-error learning. We evaluate the architecture on a real robot that learns to climb over obstacles.
Subject Adaptive Agents and Intelligent Robotics
Keyword(s) Machine Learning
Qualitative Models
Rescue Robots
Copyright notice © 2016 Cognitive Systems Foundation
ISSN 2324-8416
Versions
Version Filter Type
Access Statistics: 9 Abstract Views  -  Detailed Statistics
Created: Tue, 26 Nov 2019, 08:38:00 EST by Catalyst Administrator
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us