Projects per year
Abstract
Autonomous robots often require domain knowledge to act intelligently in their environment. This is particularly true for robots that use automated planning techniques, which require symbolic representations of the operating environment and the robot's capabilities. However, the task of specifying domain knowledge by hand is tedious and prone to error. As a result, we aim to automate the process of acquiring general common sense knowledge of objects, relations, and actions, by extracting such information from large amounts of natural language text, written by humans for human readers. We present two methods for knowledge acquisition, requiring only limited human input, which focus on the inference of spatial relations from text. Although our approach is applicable to a range of domains and information, we only consider one type of knowledge here, namely object locations in a kitchen environment. As a proof of concept, we test our approach using an automated planner and show how the addition of common sense knowledge can improve the quality of the generated plans.
Original language | English |
---|---|
Title of host publication | Robotics and Automation (ICRA), 2014 IEEE International Conference on |
Publisher | Institute of Electrical and Electronics Engineers |
Pages | 3749-3756 |
Number of pages | 8 |
DOIs | |
Publication status | Published - 1 May 2014 |
Keywords / Materials (for Non-textual outputs)
- control engineering computing
- knowledge acquisition
- mobile robots
- path planning
- automated planning techniques
- autonomous robots
- domain knowledge
- general common sense knowledge
- human readers
- kitchen environment
- natural language text
- object locations
- proof of concept
- robot planning
- symbolic representations
- Abstracts
- Ontologies
- Pattern matching
- Planning
- Robot sensing systems
- Syntactics
Fingerprint
Dive into the research topics of 'Extracting common sense knowledge from text for robot planning'. Together they form a unique fingerprint.Projects
- 1 Finished
-
Xperience - 'Robotes Bootstrapped through Learning from Experience'
Steedman, M., Geib, C. & Petrick, R.
1/01/10 → 31/12/15
Project: Research