Paper Details: Downloads: 425
Serial Number: P1130502001
Title: Using Knowledge about Predicted Robot Behaviors to Solve Ambiguity in Natural Language Instructions
Authors: Fangju Wang and Shaidah Jusoh and Simon Yang
Abstract: Natural language interfaces are desirable for communicating with robots, especially for service robots, which are built for helping people in homes, oÆces, etc. Natural language interfaces can make robots more accessible to people who do not have formal training in robot control. A major challenge in building a natual language interface is ambiguity in natural language instructions. In our research, we develop a new approach for handling ambiguity in robot natural language interfaces. This approach is based on the possibility theory of fuzzy sets. It is characterized by using the knowledge about predicted behaviors of robots.
Keywords: Robot, Natural language interface, fuzzy set, possibility theory.
Journal/Conference: International Journal of Automation, Robotics and Autonomous Systems
Volume: 5
Issue: 1
Submission Date: 1/1/2005 12:00:00 AM
Review Date: 3/1/2005 12:00:00 AM
Publishing Date: 6/1/2005 12:00:00 AM
Article Downloads: 425
Download:

Facebook