Tom Williams and Matthias Scheutz
Robotics and Autonomous Systems
Natural language is a flexible and powerful control modality which can transform a wheelchair from a vehicle into a genuine helper. While autonomous wheelchairs are increasingly designed to use natural language for control, most of them only handle a small number of rigid commands. To establish the state-of-the-art in language-enabled wheelchairs and determine how to improve natural language capabilities, we introduce a framework for analyzing and classifying properties of language-enabled wheelchairs. We then apply the framework to the twenty-four most recent natural language-enabled wheelchair projects, in order to compare their achievements and identify areas for improvement.