Processing Natural Language About Ongoing Actions in ECG

TitleProcessing Natural Language About Ongoing Actions in ECG
Publication TypeConference Paper
Year of Publication2017
AuthorsDoubleday, S., Trott S., & Feldman J.
Published inProceedings of the AAAI 2017 Spring Symposium on Computational Construction Grammar and Natural Language Understanding
KeywordsAI, ECG, LCAS, NLU, Petrinet, X-net

Actions may not proceed as planned; they may be interrupted, resumed or overridden. This is a challenge to handle in a natural language understanding system. We describe extensions to an existing implementation for the control of autonomous systems by natural language, which enables such systems to handle incoming language requests regarding actions. Language Communication with Autonomous Systems (LCAS) has been extended with support for X-nets, parameterized executable schemas representing actions. X-nets enable the system to control actions at a desired level of granularity, while providing a mechanism for language requests to be processed asynchronously. Standard semantics implemented include requests to stop, continue, or override the existing action. The specific domain demonstrated is the control of motion of a simulated robot, but the approach is general, and could be applied to other domains.


We thank Luca Gilardi for his help in designing the system framework, and for the creation of the GUI for the ECG Workbench Editor. Professor William Knottenbelt and his students at Imperial College, London, created and have maintained PIPE through many releases. Sarah Tattersall performed a thorough refactoring of the PIPE code base, without which the external system extensions would not have been possible.

ICSI Research Group