Authors:
Describes a software architecture (the PetBrain, built on OpenCog/OpenCogPrime) that enables a virtual agent in an online virtual world to carry out simple English-language interactions grounded in its perceptions and actions. It shows how embodiment and perception guide anaphor resolution and how natural-language generation answers simple questions about the observed world, integrating language with perception, motor control, and attention in an integrative-AGI design.
A foundational embodied-NLP source linking the semantic-parsing pipeline (Link Grammar / RelEx / NLG) to grounded perception–action in the OpenCog/PRIMUS lineage — relevant to the Semantic Parsing Deep Dive's account of grounded, logically-manipulable language representations.