A Robotic World Model Framework Designed to Facilitate Human-robot Communication

Meghann Lomas,  Ernest Cross,  Jonathan Darvill,  Robert Garrett,  Michael Kopack,  Kenneth Whitebread
Lockheed Martin Advanced Technology Laboratories


Abstract

We describe a novel world model framework designed to support situated human-robot communication through improved mutual knowledge about the physical world. This work focuses on enabling a robot to store and use semantic information from a human located in the same environment as the robot and respond using human-understandable terminology. This facilitates information sharing between a robot and a human and subsequently promotes team-based operations. Herein, we present motivation for our world model, an overview of the world model, a discussion of proof-of-concept simulations, and future work.