Spatial Representation and Reasoning for Human-Robot Collaboration

Abstract

How should a robot represent and reason about spatial information when it needs to collaborate effectively with a human? The form of spatial representation that is useful for robot navigation may not be useful in higher-level reasoning or working with humans as a team member. To explore this question, we have extended previous work on how children and robots learn to play hide and seek to a human-robot team covertly approaching a moving target. We used the cognitive modeling system, ACT-R, with an added spatial module to support the robot’s spatial reasoning. The robot interacted with a team member through voice, gestures, and movement during the team’s covert approach of a moving target. This paper describes the new robotic system and its integration of metric, symbolic, and cognitive layers of spatial representation and reasoning for its individual and team behavior. similar to a person’s will be able to collaborate with a person better than a computational system that does not. Furthermore, such a system will be less likely to exhibit unreasonable behavior, which is a sure benefit in any strategic domain. Our principal goal in this project is to show how the scientifically-principled integration of computational cognitive models can facilitate human-robot interaction and, specifically, how different spatial representations need to be integrated for coherent human-robot interaction (HRI). Note that this paper reports a systems approach to HRI; psychological studies and usability tests will be performed in the future. In addition, our engineering goal in this project is to create a system that can covertly approach a moving target with a team member in a laboratory scenario inspired by the Marine RECON task.