Commonsense Scene Semantics for Cognitive Robotics: Towards Grounding Embodied Visuo-Locomotive Interactions

TL;DR

This paper introduces a comprehensive model that combines visual processing and human-centered spatial reasoning to improve understanding of robot interactions with environments.

Contribution

It presents an integrative methodology for grounding visuo-spatial and locomotive interactions in robotics using AI-based semantic models.

Findings

Effective semantic grounding demonstrated in object interactions

Successful indoor movement understanding

Framework bridges low-level vision and high-level spatial reasoning

Abstract

We present a commonsense, qualitative model for the semantic grounding of embodied visuo-spatial and locomotive interactions. The key contribution is an integrative methodology combining low-level visual processing with high-level, human-centred representations of space and motion rooted in artificial intelligence. We demonstrate practical applicability with examples involving object interactions, and indoor movement.