OdorSpaces: Visual-Olfactory Embodied Navigation in 3D Environments
OdorSpaces: Visual-Olfactory Embodied Navigation in 3D Environments
This paper presents OdorSpaces, a physics-based simulation platform and benchmark for visual-olfactory embodied navigation. It augments Habitat-Sim with 3D odor transport simulated by Fire Dynamics Simulator (FDS), enabling agents to navigate using both visual and olfactory observations. The benchmark covers two source-seeking scenarios: invisible gas-leak localization and early-stage fire localization. An initial multimodal reinforcement learning policy integrating visual and olfactory data over time is developed, showing superior performance against diverse baselines in these tasks.
The authors introduce OdorSpaces, a simulation benchmark integrating visual and olfactory data for embodied navigation, demonstrating that their multimodal reinforcement learning policy outperforms baselines in gas-leak and fire-localization scenarios within Habitat-Sim augmented by Fire Dynamics Simulator.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Environment
- SimulationOpen source S2
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
arXiv:2611.05738v1 Announce Type: new Abstract: Olfaction provides critical information about airborne chemical cues, yet it remains largely underexplored in embodied AI, with limited prior machine olfaction work focusing primarily on semantic odor perception. We introduce OdorSpaces, a physics-based simulation platform and benchmark for visual-olfactory embodied navigation. OdorSpaces augments Habitat-Sim with 3D od
Open source S2
Source：arXiv Robotics — research abstracts · arxiv.org