Joel Loo

Papers

Total Citations

H-Index

About

Joel Loo is a robotics researcher whose work centers on enabling autonomous systems to understand and navigate open-world environments through natural language instructions. His primary research areas include semantic navigation, scene graph construction, and the integration of foundation models into robotic perception and planning pipelines. Loo’s most notable contribution is the development of OSG Navigator, a modular system that leverages open scene graphs to allow robots to search novel environments for target objects specified in natural language—a critical step toward general-purpose, open-world object-goal navigation. This work, published in 2025, has already garnered 4 citations, signaling its early impact on the field. By combining large language models with structured scene representations, Loo addresses the long-standing challenge of bridging high-level human commands with low-level robotic actions in unfamiliar settings. His research is particularly relevant for applications in service robotics, assistive technology, and autonomous exploration. As a rising figure in embodied AI, Loo’s contributions are shaping how robots can move beyond constrained lab settings into the dynamic, unstructured spaces of everyday life.