Yicheng Jiang

Papers

Total Citations

H-Index

About

Yicheng Jiang is a rising researcher in embodied AI and urban navigation, whose work bridges computer vision and robotics to enable autonomous agents to operate in complex, real-world environments. His most-cited paper, "CityWalker: Learning Embodied Urban Navigation from Web-Scale Videos" (2025, 4 citations), tackles the critical challenge of navigating dynamic, map-free urban spaces—a problem that has stymied traditional visual navigation systems. By leveraging web-scale video data, Jiang’s approach teaches agents to reason spatially and follow common-sense norms, such as avoiding off-street obstacles or adapting to pedestrian flows, without relying on pre-built maps. This work directly addresses the limitations of existing methods in off-street and unstructured settings, paving the way for more robust autonomous deployment in cities. Though early in his career, Jiang’s focus on learning from diverse, real-world video sources marks a significant step toward generalizable embodied intelligence. His research holds promise for applications in delivery robots, assistive navigation for the visually impaired, and autonomous vehicles, positioning him as a key contributor to the next generation of urban AI systems.