Anuj Pokhrel

Falling Short: How Lower Camera Height Degrades Image-to-OpenStreetMap Localization for Sidewalk Robots

This work introduces L OW L OC, a cross-view I2O evaluation dataset that is the first to explicitly provide viewpoints of different camera heights on sidewalk robots and characterize camera-height sensitivity in I2O localization.

Evaluating Early Multimodal Fusion in Discrete Latent World Models for Robot Navigation

Hydra: A Navigation World Action Model with Discrete Latent Planning and Continuous Flow-Matching Execution

Hydra, a discrete World Action Model that tackles this by establishing a unified latent manifold over visual states, physical poses, and control actions, outperforms state-of-the-art navigation world models in goal-directed planning, while matching or exceeding the closed-loop execution capabilities of leading reactive...