NuRF: Nudging the Particle Filter in Radiance Fields for Robot Visual Localization

Abstract

Can we localize a robot on a map only using monocular vision? This study presents neural radiance field (NuRF), an adaptive and nudged particle filter framework in radiance fields for six degree-of-freedom (6-DoF) robot visual localization. NuRF leverages recent advancements in radiance fields and visual place recognition. Conventional visual place recognition meets the challenges of data sparsity and artifact-induced inaccuracies. By utilizing radiance field-generated novel views, NuRF enhances visual localization performance and combines coarse global localization with the fine-grained pose tracking of a particle filter, ensuring continuous and precise localization. Experimentally, our method converges seven times faster than existing Monte Carlo-based methods and achieves localization accuracy within 1 m, offering an efficient and resilient solution for indoor visual localization.