Newest papers

RangeNav: Task-Aligned LiDAR Range-Image Pretraining for Robot Navigation

Samuel Foo Enze, Manthan Patel, Fan Yang, Linus Kramer, Pascal Roth, Marco Hutter, César Cadena · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.29845 · cited by 0 works (OpenAlex) · open-access PDF

French

Les robots mobiles de navigation autonomes opérant dans des foules dense et des sites troublés nécessitent une perception de 360 $ environ pour réagir aux dangers environnants. Alors que le LiDAR panoramique offre une couverture complète sur le terrain, les représentations d'apprentissage au sein de l'apprentissage du renforcement (RL) sont inefficaces sous des récompenses rares, et les encodeurs point-cloud prétrainés 3D existants sont trop lourds en mémoire pour la simulation RL massivement parallèle. Nous présentons RangeNav, un encodeur d'image de gamme compact (183k paramètres) LiDAR prétrainé entièrement sur les données synthétiques. RangeNav utilise des convol

Machine translation (Meta M2M100); the English original follows.

Abstract

Autonomous mobile robots operating in dense crowds and cluttered sites require $360^\circ$ perception to react to surrounding hazards. While panoramic LiDAR offers full-surround coverage, learning representations within reinforcement learning (RL) is sample-inefficient under sparse rewards, and existing 3D pretrained point-cloud encoders are too memory-heavy for massively parallel RL simulation. We present RangeNav, a compact (183k parameters) LiDAR range-image encoder pretrained entirely on synthetic data. RangeNav uses anisotropic convolutions and circular padding to handle the extreme aspect ratio and azimuthal seam of range images. We pretrain the network using a navigation-aligned objective combining range reconstruction, semantic segmentation, and bird's-eye-view (BEV) prediction, then freeze it for policy learning. In simulation, adding semantic and BEV supervision increases success rates over a reconstruction-only baseline by an average of 32.03 and 23.73 percentage points (pp) in procedurally generated training and out-of-distribution realistic test environments, respectively. Zero-shot real-world deployments on a Unitree B2-W quadruped demonstrate robust navigation through dense crowds, low-light settings, and construction zones where limited-field-of-view depth policies struggle. Code and models will be open-sourced.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Lidar, Computer science, Artificial intelligence, Computer vision, Encoder, Ranging

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange