Ultra-Fusion: A Resilient Tightly-Coupled Multi-Sensor Fusion SLAM Framework under Sensor Degradation and Spatiotemporal Perturbation

LiteVLoc: Map-Lite Visual Localization for Image Goal Navigation

Fusing GNSS and wheel measurements based on FAST-LIO and IKFOM

FAEL: Fast Autonomous Exploration for Large-Scale Environments with a Mobile Robot

ScaRF-SLAM: A Framework for Globally Consistent Online Visual Dense Reconstruction

Map, localize, and simulate — 3D Gaussian Splatting for robotics

Nav2 planners & controllers battle head-to-head. ROS 2 · ONNX · browser game.

Official implementation of paper "VLM³: Vision Language Models Are Native 3D Learners".

UMI-3D SLAM and Data Processing Pipeline: https://umi-3d.github.io/

ROS 2 sensor fusion SDK: UKF, 3D native, proper GNSS, zero manual tuning. Apache 2.0.

Official implementation of "Depth Any Panoramas: A Foundation Model for Panoramic Depth Estimation".

Code for "Efficient Robot Navigation Inspired by Honeybee Learning Flights"

Efficient Universal Perception Encoder: a single on-device vision encoder with versatile representations that match or exceed specialized experts across multiple task domains.

Fast Incremental Euclidean Distance Fields for Online Motion Planning of Aerial Robots

A flexible submap-based framework towards spatio-temporally consistent volumetric mapping and scene understanding.

Code for the CVPR2021 paper "Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition"

Training, optimization and deployment of Object Detection model with dinov2 backbone for efficient inference on NVIDIA Jetson

High performance volumetric occupancy and TSDF mapping

Official implementation of "LSGS-Loc: Towards Robust 3DGS-Based Visual Localization for Large-Scale UAV Scenarios". Code coming soon!

Free open-source photogrammetry software tools

[CVPR 2026 Highlight] ULF-Loc: Unbiased Landmark Feature for Robust Visual Localization with 3D Gaussian Splatting

Joint deep network for feature line detection and description

Pure-Rust visual & visual-inertial SLAM for GPS-denied robots and UAVs — beats ORB-SLAM2's published KITTI ATE on seq00/09, OV2SLAM-class accuracy on EuRoC

[RSS 2026] Causal video-action world model for generalist robot control

Scaling Text-to-Image Diffusion Transformers with Representation Autoencoders

Lightweight, GPU-free Python CLI to benchmark LLM-generated robotics: perception (GNSS, LiDAR, vision), planning (grid A*), control (1st-order tracking). Swappable model hooks + JSON.

TIPSv2 (CVPR'26) and TIPS (ICLR'25)