small, powerful and beautiful slam tricks with theory and practice

[IEEE T-RO 2025] iKalibr: Multi-Sensor Calibration (Extrinsics & Time Offsets)

[TRO 2025] NeuPAN: Direct Point Robot Navigation with End-to-End Model-based Learning.

[RSS 2026] OpenFrontier: General Navigation with Visual-Language Grounded Frontiers

Modern robot motion planning library based on Pinocchio.

(ICRA 2025) LiLoc: Lifelong Localization using Adaptive Submap Joining and Egocentric Factor Graph

VLX-Go: open-world embodied AI navigation

SLAMFormer-Infinity: Infinite SLAM Transformer for Unbounded Frontend and Backend Processing

[CVPR2026]Official implementation of "StereoWorld: Geometry-Aware Monocular-to-Stereo Video Generation"

StereoVLA is powered by stereo vision and supports flexible deployment with high tolerance to camera pose variations.

MAC (Multi-Agent CAD): A decoupled multi-agent framework for text-to-CAD generation via constrained test-time compute

RLinf: Reinforcement Learning Infrastructure for Embodied and Agentic AI

[TMECH'2024] PALoc: Advancing SLAM Benchmarking with Prior-Assisted 6-DoF Trajectory Generation and Uncertainty Estimation

Universal Trajectory Optimization Framework for Differential Drive Robot Class

Generate production-quality SVG+PNG technical diagrams from natural language. 7 styles, UML support, and AI/Agent workflow patterns.

This library is an implementation of the algorithm described in Distributed Trajectory Estimation with Privacy and Communication Constraints: a Two-Stage Distributed Gauss-Seidel Approach.

First Attempt on IMU "Foundation Model"

This repo accompanies the research paper, ARKitScenes - A Diverse Real-World Dataset for 3D Indoor Scene Understanding Using Mobile RGB-D Data and contains the data, scripts to visualize and proces…

A simple system that can relocalize a robot on a built map is developed in this system. The system is based on LIO-SAM.

VidMap: Exploiting Temporal Structure for Video-Based Structure-from-Motion (ECCV 2026)

Implementation of Tightly Coupled 3D Lidar Inertial Odometry and Mapping (LIO-mapping)

[RA-L 2022] An efficient and probabilistic adaptive voxel mapping method for LiDAR odometry

A highly robust and accurate LiDAR-only, LiDAR-inertial odometry

Agent-native framework for rapid sim-to-real deployment of embodied AI for field robotics. Closed-loop, one-to-one workflow with hardware, self validating. Checkout develop branch for latest features.

[TRO2025] A LiDAR-based complete global localization method.

SuperMap is a living spatial memory for embodied AI — it perceives the world, remembers its evolution, and supports reasoning and action.