leggedrobotics/open3d_slam

Pointcloud-based graph SLAM written in C++ using open3D library.

What it solves

It provides a simplified, pointcloud-based SLAM (Simultaneous Localization and Mapping) system designed specifically for education. It removes the complexity of advanced, specialized algorithms to help newcomers enter the field of robotic mapping and localization.

How it works

Built on top of the Open3D library for 3D data processing, the system uses well-established algorithms in their basic form. It processes pointclouds to either build a map from scratch or localize a robot within an existing map, which can then be further extended with new measurements. It does not require additional sensor input like an IMU.

Who it’s for

Students, researchers, and developers new to SLAM who want a clear, accessible implementation of mapping and localization using pointclouds.

Highlights

  • Pointcloud-based mapping and localization.
  • Built on the high-performance Open3D library.
  • C++ library with ROS integration via a catkin wrapper.
  • No IMU required for operation.

Related

  • Project
    vikashplus/robohive

    A unified framework for robot learning that provides a collection of MuJoCo-simulated environments and tasks exposed via the OpenAI-Gym API.

  • Project
    Teddy-Liao/walk-these-ways-go2

    A Sim-to-Real locomotion starter kit for the Unitree Go2 quadruped robot that adapts the walk-these-ways framework to the unitree_sdk2 architecture.

  • Project
    pieeg-club/ironbci

    A low-cost, open-source brain-computer interface platform featuring 8-channel BLE hardware and SDKs for Python and Android.

  • Project
    xavierpuigf/virtualhome

    An interactive simulation platform for household activities that allows researchers to train embodied AI agents and generate synthetic video data via a Python API.