A repeatable checklist for comparing open-source SLAM and visual-inertial odometry systems before integrating them into a UAV stack.
how to Reviewed 8/14/2026
Which open-source SLAM or VIO system should I evaluate first for my UAV's sensors, compute and mission profile?
Open each candidate repository and record the required sensors: monocular, stereo, RGB-D, LiDAR or IMU. A LiDAR SLAM system is not interchangeable with a visual one.
Record the SPDX license from the official API and the last activity/archive flag. Unknown licenses stay Unknown.
Run each estimator on the dataset its official documentation recommends before any hardware work. Record the version you used.
Check the official README for compute requirements and compare them with your onboard hardware before adoption.
Confirm whether the system ships ROS/ROS 2 interfaces or a documented API for your flight stack.
This guide links the repositories; it does not replace their official setup instructions and it does not certify flight behavior.
ORB-SLAM3 is the reference open-source visual and visual-inertial SLAM library supporting monocular, stereo, RGB-D and IMU inputs with multi-map operation for UAV state estimation.
Real-Time SLAM for Monocular, Stereo and RGB-D Cameras, with Loop Detection and Relocalization Capabilities
VINS-Mono is a monocular visual-inertial state estimator from HKUST Aerial Robotics, commonly used as a UAV localization baseline on ROS.
An optimization-based multi-sensor state estimator
Visual Inertial Odometry with SLAM capabilities and 3D Mesh generation.
An open source platform for visual-inertial navigation research.
LIO-SAM: Tightly-coupled Lidar Inertial Odometry via Smoothing and Mapping
A computationally efficient and robust LiDAR-inertial odometry (LIO) package
Cartographer is a system that provides real-time simultaneous localization and mapping (SLAM) in 2D and 3D across multiple platforms and sensor configurations.
MRS UAV Core is CTU MRS's ROS metapackage for UAV control, state estimation, trajectory generation and tracking, bundling PX4 integration and simulation tooling.