A robust, low-drift, real-time SLAM package for the Livox Horizon LiDAR, designed for highway autonomous driving scenarios.
Horizon Highway SLAM is a specialized SLAM (Simultaneous Localization and Mapping) package built for the Livox Horizon LiDAR sensor. It creates real-time maps and tracks a vehicle's position in highway environments while compensating for motion distortion and sensor limitations. The framework solves key perception problems for autonomous driving, such as maintaining accuracy at high speeds and fusing LiDAR with IMU data.
Autonomous vehicle engineers and robotics researchers working with Livox Horizon LiDAR who need a production-ready SLAM solution for highway testing and development. It's particularly relevant for teams building Level 3/4 autonomous driving systems.
Developers choose this package because it's specifically optimized for the Livox Horizon sensor and highway conditions, offering lower drift and better performance than generic SLAM solutions. The Docker-based deployment and precompiled binary make it easy to integrate into existing autonomous driving stacks.
Horizon_Highway_Slam Demo in Docker
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Tailored for 0-80 km/h speeds with motion distortion compensation, directly addressing key challenges in highway SLAM as highlighted in the README's feature list.
Implements tight coupling of LiDAR and IMU data to prevent scene degradation, improving accuracy for autonomous driving systems, with configurable IMU modes described in the launch parameters.
Processes data in real time for localization updates, crucial for autonomous navigation, as demonstrated in the provided demo videos and rosbag examples.
Offers a Docker-based installation method that simplifies environment setup and reproducibility, reducing dependency conflicts, though visualization requires extra steps.
Exclusively supports Livox Horizon LiDAR and its internal IMU, making it incompatible with other sensors, as noted in the README's restrictions on IMU usage.
Available only as a precompiled library, preventing code inspection, customization, and debugging for researchers or developers wanting to tweak algorithms.
Requires running Rviz separately on the host when using Docker, adding complexity to the workflow and potential configuration headaches for real-time monitoring.