LIO-SAM on Yonohub
Ahmed Radwan
Sr Software Engineer, COM-IoT Technologies. ?Ex: Avelabs, Yonohub Developer Advocate.
We are pleased to announce the release of LIO-SAM Mapping ready-to-use Block on YonoArc. You can use the block to create SUPER ACCURATE 3D maps using a pre-recorded rosbag. We also included a ROSBag Player block which is compatible with LIO-SAM block.
Learning Objectives
- What’s LIO-SAM Algorithm and what’s the difference between NDT mapping and LIO-SAM?
- Why use LIO-SAM package on Yonohub?
- Creating a 3D map using a prerecorded rosbag.
Introduction
3D mapping is an essential part of autonomous driving as localization algorithms depend on 3D maps to get better estimates for the location that can be accurate enough for an autonomous vehicle to operate and navigate through a certain path.
NDT Mapping is a good mapping algorithm and it’s available as a package in Autoware.ai but it has some defects. One of them is loop closure detection and correction which the NDT mapping algorithm fails to detect and correct the trajectory and the map when a loop closure is there. Also, the map created from the NDT mapping algorithm tends to shift over distance.
LIO-SAM (stands for Lidar Inertial Odometry via Smoothing and Mapping) is a highly accurate, real-time mobile robot trajectory estimator and map-builder. It achieves this performance by considering many factors such as loop closure detection, trajectory correction, and IMU measurement bias estimation.
LIO-SAM basically works by receiving data from a 3D lidar, an IMU, and optionally a GPS then estimates the robot state and trajectory by formulating the problem as a maximum a posteriori problem. the algorithm then uses a factor graph to solve the problem. The variable here which is the robot state can be known by four factors which are IMU preintegration factors, lidar odometry factors, GPS factors, and loop closure factors.
So, in terms of speed, accuracy, loop closure detection, and trajectory correction, LIO-SAM will be much better than NDT.
LIO-SAM on Yonohub
LIO-SAM block is available on Yonohub (completely free) to use without the need for any kind of installation. You’ll just need to upload your rosbag file and use LIO-SAM Mapping block with the LIO-SAM compatible ROSBag player and you are good to go. The following video is a demo of the LIO-SAM Mapping block running on YonoArc:
Yonohub
Yonohub is the first cloud-based system for designing, sharing, and evaluating autonomous vehicle algorithms using just blocks. Yonohub features a drag-and-drop tool to build complex systems consisting of many blocks, a marketplace to share and monetize blocks, a builder for custom environments, and much more.
Get $25 free credits when you sign up now. For researchers and labs, contact us to learn more about Yonohub sponsorship options. Yonohub: A Cloud Collaboration Platform for Autonomous Vehicles, Robotics, and AI Development. www.yonohub.com
If you liked this article, please consider following us on Twitter at @yonohub, email us directly, or find us on LinkedIn. I’d love to hear from you if I can help you or your team with how to use YonoHub.
References
https://arxiv.org/abs/2007.00258
https://github.com/TixiaoShan/LIO-SAM?