Module 09 · Phase 4: Localization and autonomy
SLAM Selection, Configuration & Troubleshooting
Pick a SLAM system for the sensors and compute you have, configure it, and diagnose drift and tracking loss. ORB-SLAM3, VINS-Fusion, OpenVINS, LIO-SAM, FAST-LIO2 and KISS-ICP are compared from their official documentation and repositories.
This overview is open. 5 of 5 guides are written; the others show Coming soon until they are complete and tested.
Questions this module answers
- Which SLAM algorithm fits my sensors, environment and compute?
- When should I use VIO instead of LiDAR SLAM?
- Why does SLAM drift or fail in repetitive environments?
- How do I diagnose tracking loss and evaluate map quality and localization accuracy?
What it covers
- Visual, monocular, stereo, RGB-D and visual-inertial SLAM
- LiDAR SLAM and LiDAR-inertial odometry
- Loop closure, relocalization, pose-graph optimization
- Drift, mapping, failure detection
- Six-system comparison verified against official sources
Guides and design decisions
A. Engineering Guides (3)
B. Design & Selection (2)
Diagnostics, tools and challenges
C. Troubleshooting (3)
- SLAM driftsThe map bends or the trajectory does not close on revisit.
- SLAM fails in repetitive environmentsCorridors or aisles are matched to the wrong place.
- Visual tracking is lostThe visual system reports lost tracking or reinitializes.
D. Interactive tools (1)
- SLAM Selection GuideCompare SLAM approaches by sensors, environment, accuracy, compute, mapping needs, real-time limits and engineering effort.
E. Design challenges (2)
- Challenge 1: Design an Autonomous Mobile RobotIndoor differential-drive robot: sensors, motor sizing, compute, power, ROS 2, LiDAR+IMU, SLAM, Nav2, tuning and validation.
- Challenge 4: UAV GNSS-Denied NavigationCamera, LiDAR and IMU on a UAV: compute, calibration, VIO/LIO, estimation, flight-controller integration and failure handling.
Practical deliverables
- SLAM selection guide
- Verified comparison of six SLAM projects
- Real configuration examples
Read first: existing FixTheRobot lessons and labs
This module builds on these instead of repeating them.
- LIO-SAM SLAM · Lesson
- FAST-LIO and FAST-LIO2 SLAM · Lesson
- ORB-SLAM Architecture SLAM · Lesson
- Failure Modes in Visual SLAM SLAM · Lesson
- SLAM Toolbox and Mapping Workflows SLAM · Lesson