LVI-Q: Robust LiDAR-Visual-Inertial-Kinematic Odometry for Quadruped Robots Using Tightly-Coupled and Efficient Alternating Optimization
Kevin Christiansen Marsim, Minho Oh, Byeongho Yu, Seungjae Lee, I Made Aswin Nahrendra, HYUNGTAE LIM, Hyun Myung
AI summary
Problem
Existing LiDAR-visual-inertial odometry systems struggle with sensor degeneracy, rapid motion, and illumination changes due to rigid fusion strategies that cannot adapt to varying sensor reliability in challenging environments.
Approach
The system dynamically alternates between a filter-based LiDAR-inertial-kinematic odometry module and a factor-graph-based visual-inertial-kinematic odometry module, using foot-preintegration and depth consistency factors to constrain pose estimation based on real-time sensor availability.
Key results
- Proposes a VIKO module utilizing kinematic and visual feature depth consistency factors for consistent scale tracking.
- Develops a LIKO module coupling foot-preintegration residuals with LiDAR point-to-plane residuals in an error-state Kalman filter.
- Demonstrates robust, drift-free odometry across public and long-term datasets on multiple quadruped platforms.
- Achieves low-latency computation (<20 ms) while maintaining tight sensor coupling in challenging environments.
Why it matters
Enables reliable autonomous navigation for legged robots in unstructured or dynamic environments where traditional sensor fusion fails, advancing real-world deployment of quadruped systems.
Abstract
Autonomous navigation for legged robots in com- plex and dynamic environments relies on robust simultaneous localization and mapping (SLAM) systems to accurately map surroundings and localize the robot, ensuring safe and efficient operation. While prior sensor fusion-based SLAM approaches have integrated various sensor modalities to improve their robustness, these algorithms are still susceptible to estimation drift in challenging environments due to their reliance on unsuitable fusion strategies. Therefore, we propose a robust LiDAR-visual-inertial-kinematic odometry system that inte- grates information from multiple sensors, such as a camera, LiDAR, inertial measurement unit (IMU), and joint encoders, for visual and LiDAR-based odometry estimation. Our system employs a fusion-based pose estimation approach that runs optimization-based visual-inertial-kinematic odometry (VIKO) and filter-based LiDAR-inertial-kinematic odometry (LIKO) based on measurement availability. In VIKO, we utilize the foot- preintegration technique and robust LiDAR-visual depth consis- tency using superpixel clusters in a sliding window optimization. In LIKO, we incorporate foot kinematics and employ a point-to- plane residual in an error-state iterative Kalman filter (ESIKF). Compared with other sensor fusion-based SLAM algorithms, our approach shows robust performance across public and long- term datasets.