Towards High-Performance Solid-State-LiDAR-Inertial Odometry and Mapping

Kailai Li,Meng Li,Uwe D. Hanebeck

DOI: https://doi.org/10.1109/LRA.2021.3070251

2021-04-28

Abstract:We present a novel tightly-coupled LiDAR-inertial odometry and mapping scheme for both solid-state and mechanical LiDARs. As frontend, a feature-based lightweight LiDAR odometry provides fast motion estimates for adaptive keyframe selection. As backend, a hierarchical keyframe-based sliding window optimization is performed through marginalization for directly fusing IMU and LiDAR measurements. For the Livox Horizon, a newly released solid-state LiDAR, a novel feature extraction method is proposed to handle its irregular scan pattern during preprocessing. LiLi-OM (Livox LiDAR-inertial odometry and mapping) is real-time capable and achieves superior accuracy over state-of-the-art systems for both LiDAR types on public data sets of mechanical LiDARs and in experiments using the Livox Horizon. Source code and recorded experimental data sets are available at <a class="link-external link-https" href="https://github.com/KIT-ISAS/lili-om" rel="external noopener nofollow">this https URL</a>.

Robotics

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is how to achieve high - performance Solid - State LiDAR - Inertial Odometry and Mapping. Specifically, the researchers aim to:

1. **Propose a tightly - coupled LiDAR - IMU fusion scheme**: By combining the data of solid - state LiDAR (such as Livox Horizon) and Inertial Measurement Unit (IMU), to improve the accuracy and real - time performance in the 6 - Degree - of - Freedom (6 - DoF) self - motion estimation of mobile platforms (such as robots or autonomous vehicles).

2. **Handle the scanning patterns unique to solid - state LiDAR**: Different from traditional mechanical LiDAR, solid - state LiDAR has irregular and non - repetitive scanning patterns. Therefore, new feature extraction methods need to be developed to adapt to this new type of sensor.

3. **Achieve real - time processing capabilities**: Ensure that the system can operate at a high frequency (for example, 10 Hz) in practical applications and remain efficient under limited computing resources.

4. **Improve robustness in complex environments**: Especially in scenarios of rapid motion or insufficient features, compensate for the distortion of LiDAR point clouds through IMU data, thereby improving the robustness and precision of the system.

### Main contributions

To achieve the above goals, the authors have made the following innovations:

- **Feature extraction method for solid - state LiDAR**: A new two - stage feature extraction algorithm has been designed, specifically for handling the irregular scanning pattern of Livox Horizon (see Section 3 for details).

- **Sliding - window optimization framework based on key - frames**: A hierarchical key - frame selection and optimization strategy has been proposed, directly fusing LiDAR and IMU measurements together for joint optimization (see Section 4 for details).

- **Open - source code and data set**: The complete source code and a new data set recorded using Livox Horizon and Xsens MTi - 670 IMU are provided, so that other researchers can reproduce and further improve the system.

These contributions make the system not only suitable for traditional mechanical LiDAR, but also specifically optimized for emerging solid - state LiDAR, providing a cost - effective solution for mobile perception tasks.