LPVIMO-SAM: Tightly-coupled LiDAR/Polarization Vision/Inertial/Magnetometer/Optical Flow Odometry via Smoothing and Mapping

TL;DR

The authors address robust state estimation and mapping in LiDAR-degraded and low-texture environments by proposing LPVIMO-SAM, a tightly-coupled system that fuses LiDAR, polarization vision, IMU, magnetometer, and optical flow within a factor-graph optimization framework. The architecture comprises two subsystems: PVIS, which leverages polarization cues to strengthen visual feature extraction, and LIMOS, which uses LiDAR features plus magnetometer heading and optical-flow-derived velocity and height factors to reduce drift. Key contributions include magnetometer heading priors, optical-flow velocity/height factors, and polarization-based feature augmentation integrated into iSAM2 optimization, along with extensive outdoor experiments against multiple baselines. Results show improved accuracy and robustness in various challenging scenarios, including LiDAR-degraded and low-texture environments, with LPVIMO-SAM significantly outperforming several existing LVIS/SLAM systems in several metrics. The work demonstrates practical impact for robust navigation in complex real-world settings where single-sensor or standard LVIS systems struggle.

Abstract

We propose a tightly-coupled LiDAR/Polarization Vision/Inertial/Magnetometer/Optical Flow Odometry via Smoothing and Mapping (LPVIMO-SAM) framework, which integrates LiDAR, polarization vision, inertial measurement unit, magnetometer, and optical flow in a tightly-coupled fusion. This framework enables high-precision and highly robust real-time state estimation and map construction in challenging environments, such as LiDAR-degraded, low-texture regions, and feature-scarce areas. The LPVIMO-SAM comprises two subsystems: a Polarized Vision-Inertial System and a LiDAR/Inertial/Magnetometer/Optical Flow System. The polarized vision enhances the robustness of the Visual/Inertial odometry in low-feature and low-texture scenarios by extracting the polarization information of the scene. The magnetometer acquires the heading angle, and the optical flow obtains the speed and height to reduce the accumulated error. A magnetometer heading prior factor, an optical flow speed observation factor, and a height observation factor are designed to eliminate the cumulative errors of the LiDAR/Inertial odometry through factor graph optimization. Meanwhile, the LPVIMO-SAM can maintain stable positioning even when one of the two subsystems fails, further expanding its applicability in LiDAR-degraded, low-texture, and low-feature environments. Code is available on https://github.com/junxiaofanchen/LPVIMO-SAM.

Figures (6)

• Figure 1: Flow chart of the LPVIMO-SAM system.
• Figure 2: Left: Polarized image sensor PHX050S-PC, Right: Pixel distribution map.
• Figure 3: The factor graph framework of the LiDAR/Polari-zation Vision/Inertial/Magnetometer/Optical Flow navigation system (adaptive modified from shan2021lvisam).
• Figure 4: Trajectory comparison charts for each scene.
• Figure 5: Comparison chart of the estimation error of the three-dimensional distance and the errors of the trajectory in each direction in scene d, where the 3D distance is $\sqrt{\mathrm{p}_{x}^{2}+\mathrm{p}_{y}^{2}+\mathrm{p}_{z}^{2}}$.