SR-LIVO: LiDAR-Inertial-Visual Odometry and Mapping With Sweep Reconstruction

Existing LiDAR-inertial-visual odometry and mapping (LIV-OAM) systems mainly utilize the LiDAR-inertial odometry (LIO) module for structure reconstruction and the LiDAR-assisted visual-inertial odometry (VIO) module for color rendering. However, the performance of existing LiDAR-assisted VIO module doesn't match the accuracy delivered by LIO systems in the scenarios containing rich textures and geometric structures (i.e., without failure mode for both camera and LiDAR). This paper introduces SR-LIVO, an advanced and novel LIV-OAM system employing sweep reconstruction to align reconstructed sweeps with image timestamps. This allows the LIO module to accurately determine states at all imaging moments, enhancing pose accuracy and processing efficiency. Experimental results on two public datasets demonstrate that: 1) our SR-LIVO outperforms the existing state-of-the-art LIV-OAM systems in both pose accuracy, rendering performance and runtime efficiency; 2) In scenarios with rich textures and geometric structures, the LIO framework can provide more accurate pose than existing LiDAR-assisted VIO framework, and thus helps rendering. We have released our source code to contribute to the community development in this field.