AdaLIO: Robust Adaptive LiDAR-Inertial Odometry in Degenerate Indoor Environments

Abstract: In recent years, the demand for mapping construction sites or buildings using light detection and ranging~(LiDAR) sensors has been increased to model environments for efficient site management. However, it is observed that sometimes LiDAR-based approaches diverge in narrow and confined environments, such as spiral stairs and corridors, caused by fixed parameters regardless of the changes in the environments. That is, the parameters of LiDAR (-inertial) odometry are mostly set for open space; thus, if the same parameters suitable for the open space are applied in a corridor-like scene, it results in divergence of odometry methods, which is referred to as \textit{degeneracy}. To tackle this degeneracy problem, we propose a robust LiDAR inertial odometry called \textit{AdaLIO}, which employs an adaptive parameter setting strategy. To this end, we first check the degeneracy by checking whether the surroundings are corridor-like environments. If so, the parameters relevant to voxelization and normal vector estimation are adaptively changed to increase the number of correspondences. As verified in a public dataset, our proposed method showed promising performance in narrow and cramped environments, avoiding the degeneracy problem.

• The paper introduces an adaptive parameter strategy that dynamically adjusts voxel size and normal estimation to enhance pose estimation in constrained indoor spaces.
• It demonstrates superior performance over methods like Faster-LIO on the HILTI-Oxford dataset, especially in degenerate conditions.
• The research paves the way for robust SLAM in challenging indoor environments, potentially benefiting construction mapping and robotic navigation.

Overview of AdaLIO: A Robust Adaptive LiDAR-Inertial Odometry

The paper presents AdaLIO, an innovative approach to LiDAR-inertial odometry that aims to address the challenges posed by degenerate indoor environments, such as narrow corridors and spiral staircases. Traditional LiDAR-based odometry systems often struggle in confined spaces due to fixed parameter settings that fail to adapt to varying environmental conditions. AdaLIO introduces an adaptive parameter setting strategy designed to maintain effective correspondence estimation in narrow and cramped settings, effectively mitigating the degeneracy that leads to pose estimation divergence.

Methodological Innovations

AdaLIO's core contribution lies in its adaptive parameter setting strategy. The method dynamically adjusts parameters related to voxelization and normal vector estimation. This adaptive mechanism increases the number of correspondences in environments identified as degenerate, thus enhancing the stability and accuracy of ego-motion estimation. Specifically, the strategy involves changing the voxelization size and parameters for normal vector estimation based on the detected environmental conditions. These modifications aim to optimize the data association process, preserving correspondence quality even in environments with reduced spatial volume.

Experimental Validation

The efficacy of AdaLIO is validated using the HILTI-Oxford dataset. Through both qualitative and quantitative evaluations, AdaLIO demonstrates superior performance over state-of-the-art methods like Faster-LIO, particularly in scenarios involving narrow and cramped indoor environments. In Exp11, for example, AdaLIO successfully navigated a tight spiral staircase that caused Faster-LIO to diverge. The adaptive adjustments in parameter settings allowed AdaLIO to maintain a stable estimation of poses in challenging settings where traditional fixed-parameter approaches failed.

Quantitatively, AdaLIO reported higher scores in the HILTI-Oxford SLAM Challenge’s validation set, particularly for sequences involving narrow spaces. Despite similar performances in open areas, AdaLIO's adaptive capabilities provided significant improvements within degenerate conditions, proving its robustness and reliability in such scenarios.

Implications and Future Directions

AdaLIO offers a significant advancement for robotic mapping and navigation systems operating in unpredictable indoor environments. By effectively tackling degeneracy through adaptive parameterization, it paves the way for more robust SLAM systems. This approach could be particularly beneficial for applications involving construction site mapping and indoor navigation, where environmental conditions can vary drastically.

Future work could expand upon AdaLIO by integrating additional sensor modalities, such as camera inputs or ultra-wideband sensors, to further enhance robustness and accuracy in even more challenging environments. The development of a degeneracy-robust SLAM framework incorporating AdaLIO principles could also provide a comprehensive solution, fostering advancements in various fields of autonomous robotics.