GenZ-LIO: Generalizable LiDAR-Inertial Odometry Beyond Indoor--Outdoor Boundaries

Abstract

Light detection and ranging (LiDAR)-inertial odometry (LIO) enables accurate localization and mapping for autonomous navigation in various scenes. However, its performance remains sensitive to variations in spatial scale, which refers to the spatial extent of the scene reflected in the distribution of point ranges in a LiDAR scan. Transitions between confined indoor and expansive outdoor spaces induce substantial variations in point density, which may reduce robustness and computational efficiency. To address this issue, we propose GenZ-LIO, a LIO framework generalizable across both indoor and outdoor environments. GenZ-LIO comprises three key components. First, inspired by the principle of the proportional-integral-derivative (PID) controller, it adaptively regulates the voxel size for downsampling via feedback control, driving the voxelized point count toward a scale-informed setpoint while enabling stable and efficient processing across varying scene scales. Second, we formulate a hybrid-metric state update that jointly leverages point-to-plane and point-to-point residuals to mitigate LiDAR degeneracy arising from directionally insufficient geometric constraints. Third, to alleviate the computational burden introduced by point-to-point matching, we introduce a voxel-pruned correspondence search strategy that discards non-promising voxel candidates and reduces unnecessary computations. Experimental results demonstrate that GenZ-LIO achieves robust odometry estimation and improved computational efficiency across confined indoor, open outdoor, and transitional environments. Our code will be made publicly available upon publication.

Figures (11)

• Figure 1: Estimated trajectory and mapping result of our GenZ-LIO on the [1]Handheld-A-01 sequence of our [1]NarrowWide dataset, introduced in this work, where the platform traverses environments with substantially different spatial scales. The trajectory is color-coded by the proposed scale indicator $\bar{m}_t$, which reflects the spatial extent of the surrounding scene. By incorporating this indicator, GenZ-LIO adapts to varying scene scales, enabling consistent odometry estimation while maintaining computational efficiency across both confined indoor and open outdoor environments.
• Figure 2: Comparison of adaptive voxelization strategies. (a) LOCUS 2.0 reinke2022ral updates the voxel size for next scan, denoted by $d_{t+1}$, using the ratio of current voxelized point count $N_t$ to the fixed desired point count $N_\mathrm{desired}^\mathrm{fixed}$. (b) AdaLIO lim2023ur adaptively selects between pre-defined coarse voxel size $d_\mathrm{coarse}^\mathrm{fixed}$ and the fine voxel size $d_\mathrm{fine}^\mathrm{fixed}$ by checking whether the coarse voxelization yields fewer points than a fixed threshold $\tau_N^\mathrm{fixed}$. (c) LIVOX-CAM cheng2025ral adjusts the voxel size $d_t$ by first performing a temporary voxelization with a fixed voxel size $d_\mathrm{temp}^\mathrm{fixed}$ and then updating it based on the ratio between the temporary point count $N_{\mathrm{temp},t}$ and desired point count $N_\mathrm{desired}^\mathrm{fixed}$. This update follows a volume-based scaling strategy rather than a linear scaling in LOCUS 2.0 reinke2022ral. (d) Our method adaptively computes the desired point count $N_{\mathrm{desired},t}$ based on the scale indicator $\bar{m}_t$ and determines the corresponding voxel size $d_t$ via proportional-derivative (PD) control with sensitivity-informed gain scheduling. This process corresponds to Fig. \ref{['fig:genz-lio_flowchart']}(b).
• Figure 3: System overview of GenZ-LIO. (a) In the preprocessing stage, forward propagation uses IMU measurements to propagate the state and covariance, and backward propagation removes motion distortion from the LiDAR scan, yielding the deskewed scan $\mathcal{S}_t$. (b) The scan $\mathcal{S}_t$ is voxelized using the voxel size of previous timestep to produce the temporary voxelized scan $\mathcal{V}_{\mathrm{temp},t}$. The median range $m_t$ of points in $\mathcal{V}_{\mathrm{temp},t}$ is inserted into a sliding window to compute the scene scale indicator $\bar{m}_t$. Based on $\bar{m}_t$, a target number of voxelized points is set as a scale-informed control setpoint, and the voxel size $d_t$ is adaptively adjusted via a PD controller with gain scheduling. The updated $d_t$ is then used for bi-resolution voxelization of $\mathcal{S}_t$, yielding $\mathcal{V}_{\mathrm{merge},t}$ with $d_t/2$ for map integration with reduced discretization error and $\mathcal{V}_t$ with $d_t$ for state update. (c) The voxelized scan $\mathcal{V}_t$ is aligned with the voxel map for a voxel-pruned correspondence search, which avoids unnecessary traversal of neighboring voxels. This process produces the point-to-plane correspondence set $\mathcal{C}_\mathrm{pl}$ and point-to-point correspondence set $\mathcal{C}_\mathrm{po}$, which are used in the hybrid-metric lee2024ral state update. Finally, the transformed $\mathcal{V}_{\mathrm{merge},t}$ is integrated into the voxel map. For clarity, the control flow of the scale-aware adaptive voxelization is further illustrated in Fig. \ref{['fig:adap_vox_diagram_comparison']}(d), where the bi-resolution voxelization step is omitted for comparison with other adaptive voxelization strategies.
• Figure 4: Our scale indicator $\bar{m}_t$ mapped onto the estimated trajectory for the [1]Corridor 02 sequence of the [1]SuperLoczhao2025icra dataset, demonstrating its variation with the scene’s spatial scale.
• Figure 5: Candidate voxel selection based on the region occupied by a query point within its corresponding root voxel. The root voxel is divided into 27 regions, and the occupied region falls into one of four cases: (a) center case, selecting no neighboring voxel; (b) surface case, selecting one surface-sharing neighboring voxel; (c) edge case, selecting three edge-sharing neighboring voxels; and (d) corner case, selecting seven corner-sharing neighboring voxels. The selected neighboring voxels, together with the root voxel, are considered candidate voxels for correspondence search. In Algorithm \ref{['alg: pruning strategy']}, the function $\texttt{GetCandidateVoxels}$ selects the candidate voxels based on these sharing relations.