UL-VIO: Ultra-lightweight Visual-Inertial Odometry with Noise Robust Test-time Adaptation
This work addresses deployment challenges for autonomous robots on limited hardware, offering a lightweight and adaptable solution, though it is incremental in combining model compression with test-time adaptation.
The paper tackles the problem of deploying visual-inertial odometry (VIO) on resource-constrained devices by proposing UL-VIO, an ultra-lightweight network (<1M) that achieves a 36X smaller size than state-of-the-art with only a 1% increase in error on the KITTI dataset, and it introduces noise-robust test-time adaptation to handle environmental shifts.
Data-driven visual-inertial odometry (VIO) has received highlights for its performance since VIOs are a crucial compartment in autonomous robots. However, their deployment on resource-constrained devices is non-trivial since large network parameters should be accommodated in the device memory. Furthermore, these networks may risk failure post-deployment due to environmental distribution shifts at test time. In light of this, we propose UL-VIO -- an ultra-lightweight (<1M) VIO network capable of test-time adaptation (TTA) based on visual-inertial consistency. Specifically, we perform model compression to the network while preserving the low-level encoder part, including all BatchNorm parameters for resource-efficient test-time adaptation. It achieves 36X smaller network size than state-of-the-art with a minute increase in error -- 1% on the KITTI dataset. For test-time adaptation, we propose to use the inertia-referred network outputs as pseudo labels and update the BatchNorm parameter for lightweight yet effective adaptation. To the best of our knowledge, this is the first work to perform noise-robust TTA on VIO. Experimental results on the KITTI, EuRoC, and Marulan datasets demonstrate the effectiveness of our resource-efficient adaptation method under diverse TTA scenarios with dynamic domain shifts.