发表机构
University of Michigan（密歇根大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
GINIO提出SO(3)-等变接口用于神经惯性里程计，通过末帧对齐和谱协方差实现任意IMU旋转下的张量一致测量，显著提升ATE精度与鲁棒性。
AI 中文摘要
神经惯性里程计日益将网络用作滤波管道内的学习测量。此类测量应在任意IMU安装约定下一致变换：其均值必须作为向量变换，其协方差必须作为二阶张量相合变换。我们提出GINIO，一种用于在IMU测量帧任意旋转下进行神经惯性里程计的几何SO(3)-等变接口。给定校准的IMU窗口，我们的框架预测遵循这些张量定律的运动测量和不确定性。为支持高效的传感器帧学习，我们引入末帧对齐（LFA），一种确定性预处理步骤，对于SO(3)-等变预测器，其在理论上等价于世界帧训练。连接的估计器跟踪传感器局部状态（如IMU偏置），将干扰估计与学习测量所施加的几何定律分离。我们在滤波器连接的NIO、AirIO式循环空中预测、EqNIO式全SO(3)规范化以及ResNet式时间骨干中实例化相同接口。在TLIO上，GINIO实现2.018 m的ID/SO(3) ATE，而EqNIO退化至76.389 m，且使用11.6倍更少的FLOPs。在NanoBench上，我们的AirIO式实例化在无外部姿态输入下将ATE从5.579 m提升至1.430 m，我们的ResNet式实例化达到0.581 m ATE，而ResNet1D为0.645 m。在Fetch上，GINIO经验性地将未见物理重装的ATE从8.15 m降至0.50 m而无需重新训练，展示了超出精确坐标帧保证的鲁棒性。对于不确定性，谱协方差将协方差等变误差相比对角头部降低了三个数量级以上。
英文摘要
Neural inertial odometry increasingly uses networks as learned measurements inside filtering pipelines. Such measurements should transform consistently under arbitrary IMU mounting conventions: their mean must transform as a vector, and their covariance must transform congruently as a second-order tensor. We present GINIO, a geometric SO(3)-equivariant interface for neural inertial odometry under arbitrary rotations of the IMU measurement frame. Given calibrated IMU windows, our framework predicts a motion measurement and uncertainty obeying these tensorial laws. To support efficient sensor-frame learning, we introduce Last-Frame Alignment (LFA), a deterministic preprocessing step that is provably equivalent to world-frame training for SO(3)-equivariant predictors. The connected estimator tracks sensor-local states such as IMU bias, separating nuisance estimation from the geometric law enforced by the learned measurement. We instantiate the same interface in filter-connected NIO, AirIO-style recurrent aerial prediction, EqNIO-style full-SO(3) canonicalization, and ResNet-style temporal backbones. On TLIO, GINIO achieves 2.018 m ID/SO(3) ATE while EqNIO degrades to 76.389 m, using 11.6x fewer FLOPs. On NanoBench, our AirIO-style instantiation improves ATE from 5.579 m to 1.430 m without external attitude input, and our ResNet-style instantiation reaches 0.581 m ATE versus 0.645 m for ResNet1D. On Fetch, GINIO empirically reduces unseen physical-remount ATE from 8.15 m to 0.50 m without retraining, demonstrating robustness beyond the exact coordinate-frame guarantee. For uncertainty, spectral covariance reduces covariance-equivariance error by over three orders of magnitude compared with a diagonal head.
CommentsAccepted at the 10th Conference on Robot Learning (CoRL 2026). 28 pages, 14 figures
