Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2507.17089 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909836273254400 |
|---|---|
| author | Zhang, Shanshan Zhang, Qi Wang, Siyue Wen, Tianshui Wu, Liqin Zhou, Ziheng Hong, Xuemin Peng, Ao Zheng, Lingxiang Yang, Yu |
| author_facet | Zhang, Shanshan Zhang, Qi Wang, Siyue Wen, Tianshui Wu, Liqin Zhou, Ziheng Hong, Xuemin Peng, Ao Zheng, Lingxiang Yang, Yu |
| contents | Researchers have increasingly adopted Transformer-based models for inertial odometry. While Transformers excel at modeling long-range dependencies, their limited sensitivity to local, fine-grained motion variations and lack of inherent inductive biases often hinder localization accuracy and generalization. Recent studies have shown that incorporating large-kernel convolutions and Transformer-inspired architectural designs into CNN can effectively expand the receptive field, thereby improving global motion perception. Motivated by these insights, we propose a novel CNN-based module called the Dual-wing Adaptive Dynamic Mixer (DADM), which adaptively captures both global motion patterns and local, fine-grained motion features from dynamic inputs. This module dynamically generates selective weights based on the input, enabling efficient multi-scale feature aggregation. To further improve temporal modeling, we introduce the Spatio-Temporal Gating Unit (STGU), which selectively extracts representative and task-relevant motion features in the temporal domain. This unit addresses the limitations of temporal modeling observed in existing CNN approaches. Built upon DADM and STGU, we present a new CNN-based inertial odometry backbone, named Next Era of Inertial Odometry (IONext). Extensive experiments on six public datasets demonstrate that IONext consistently outperforms state-of-the-art (SOTA) Transformer- and CNN-based methods. For instance, on the RNIN dataset, IONext reduces the average ATE by 10% and the average RTE by 12% compared to the representative model iMOT. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_17089 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | IONext: Unlocking the Next Era of Inertial Odometry Zhang, Shanshan Zhang, Qi Wang, Siyue Wen, Tianshui Wu, Liqin Zhou, Ziheng Hong, Xuemin Peng, Ao Zheng, Lingxiang Yang, Yu Computer Vision and Pattern Recognition Robotics Researchers have increasingly adopted Transformer-based models for inertial odometry. While Transformers excel at modeling long-range dependencies, their limited sensitivity to local, fine-grained motion variations and lack of inherent inductive biases often hinder localization accuracy and generalization. Recent studies have shown that incorporating large-kernel convolutions and Transformer-inspired architectural designs into CNN can effectively expand the receptive field, thereby improving global motion perception. Motivated by these insights, we propose a novel CNN-based module called the Dual-wing Adaptive Dynamic Mixer (DADM), which adaptively captures both global motion patterns and local, fine-grained motion features from dynamic inputs. This module dynamically generates selective weights based on the input, enabling efficient multi-scale feature aggregation. To further improve temporal modeling, we introduce the Spatio-Temporal Gating Unit (STGU), which selectively extracts representative and task-relevant motion features in the temporal domain. This unit addresses the limitations of temporal modeling observed in existing CNN approaches. Built upon DADM and STGU, we present a new CNN-based inertial odometry backbone, named Next Era of Inertial Odometry (IONext). Extensive experiments on six public datasets demonstrate that IONext consistently outperforms state-of-the-art (SOTA) Transformer- and CNN-based methods. For instance, on the RNIN dataset, IONext reduces the average ATE by 10% and the average RTE by 12% compared to the representative model iMOT. |
| title | IONext: Unlocking the Next Era of Inertial Odometry |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2507.17089 |