Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2503.06565 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913727170740224 |
|---|---|
| author | Li, Shijie Xu, Xun Yeo, Si Yong Yang, Xulei |
| author_facet | Li, Shijie Xu, Xun Yeo, Si Yong Yang, Xulei |
| contents | Motion forecasting is a crucial component of autonomous driving systems, enabling the generation of accurate and smooth future trajectories to ensure safe navigation to the destination. In previous methods, potential future trajectories are often absent in the scene encoding stage, which may lead to suboptimal outcomes. Additionally, prior approaches typically employ transformer architectures for spatiotemporal modeling of trajectories and map information, which suffer from the quadratic scaling complexity of the transformer architecture. In this work, we propose an interaction-based method, named Future-Aware Interaction Network, that introduces potential future trajectories into scene encoding for a comprehensive traffic representation. Furthermore, a State Space Model (SSM), specifically Mamba, is introduced for both spatial and temporal modeling. To adapt Mamba for spatial interaction modeling, we propose an adaptive reordering strategy that transforms unordered data into a structured sequence. Additionally, Mamba is employed to refine generated future trajectories temporally, ensuring more consistent predictions. These enhancements not only improve model efficiency but also enhance the accuracy and diversity of predictions. We conduct comprehensive experiments on the widely used Argoverse 1 and Argoverse 2 datasets, demonstrating that the proposed method achieves superior performance compared to previous approaches in a more efficient way. The code will be released according to the acceptance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_06565 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Future-Aware Interaction Network For Motion Forecasting Li, Shijie Xu, Xun Yeo, Si Yong Yang, Xulei Computer Vision and Pattern Recognition Motion forecasting is a crucial component of autonomous driving systems, enabling the generation of accurate and smooth future trajectories to ensure safe navigation to the destination. In previous methods, potential future trajectories are often absent in the scene encoding stage, which may lead to suboptimal outcomes. Additionally, prior approaches typically employ transformer architectures for spatiotemporal modeling of trajectories and map information, which suffer from the quadratic scaling complexity of the transformer architecture. In this work, we propose an interaction-based method, named Future-Aware Interaction Network, that introduces potential future trajectories into scene encoding for a comprehensive traffic representation. Furthermore, a State Space Model (SSM), specifically Mamba, is introduced for both spatial and temporal modeling. To adapt Mamba for spatial interaction modeling, we propose an adaptive reordering strategy that transforms unordered data into a structured sequence. Additionally, Mamba is employed to refine generated future trajectories temporally, ensuring more consistent predictions. These enhancements not only improve model efficiency but also enhance the accuracy and diversity of predictions. We conduct comprehensive experiments on the widely used Argoverse 1 and Argoverse 2 datasets, demonstrating that the proposed method achieves superior performance compared to previous approaches in a more efficient way. The code will be released according to the acceptance. |
| title | Future-Aware Interaction Network For Motion Forecasting |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.06565 |