Neural LiDAR Odometry With Feature Association and Reuse for Unstructured Environments
文献类型：期刊论文
| 作者 | Qian, Liangshu2,3; Li, Wei1,3; Hu, Yu1,3 |
| 刊名 | JOURNAL OF FIELD ROBOTICS
 |
| 出版日期 | 2025-06-16 |
| 页码 | 18 |
| 关键词 | deep learning LiDAR odometry unstructured environments |
| ISSN号 | 1556-4959 |
| DOI | 10.1002/rob.22607 |
| 英文摘要 | Odometry plays a crucial role in autonomous tasks of field robots, providing accurate position and orientation derived from sequential sensor observations. Odometry based on Light Detection and Ranging (LiDAR) sensors has demonstrated widespread applicability in environments with rich structured features, such as urban and indoor settings. However, for unstructured environments like scrubland and rural roads, the extraction, description, and correct matching of LiDAR features between frames become challenging. Due to the lack of flat surfaces and straight lines, the existing odometry approaches, whether using hand-crafted features such as edge and planar points or learned features through networks, will face the problem of decreased positioning accuracy and potential failure. Therefore, we propose a neural LiDAR odometry based on Trans-frame Association to extract more effective features for pose estimation in unstructured environments. The Trans-frame Association module contains a fully interactive frame Transformer and a scan-aware Swin Transformer. The former applies cross-attention to features extracted from two consecutive frames, thus enhancing the accuracy and robustness of feature correspondences by considering the contextual information. The latter restricts the attention mechanism to shift along the scan lines of LiDAR, thereby leveraging the sensor's inherent higher horizontal resolution. Our Transformer has linear complexity, which guarantees the module can meet real-time requirements. Additionally, we design a Reuse Refinement Pyramid architecture to further improve the accuracy of pose estimation by reusing multiresolution features. We conducted extensive experiments on the RELLIS-3D data set and our Matian Ridge data set collected in a representative unstructured scene. The results demonstrate that our network outperforms recent learning-based LiDAR odometry methods in terms of accuracy. The code is available at . |
| 资助项目 | This study was supported by Beijing Natural Science Foundation (Grant No. L243008), and in part by National Natural Science Foundation of China (Grant Nos. 62003323 and 62176250).[L243008] ; Beijing Natural Science Foundation[62003323] ; Beijing Natural Science Foundation[62176250] ; National Natural Science Foundation of China |
| WOS研究方向 | Robotics |
| 语种 | 英语 |
| WOS记录号 | WOS:001508682300001 |
| 出版者 | WILEY |
| 源URL | [http://119.78.100.204/handle/2XEOYT63/42367]  |
| 专题 | 中国科学院计算技术研究所期刊论文_英文 |
| 通讯作者 | Li, Wei; Hu, Yu |
| 作者单位 | 1.Univ Chinese Acad Sci, Sch Comp Sci & Technol, Beijing, Peoples R China 2.Univ Chinese Acad Sci, Hangzhou Inst Adv Study, Hangzhou, Peoples R China 3.Chinese Acad Sci, Res Ctr Intelligent Comp Syst, Inst Comp Technol, Beijing, Peoples R China |
| 推荐引用方式 GB/T 7714 | Qian, Liangshu,Li, Wei,Hu, Yu. Neural LiDAR Odometry With Feature Association and Reuse for Unstructured Environments[J]. JOURNAL OF FIELD ROBOTICS,2025:18. |
| APA | Qian, Liangshu,Li, Wei,&Hu, Yu.(2025).Neural LiDAR Odometry With Feature Association and Reuse for Unstructured Environments.JOURNAL OF FIELD ROBOTICS,18. |
| MLA | Qian, Liangshu,et al."Neural LiDAR Odometry With Feature Association and Reuse for Unstructured Environments".JOURNAL OF FIELD ROBOTICS (2025):18. |
入库方式： OAI收割
来源：计算技术研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。