BEV-Patch-PF: Particle Filtering with BEV-Aerial Feature Matching for Off-Road Geo-Localization
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915847116685312 |
|---|---|
| author | Lee, Dongmyeong Quattrociocchi, Jesse Ellis, Christian Rana, Rwik Adkins, Amanda Uccello, Adam Warnell, Garrett Biswas, Joydeep |
| author_facet | Lee, Dongmyeong Quattrociocchi, Jesse Ellis, Christian Rana, Rwik Adkins, Amanda Uccello, Adam Warnell, Garrett Biswas, Joydeep |
| contents | We propose BEV-Patch-PF, a GPS-free sequential geo-localization system that integrates a particle filter with learned bird's-eye-view (BEV) and aerial feature maps. From onboard RGB and depth images, we construct a BEV feature map. For each 3-DoF particle pose hypothesis, we crop the corresponding patch from an aerial feature map computed from a local aerial image queried around the approximate location. BEV-Patch-PF computes a per-particle log-likelihood by matching the BEV feature to the aerial patch feature. On two real-world off-road datasets, our method achieves 9.7x lower absolute trajectory error (ATE) on seen routes and 6.6x lower ATE on unseen routes than a retrieval-based baseline, while maintaining accuracy under dense canopy and shadow. The system runs in real time at 10 Hz on an NVIDIA Tesla T4, enabling practical robot deployment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_15111 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | BEV-Patch-PF: Particle Filtering with BEV-Aerial Feature Matching for Off-Road Geo-Localization Lee, Dongmyeong Quattrociocchi, Jesse Ellis, Christian Rana, Rwik Adkins, Amanda Uccello, Adam Warnell, Garrett Biswas, Joydeep Robotics Computer Vision and Pattern Recognition We propose BEV-Patch-PF, a GPS-free sequential geo-localization system that integrates a particle filter with learned bird's-eye-view (BEV) and aerial feature maps. From onboard RGB and depth images, we construct a BEV feature map. For each 3-DoF particle pose hypothesis, we crop the corresponding patch from an aerial feature map computed from a local aerial image queried around the approximate location. BEV-Patch-PF computes a per-particle log-likelihood by matching the BEV feature to the aerial patch feature. On two real-world off-road datasets, our method achieves 9.7x lower absolute trajectory error (ATE) on seen routes and 6.6x lower ATE on unseen routes than a retrieval-based baseline, while maintaining accuracy under dense canopy and shadow. The system runs in real time at 10 Hz on an NVIDIA Tesla T4, enabling practical robot deployment. |
| title | BEV-Patch-PF: Particle Filtering with BEV-Aerial Feature Matching for Off-Road Geo-Localization |
| topic | Robotics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.15111 |