Event-Only Drone Trajectory Forecasting with RPM-Modulated Kalman Filtering
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910038059122688 |
|---|---|
| author | M., Hari Prasanth S. Habibiroudkenar, Pejman Alamikkotervo, Eerik Bouzoulas, Dimitrios Ojala, Risto |
| author_facet | M., Hari Prasanth S. Habibiroudkenar, Pejman Alamikkotervo, Eerik Bouzoulas, Dimitrios Ojala, Risto |
| contents | Event cameras provide high-temporal-resolution visual sensing that is well suited for observing fast-moving aerial objects; however, their use for drone trajectory prediction remains limited. This work introduces an event-only drone forecasting method that exploits propeller-induced motion cues. Propeller rotational speed are extracted directly from raw event data and fused within an RPM-aware Kalman filtering framework. Evaluations on the FRED dataset show that the proposed method outperforms learning-based approaches and vanilla kalman filter in terms of average distance error and final distance error at 0.4s and 0.8s forecasting horizons. The results demonstrate robust and accurate short- and medium-horizon trajectory forecasting without reliance on RGB imagery or training data. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_01997 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Event-Only Drone Trajectory Forecasting with RPM-Modulated Kalman Filtering M., Hari Prasanth S. Habibiroudkenar, Pejman Alamikkotervo, Eerik Bouzoulas, Dimitrios Ojala, Risto Computer Vision and Pattern Recognition Robotics Image and Video Processing Event cameras provide high-temporal-resolution visual sensing that is well suited for observing fast-moving aerial objects; however, their use for drone trajectory prediction remains limited. This work introduces an event-only drone forecasting method that exploits propeller-induced motion cues. Propeller rotational speed are extracted directly from raw event data and fused within an RPM-aware Kalman filtering framework. Evaluations on the FRED dataset show that the proposed method outperforms learning-based approaches and vanilla kalman filter in terms of average distance error and final distance error at 0.4s and 0.8s forecasting horizons. The results demonstrate robust and accurate short- and medium-horizon trajectory forecasting without reliance on RGB imagery or training data. |
| title | Event-Only Drone Trajectory Forecasting with RPM-Modulated Kalman Filtering |
| topic | Computer Vision and Pattern Recognition Robotics Image and Video Processing |
| url | https://arxiv.org/abs/2603.01997 |