Beyond Features: How Dataset Design Influences Multi-Agent Trajectory Prediction Performance
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913930128916480 |
|---|---|
| author | Demmler, Tobias Häringer, Jakob Tamke, Andreas Dang, Thao Hegai, Alexander Mikelsons, Lars |
| author_facet | Demmler, Tobias Häringer, Jakob Tamke, Andreas Dang, Thao Hegai, Alexander Mikelsons, Lars |
| contents | Accurate trajectory prediction is critical for safe autonomous navigation, yet the impact of dataset design on model performance remains understudied. This work systematically examines how feature selection, cross-dataset transfer, and geographic diversity influence trajectory prediction accuracy in multi-agent settings. We evaluate a state-of-the-art model using our novel L4 Motion Forecasting dataset based on our own data recordings in Germany and the US. This includes enhanced map and agent features. We compare our dataset to the US-centric Argoverse 2 benchmark. First, we find that incorporating supplementary map and agent features unique to our dataset, yields no measurable improvement over baseline features, demonstrating that modern architectures do not need extensive feature sets for optimal performance. The limited features of public datasets are sufficient to capture convoluted interactions without added complexity. Second, we perform cross-dataset experiments to evaluate how effective domain knowledge can be transferred between datasets. Third, we group our dataset by country and check the knowledge transfer between different driving cultures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_05098 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Beyond Features: How Dataset Design Influences Multi-Agent Trajectory Prediction Performance Demmler, Tobias Häringer, Jakob Tamke, Andreas Dang, Thao Hegai, Alexander Mikelsons, Lars Robotics Artificial Intelligence Machine Learning Accurate trajectory prediction is critical for safe autonomous navigation, yet the impact of dataset design on model performance remains understudied. This work systematically examines how feature selection, cross-dataset transfer, and geographic diversity influence trajectory prediction accuracy in multi-agent settings. We evaluate a state-of-the-art model using our novel L4 Motion Forecasting dataset based on our own data recordings in Germany and the US. This includes enhanced map and agent features. We compare our dataset to the US-centric Argoverse 2 benchmark. First, we find that incorporating supplementary map and agent features unique to our dataset, yields no measurable improvement over baseline features, demonstrating that modern architectures do not need extensive feature sets for optimal performance. The limited features of public datasets are sufficient to capture convoluted interactions without added complexity. Second, we perform cross-dataset experiments to evaluate how effective domain knowledge can be transferred between datasets. Third, we group our dataset by country and check the knowledge transfer between different driving cultures. |
| title | Beyond Features: How Dataset Design Influences Multi-Agent Trajectory Prediction Performance |
| topic | Robotics Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2507.05098 |