Beyond Features: How Dataset Design Influences Multi-Agent Trajectory Prediction Performance

TL;DR

This study investigates how dataset design, feature selection, and geographic diversity affect multi-agent trajectory prediction accuracy, revealing that extensive features may not be necessary and that domain transfer varies across datasets and cultures.

Contribution

The paper provides a systematic analysis of dataset design choices and their impact on trajectory prediction, introducing a new dataset and evaluating feature importance and transferability.

Findings

Supplementary features do not significantly improve performance.

Limited features in public datasets suffice for complex interactions.

Domain transfer effectiveness varies across datasets and cultures.

Abstract

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.