CARLA-Round: A Multi-Factor Simulation Dataset for Roundabout Trajectory Prediction

TL;DR

CARLA-Round is a systematically designed simulation dataset for roundabout vehicle trajectory prediction, varying weather and traffic conditions to analyze their effects on prediction accuracy and facilitate model development.

Contribution

The paper introduces CARLA-Round, a novel structured simulation dataset for roundabout trajectory prediction with explicit annotations and controlled scenarios, enabling detailed analysis of environmental factors.

Findings

Traffic density significantly affects prediction difficulty.

Weather impacts are non-linear and less predictable.

The best model achieves 0.312m ADE on real-world data.

Abstract

Accurate trajectory prediction of vehicles at roundabouts is critical for reducing traffic accidents, yet it remains highly challenging due to their circular road geometry, continuous merging and yielding interactions, and absence of traffic signals. Developing accurate prediction algorithms relies on reliable, multimodal, and realistic datasets; however, such datasets for roundabout scenarios are scarce, as real-world data collection is often limited by incomplete observations and entangled factors that are difficult to isolate. We present CARLA-Round, a systematically designed simulation dataset for roundabout trajectory prediction. The dataset varies weather conditions (five types) and traffic density levels (spanning Level-of-Service A-E) in a structured manner, resulting in 25 controlled scenarios. Each scenario incorporates realistic mixtures of driving behaviors and provides explicit annotations that are largely absent from existing datasets. Unlike randomly sampled simulation data, this structured design enables precise analysis of how different conditions influence trajectory prediction performance. Validation experiments using standard baselines (LSTM, GCN, GRU+GCN) reveal traffic density dominates prediction difficulty with strong monotonic effects, while weather shows non-linear impacts. The best model achieves 0.312m ADE on real-world rounD dataset, demonstrating effective sim-to-real transfer. This systematic approach quantifies factor impacts impossible to isolate in confounded real-world datasets. Our CARLA-Round dataset is available at https://github.com/Rebecca689/CARLA-Round.