VRUD: A Drone Dataset for Complex Vehicle-VRU Interactions within Mixed Traffic

2026-04-01

AI Summary

This work addresses the scarcity of high-quality interaction data between vehicles and vulnerable road users (VRUs) in dense, unstructured urban environments—a critical gap limiting safety advancements in Level 4 autonomous driving systems. To bridge this gap, the authors propose an efficient data collection framework leveraging drone-based aerial footage (4K resolution at 30 Hz) to capture complex traffic dynamics in Shenzhen’s “urban villages,” characterized by severe occlusions and unregulated mixed traffic. The resulting dataset, the first of its kind focused on such challenging conditions, comprises four hours of video, 11,479 VRU trajectories, 1,939 vehicle trajectories, and 4,002 multi-agent interaction scenarios. High-risk edge cases are automatically identified using Vector Time to Collision (VTTC), and all trajectories are aligned with OpenDRIVE-compliant high-definition maps to ensure spatial accuracy and usability for autonomous driving research.

Abstract

The Operational Design Domain (ODD) of urbanoriented Level 4 (L4) autonomous driving, especially for autonomous robotaxis, confronts formidable challenges in complex urban mixed traffic environments. These challenges stem mainly from the high density of Vulnerable Road Users (VRUs) and their highly uncertain and unpredictable interaction behaviors. However, existing open-source datasets predominantly focus on structured scenarios such as highways or regulated intersections, leaving a critical gap in data representing chaotic, unstructured urban environments. To address this, this paper proposes an efficient, high-precision method for constructing drone-based datasets and establishes the Vehicle-Vulnerable Road User Interaction Dataset (VRUD), as illustrated in Figure 1. Distinct from prior works, VRUD is collected from typical "Urban Villages" in Shenzhen, characterized by loose traffic supervision and extreme occlusion. The dataset comprises 4 hours of 4K/30Hz recording, containing 11,479 VRU trajectories and 1,939 vehicle trajectories. A key characteristic of VRUD is its composition: VRUs account for about 87% of all traffic participants, significantly exceeding the proportions in existing benchmarks. Furthermore, unlike datasets that only provide raw trajectories, we extracted 4,002 multi-agent interaction scenarios based on a novel Vector Time to Collision (VTTC) threshold, supported by standard OpenDRIVE HD maps. This study provides valuable, rare edge-case resources for enhancing the safety performance of ADS in complex, unstructured urban environments. To facilitate further research, we have made the VRUD dataset open-source at: https://zzi4.github.io/VRUD/.

Problem

• Vulnerable Road Users
• mixed traffic
• unstructured urban environment
• autonomous driving
• interaction behavior

Innovation

• drone-based dataset
• Vulnerable Road Users (VRUs)
• unstructured urban traffic
• Vector Time to Collision (VTTC)
• multi-agent interaction scenarios