Interaction of Autonomous and Manually Controlled Vehicles Multiscenario Vehicle Interaction Dataset

TL;DR

IAMCV presents a comprehensive, multimodal dataset focused on inter-vehicle interactions, recorded with a driver-centric vehicle across diverse German road scenes and enriched with LiDAR, cameras, GNSS/INS, and vehicle bus data. The dataset enables detailed analysis of vehicle dynamics and sensor fusion in real-world scenarios, and its utility is demonstrated through unsupervised trajectory segmentation and online camera calibration experiments. Key contributions include a robust sensor suite description, precise synchronization and calibration procedures, and two proof-of-concept applications that highlight practical uses beyond data collection. The work advances autonomous driving research by providing a rich resource for perception, localization, and interaction modeling, with planned future enhancements such as weather annotations and bounding-box labels.

Abstract

The acquisition and analysis of high-quality sensor data constitute an essential requirement in shaping the development of fully autonomous driving systems. This process is indispensable for enhancing road safety and ensuring the effectiveness of the technological advancements in the automotive industry. This study introduces the Interaction of Autonomous and Manually-Controlled Vehicles (IAMCV) dataset, a novel and extensive dataset focused on inter-vehicle interactions. The dataset, enriched with a sophisticated array of sensors such as Light Detection and Ranging, cameras, Inertial Measurement Unit/Global Positioning System, and vehicle bus data acquisition, provides a comprehensive representation of real-world driving scenarios that include roundabouts, intersections, country roads, and highways, recorded across diverse locations in Germany. Furthermore, the study shows the versatility of the IAMCV dataset through several proof-of-concept use cases. Firstly, an unsupervised trajectory clustering algorithm illustrates the dataset's capability in categorizing vehicle movements without the need for labeled training data. Secondly, we compare an online camera calibration method with the Robot Operating System-based standard, using images captured in the dataset. Finally, a preliminary test employing the YOLOv8 object-detection model is conducted, augmented by reflections on the transferability of object detection across various LIDAR resolutions. These use cases underscore the practical utility of the collected dataset, emphasizing its potential to advance research and innovation in the area of intelligent vehicles.

Figures (10)

• Figure 1: (a) depicts the location of the three sensors and the vertical associated with each sensor. (b) Shows the horizontal of the sensors. Lastly, (c) depicts the horizontal covered by the three cameras its-vehicle.
• Figure 2: The JKU-ITS vehicle was used for the collection of the data.
• Figure 3: Route followed from Linz (Austria) to Aachen (Germany).
• Figure 4: Directory tree depicting how the folders and files comprising the dataset are organized.
• Figure 5: Raw images acquired simultaneously:(a) left camera, (b) center camera, and (c) right camera.