Zhang, Zhihe (Lanzhou University), Nie, Hongtao (Lanzhou University), Zhang, Yichi (Lanzhou University), Wei, Hao (Lanzhou university), Zhao, Rui (LanZhou University), Zhi, Peng (Lanzhou University), Zhou, Qingguo (Lanzhou University), Li, Yan (Lanzhou University)

MSDAD: A Multi-Sensor Dataset for Autonomous Driving

Scheduled for presentation during the Poster Session "Validation, simulation, and virtual testing I" (WeAT14), Wednesday, September 25, 2024, 10:30−12:30, Foyer

2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), September 24- 27, 2024, Edmonton, Canada

This information is tentative and subject to change. Compiled on December 26, 2024

Abstract

The development of autonomous driving relies significantly on high-quality datasets. Currently, there is a pressing need for abundant, diverse, and realistic datasets to drive advancements in autonomous driving technology. In this paper, we present a multi-sensor dataset specifically designed for autonomous driving. It incorporates advanced sensors such as a 128-beam Lidar, solid state Lidar, and event cameras. It not only provides precise trajectory ground truth for localization tasks but also offers real-time driver behavior for research such as reinforcement learning. To ensure the dataset reflects the real world, we took into account variations in lighting conditions, weather. This deliberate inclusion aims to provide realistic scenarios for researchers to comprehensively evaluate and optimize the performance of autonomous driving algorithms. Through this dataset, researchers can delve deeper into the challenges faced by autonomous driving technology, thereby driving innovation and progress in the field.