There is a newer version of the record available.

Published April 20, 2026 | Version v1
Dataset Open

AutoHack: A Physically Verified Multi-Bus CAN Dataset for Intrusion Detection System Evaluation

Authors/Creators

  • 1. ROR icon Korea University
  • 2. ROR icon Kookmin University

Description

This dataset is designed for evaluating Intrusion Detection Systems (IDS) in automotive networks. It features physically verified CAN traffic collected from multiple buses (C-CAN, P-CAN, and B-CAN). The dataset includes various attack scenarios such as Fuzzing, Spoofing, Replay, DoS, and UDS-based attacks, all executed in a real vehicle environment to ensure high fidelity.

This dataset is provided as an appendix for the paper submitted to USENIX VehicleSec 2026.

Files

AutoHack-Benchmark.zip

Files (309.1 MB)

Name Size
md5:06444846f024d72d6028aa8c4c5f2474
29.6 kB Preview Download
md5:78193f5f4927d7799952d35044657dc2
309.1 MB Preview Download

Additional details

Software

Programming language
Python