UAVCAN Dataset Description

TL;DR

This paper presents a public UAVCAN intrusion dataset collected on a Pixhawk4-based testbed to study security threats in unmanned vehicle CAN networks. It focuses on three injection attacks—Flooding, Fuzzy, and Replay—executed across 10 attack scenarios to emulate real-world adversarial conditions. By capturing normal and attack frames on UAVCAN within the internal CAN networks, the dataset enables evaluation of anomaly detection and intrusion-detection approaches. The work targets practical impact by providing labeled data, scenario design guidelines, and metadata to support research on UAVCAN security and defense mechanisms.

Abstract

We collected attack data from unmanned vehicles using the UAVCAN protocol, and public and described technical documents. A testbed was built with a drone using PX4, and a total of three attacks, Flooding, Fuzzy, and Replay, were performed. The attack was carried out in a total of 10 scenarios. We expect that the attack data will help develop technologies such as anomaly detection to solve the security threat problem of drones.

Figures (27)

• Figure 1: Security Threats in UAV khan2020uav
• Figure 2: Network Protocol Process of UAV
• Figure 3: UAVCAN Message Frame Description uav_msg
• Figure 4: UAVCAN Anonymous Message Description uav_msg
• Figure 5: UAVCAN Service Frame Description uav_msg