In-Vehicle Network CAN Data for Aggressive Driving Behavior and Cybersecurity Attack Detection

This dataset presents a comprehensive collection of Controller Area Network (CAN) bus data specifically designed for real-time detection of aggressive driving behaviors and cyber-attacks in in-vehicle networks (IVNs), addressing the critical need for unified safety and security solutions in modern automotive environments. The dataset was collected from 16 drivers aged 20-35 years (including 4 female drivers) across multiple vehicle models with CAN bus integration, generating data at approximately 260,000 bits per second. The data is categorized into five distinct behavioral classes: normal driving, aggressive braking, aggressive lane changes, aggressive acceleration, and aggressive situation injection through cyber-attacks, with the dataset being properly partitioned into training, validation, and testing subsets to ensure robust model development and evaluation.