The cyberattack detection strategies are essential for autonomous electric vehicles (EVs) to ensure a safe and reliable driving operation under adversarial conditions. Model-based methods show a relatively fast attack detection compared to data-driven approaches, making them a more practical solution. However, the limitations of existing methods, including model dependency, computational delays in estimators or observers, and sensitivity to residue threshold, can compromise the detection accuracy. To address these challenges, a new residue-free detection framework is proposed based on the characteristics of the zero-sequence current component of the traction motor. Also, a method based on frequency components is introduced to identify the type of cyberattack. The proposed approach is verified on a predictive current-controlled interior permanent magnet synchronous motor (IPMSM) drive using a controller hardware-in-loop (CHIL) system.
