Characterizing driver fingerprinting of new energy vehicles in risky scenarios: A naturalistic driving study

With the increasing market penetration of new energy vehicles (NEVs), unique driving behaviors and habits compared to traditional fuel vehicles have introduced new traffic safety challenges. Therefore, identifying driving behaviors in NEVs is crucial for behavioral characterization and crash risk prevention. In this context, the “driver fingerprinting (DF)" method, which uses driving behavior data to identify driver identities, has emerged as a promising technique. However, research on NEV driving behaviors and driver fingerprintings remains limited. To address this gap, we conducted a 32-day naturalistic driving experiment (NDE) to collect driving data from 16 NEV drivers. We analyzed driver fingerprinting characteristics in three typical risky scenarios: turning movements, lane-changing, and longitudinal dynamics (including longitudinal acceleration and deceleration). The results show that: 1) Younger drivers exhibit greater speed variability in these scenarios, prefer rapid acceleration, and tend to brake at higher pedal positions; 2) Younger drivers are more likely to accelerate along the X -axis, while older drivers tend to decelerate. Overall, significant differences were observed in vehicle speed, accelerator pedal position, and X -axis acceleration among drivers. • New energy vehicles' (NEVs) driving behaviors were characterized using long-term naturalistic driving data. • NEVs' driver fingerprinting in three typical risky driving scenarios was characterized for the first time. • Speed, longitudinal acceleration, vertical angular velocity significantly varied across NEV drivers.