Abstract
This disclosure describes a system that enables software-defined vehicles (SDVs), smart cars, or connected autonomous vehicles to adjust their advanced driver assistance systems (ADAS) based on a persistent driver signature. The driver signature acts as a portable digital identity or user profile representing a driver’s unique physiological and behavioral tendencies, such as typical steering and braking patterns. Instead of storing raw data like video files, the system relies on a cloud database to store a secure mathematical model (a vector or tensor) that represents these individual habits. At the beginning of each trip, the SDV identifies the driver using a near field communication (NFC) sensor and/or a biometric sensor to perform a secure biometric handshake. The SDV retrieves the driver signature of the authenticated driver from the cloud database via a network. The SDV processes the driver signature using an edge tensor processing unit (TPU) or local machine learning model to adjust the one or more ADAS sensors to match the driver's specific style, avoiding long calibration delays and nuisance alerts. Because the driver signature is stored securely in the cloud, it can be downloaded to any compatible SDV, effectively creating a portable “digital bubble” across personal, rental, and fleet vehicles.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Keys, Sean C. and Knese, Karsten, "PERSISTENT DRIVER SIGNATURES FOR SOFTWARE-DEFINED VEHICLES", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/10793