Machine Learning Based Torque Monitoring Algorithm for Preventing Unintended Acceleration and Deceleration in Vehicles

The primary purpose of E-GAS monitoring is to ensure the functional stability of the electronic controller in relation to vehicle torque. At level 2 of the E-GAS monitoring concept, the calculation of permissible torque typically relies on formula-based models that simplify real vehicle behavior, accompanied by complex calibration to enhance accuracy. However, this approach often fails to adequately account for the diverse driving scenarios encountered by the vehicle. To address this limitation, this study proposes an algorithm for calculating permissible torque using machine learning at level 2 of the E-GAS monitoring concept. The effectiveness of the algorithm is validated through the analysis of real-world vehicle driving data, confirming its practicality and applicability.