Adaptive Shared Lateral Control and Human Driver Modeling for Enhanced Driver Assistance Systems in Autonomous Vehicles

ABSTRACT

This paper proposes a novel advanced Adaptive Driver Assistance System (ADAS) that integrates human driver modeling into a shared lateral control framework for autonomous vehicles. The human driver model incorporates steering column dynamics, sensory delays, neuromuscular system effects, and vision‐based steering control, simulating real driver behaviors. We introduce a Linear Parameter Varying (LPV) polytopic controller with a human‐in‐the‐loop model to assist with lateral vehicle control, adapting dynamically to various driver profiles, as well as to changes in vehicle speed and road curvature. We illustrate the controller's performance by presenting simulation results on a Renault Megane vehicle model to show its effectiveness in reducing lateral offset errors across multiple driver profiles, highlighting its adaptability and robustness in improving driver‐vehicle interaction.