Exploring Enhancement of AR-HUD Visual Interaction Design Through Application of Intelligent Algorithms

Abstract

This study aims to optimize the visual interaction design of AR-HUD and reduce cognitive load in complex driving situations. An immersive driving simulation incorporating eye-tracking technology was utilized to analyze objective physiological indices and measure subjective cognitive load using the NASA-TLX. Additionally, a visual cognitive load index was integrated into a BP-GA neural network model for load prediction, enabling the derivation of an optimal solution for AR-HUD design. The optimized AR-HUD interface demonstrated a significant reduction in cognitive load compared to the previous prototype. The experimental group achieved a mean total score of 25.63 on the WP scale, whereas the control group scored 43.53, indicating a remarkable improvement of 41.4%. This study presents an innovative approach to optimizing AR-HUD design, effectively reducing cognitive load in complex driving situations. The findings demonstrate the potential of the proposed algorithm to enhance user experience and performance.