Augmented reality head-up displays (AR-HUDs) are mainly categorized into single optical path HUDs and dual optical path HUDs. Compared to single optical path HUDs, dual optical path HUDs have received significant attention due to their capability to separately display basic driving information and real-time navigation data. However, existing dual optical path HUDs often use orthogonal projection for presenting real-time navigation information, which may result in insufficient augmented reality effects and visual fatigue for drivers due to frequent recognition of road conditions. To address these challenges, we propose a novel dual optical path HUD structure to present more effective information to drivers while reducing costs and saving space. The system demonstrates excellent display performance in real-time navigation information presentation.

To minimize costs and save space, our system uses a single picture generation unit (PGU) to achieve dual optical path displays for both near and far fields. To enhance the augmented reality effects of real-time navigation information in the far field, the far optical path adopts oblique projection, achieving seamless ground fusion of navigation data. Based on geometric optics imaging principles, the system leverages the relationship between object distances and virtual image positions. The initial structure of the automotive head-up display system using dual optical path oblique projection in our study is designed based on this principle. Specifically, an optical wedge is introduced between the primary reflector and the PGU to achieve a small angular deflection in the near optical path. This alters the optical length of the near optical path, allowing the light emitted by the PGU to produce two virtual images at different distances in front of the driver after multiple reflections. Light from the PGU passes through a primary mirror, a secondary mirror, and a windshield before entering the human eye. Finally, it converges into two virtual images, far and near images, at different distances in front of the windshield. The near field uses orthogonal projection, while the far field uses oblique projection.

Our system integrates two separate areas on a single PGU. An optical wedge relays one area to a closer position, enabling dual optical path displays for both near and far fields. Imaging quality analysis shows that when the driver’s eye is positioned at the center of the eye box, the system’s dot spots remain within the Airy disk across the entire field of view. The maximum RMS spot radii for near and far optical paths are 21.636 and 26.210 μm, respectively, significantly smaller than their corresponding Airy disk radii. The minimum MTF values at the cut-off frequency are 0.586 and 0.524 for the far and near optical paths, respectively, both exceeding 0.5, with grid distortion under 5%, meeting the design requirements. UI simulations demonstrate minimal distortion in the effective area at the center of the UI both before and after the near optical path simulation. The UI image corresponding to the far optical path is clear and exhibits a trapezoidal effect, aligning with the driver’s visual characteristics during driving and effectively displaying relevant driving information.

In this paper, we propose an automotive HUD using dual optical path oblique projection. By introducing an optical wedge between the PGU and the primary mirror, the optical length of the near optical path is altered, enabling simultaneous dual image plane displays for both far and near fields. The near optical path adopts orthogonal projection to display basic driving information, while the far optical path adopts oblique projection for real-time navigation information fusion display on the ground. The design reduces the driver’s visual burden, enhances the visual experience, and delivers an AR-HUD with virtual image distances of 5 m (corresponding to a field of view angle of 10°×1°) and 6.5?8.5 m (corresponding to a field of view angle of 10°×2°). The system achieves excellent imaging quality and meets the modern processing level, providing a reference for the future development of automotive HUDs.