Automotive augmented reality 3D head-up display based on light-field rendering with eye-tracking

Area of Science:

• Optics and Photonics
• Human-Computer Interaction
• Automotive Technology

Background:

• Head-up displays (HUDs) are crucial for automotive safety and information delivery.
• Stereoscopic 3D displays enhance immersion and depth perception.
• Integrating eye-tracking with light-field displays offers potential for advanced automotive HUDs.

Purpose of the Study:

• To investigate the feasibility of a stereoscopy-based 3D HUD for automotive applications.
• To develop and validate a system combining eye-tracking, light-field display, and HUD optics.
• To assess the performance of the prototype in terms of viewing conditions and crosstalk.

Main Methods:

• Implemented a "light-weight" equation to map driver's eye position to a virtual eyebox plane.
• Replaced actual optics with an effective lens model for simplified simulation.
• Developed a light-field rendering algorithm utilizing processed eye-tracking data.
• Validated the system through ray-tracing simulations and prototype experimentation.

Main Results:

• Experimental results closely matched ray-tracing simulations.
• Achieved a virtual image field of view of 10° × 5°.
• Demonstrated static crosstalk of less than 1.5%.

Conclusions:

• The developed stereoscopic 3D HUD system is feasible for automotive applications.
• The system effectively delivers high-quality 3D visuals with precise eye-tracking integration.
• The prototype confirms the viability of the proposed approach for future automotive displays.