Design method of an ultra-thin two-dimensional hybrid waveguide near-eye display based on forward-ray-tracing and global optimization algorithm

Augmented reality (AR) near-eye display technologies have received significant attention due to their broad application potential in fields such as consumer electronics, industry, and healthcare. Two-dimensional (2D) geometric waveguides enable ultra-thin AR displays with wide fields of view and large exit pupil sizes, while 2D holographic waveguides enable further reductions in device thickness. However, holographic waveguides typically suffer from poor exit pupil illuminance uniformity. To address this issue, we propose a hybrid 2D pupil expansion waveguide that combines holographic and geometric elements. The design utilizes volume holographic gratings for in-coupling and out-coupling, and incorporates an array of partial reflective films for horizontal pupil expansion and beam redirection. Based on the structural characteristics of the hybrid waveguide, we establish a forward ray-tracing model to analyze light propagation and conduct simulation-driven structural optimization. Furthermore, we introduce an illuminance uniformity optimization method that integrates Particle Swarm Optimization (PSO) and Simulated Annealing (SA) algorithms. An evaluation model for energy propagation and illuminance distribution is developed, and a non-sequential ray-tracing method is employed to map the reflectance of the geometric waveguide film system to the output pupil uniformity. PSO is used to identify initial solutions, which are then refined using SA to enhance the uniformity of the exit pupil illumination. Compared to the PSO, the combined optimization algorithm is more efficient and the optimized pupil uniformity improves from 70% to 80%. The final hybrid waveguide design achieves a diagonal field of view of 45°, a thickness of 1 mm, an eye-box size of 14 mm × 10 mm, and an eye relief of 15 mm, with an overall illuminance uniformity of 80%. These results demonstrate the feasibility and effectiveness of the proposed design and optimization approach.