Unique Presentation Identifier:
12
Program Type
Undergraduate
Faculty Advisor
Dr. Jessica Young
Document Type
Poster
Location
Face-to-face
Start Date
9-4-2026 1:00 PM
End Date
9-4-2026 3:00 PM
Abstract
The goal of our work is to automate the tedious and delicate process of optical alignment (rotating mirrors, shifting lenses by millimeters, and iterating endlessly) in the context of generating vortex beams from a Gaussian laser beam. We create vortex beams by illuminating a digital hologram displayed on a spatial light modulator (SLM). Two main issues arise: misalignment and phase imperfections in the input beam. If the beam does not pass directly through the center of the SLM, or if it deviates from an ideal Gaussian profile, the resulting vortex beam becomes distorted.
Last semester, we developed two methods to automate precision alignment using only intensity images of the imperfect vortex beam captured by a scientific camera. Both methods operate on the same principle, shifting the hologram pattern on the SLM until the beam effectively passes through the center, but differ in runtime and accuracy. The first samples several hologram offsets, analyzes the resulting beam geometries, and selects the optimal center. It is highly accurate but computationally intensive. The second begins with an initial guess and iteratively refines the offset using simpler geometric metrics, making it up to an order of magnitude faster at some cost to precision. In either case, the outcome is a centered and well-aligned vortex beam.
This semester, we are focusing on phase correction. Even with perfect alignment, imperfections in the input beam prevent us from generating an ideal vortex. To address this, we modify the hologram to include combinations of common optical aberrations. By analyzing the resulting interference patterns, we aim to train a machine learning model to map observed patterns to their underlying aberrations, enabling us to compensate for them directly in the hologram.
Recommended Citation
Temple, Joseph E., "Computational Approaches to Laser Alignment and Phase Correction" (2026). ATU Scholars Symposium. 42.
https://orc.library.atu.edu/atu_rs/2026/2026/42
Included in
Computational Approaches to Laser Alignment and Phase Correction
Face-to-face
The goal of our work is to automate the tedious and delicate process of optical alignment (rotating mirrors, shifting lenses by millimeters, and iterating endlessly) in the context of generating vortex beams from a Gaussian laser beam. We create vortex beams by illuminating a digital hologram displayed on a spatial light modulator (SLM). Two main issues arise: misalignment and phase imperfections in the input beam. If the beam does not pass directly through the center of the SLM, or if it deviates from an ideal Gaussian profile, the resulting vortex beam becomes distorted.
Last semester, we developed two methods to automate precision alignment using only intensity images of the imperfect vortex beam captured by a scientific camera. Both methods operate on the same principle, shifting the hologram pattern on the SLM until the beam effectively passes through the center, but differ in runtime and accuracy. The first samples several hologram offsets, analyzes the resulting beam geometries, and selects the optimal center. It is highly accurate but computationally intensive. The second begins with an initial guess and iteratively refines the offset using simpler geometric metrics, making it up to an order of magnitude faster at some cost to precision. In either case, the outcome is a centered and well-aligned vortex beam.
This semester, we are focusing on phase correction. Even with perfect alignment, imperfections in the input beam prevent us from generating an ideal vortex. To address this, we modify the hologram to include combinations of common optical aberrations. By analyzing the resulting interference patterns, we aim to train a machine learning model to map observed patterns to their underlying aberrations, enabling us to compensate for them directly in the hologram.