Efficient gradient computation in optical system optimization

Martin Pflaum

doi:10.1117/12.3017062

17 June 2024 Efficient gradient computation in optical system optimization

Martin Pflaum

Proceedings Volume 13023, Computational Optics 2024; 130230F (2024) https://doi.org/10.1117/12.3017062
Event: SPIE Optical Systems Design, 2024, Strasbourg, France

Conference Poster

Abstract

In numerous applications within laser material processing and various other disciplines, achieving a specific irradiance distribution is a frequent objective. By optimizing the parameters of an optical system, light can be redirected to attain the desired irradiance distribution. In gradient descent-based approaches, a major challenge involves accurately and efficiently computing the objective function's gradient concerning these parameters. Algorithmic differentiable ray tracing provides a promising approach for this challenge. To calculate the gradient of the objective function, we derive a new mathematical expression that applies to a wide range of optical systems and can be evaluated using algorithmic differentiable ray tracing and automatic differentiation techniques. This novel methodology for calculating the gradient of the objective function is both computationally efficient and robust in its applicable domain. To validate our approach, we additionally provide numerical simulation results.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Martin Pflaum "Efficient gradient computation in optical system optimization", Proc. SPIE 13023, Computational Optics 2024, 130230F (17 June 2024); https://doi.org/10.1117/12.3017062

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available