Simulation of ray behavior in biconvex converging lenses using machine learning algorithms - Article

Aug 5, 2026

This study uses machine learning algorithms to simulate light-ray behavior in biconvex converging lenses with reinforcement learning for 3D lens refraction. Existing approaches often focus on image formation or ray tracing without RL methods such as proximal policy optimization (PPO) and soft actor-critic (SAC). The research evaluates and compares these two algorithms in an optical simulation setting. Results indicate PPO provides superior ray convergence, with improved stability and accuracy over SAC, making PPO a promising direction for optimizing optical ray simulators and supporting more complex optical scenarios.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Introduction

Related Works

Converging biconvex lenses

Reinforcement learning methods

Proposed Simulation

Results and Discussion

Conclusions and Future Work

Indonesian Journal of Electrical Engineering and Computer Science Vol. 38, No. 1, April 2025, pp. 357∼366 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v38.i1.pp357-366 ❒ 357 | Simulation of ray behavior in biconvex converging lenses using machine learning algorithms
Juan Deyby Carlos-Chullo, Marielena Vilca-Quispe, Whinders Joel Fernandez-Granda,
Eveling Castro-Gutierrez
Universidad Nacional de San Agustin de Arequipa, Arequipa, Peru | | | | --- | --- | --- | | Article Info
Article history:
Received May 20, 2024 Revised Oct 21, 2024 Accepted Oct 30, 2024
Keywords:
Converging biconvex lenses Machine learning Proximal policy optimization Reinforcement learning Soft actor-critic | ABSTRACT
This study used machine learning (ML) algorithms to investigate the simulation of light ray behavior in biconvex converging lenses. While earlier studies have focused on lens image formation and ray tracing, they have not applied reinforcement learning (RL) algorithms like proximal policy optimization (PPO) and soft actor-critic (SAC), to model light refraction through 3D lens models. This study addresses that gap by assessing and contrasting the performance of these two algorithms in an optical simulation context. The findings of this study suggest that the PPO algorithm achieves superior ray convergence, surpassing SAC in terms of stability and accuracy in optical simulation. Consequently, PPO offers a promising avenue for optimizing optical ray simulators. It allows for a representation that closely aligns with the behavior in biconvex converging lenses, which holds significant potential for application in more complex optical scenarios. | | | This is an open access article under the CC BY-SA license. | | | | Corresponding Author: | | | | Juan Deyby Carlos Chullo
Universidad Nacional de San Agustin de Arequipa Arequipa, Peru
Email: [jcarlosc@unsa.edu.pe](jcarlosc@unsa.edu.pe) | | | 1. INTRODUCTION Converging lenses, such as biconvex lenses, are designed to form both real and virtual images [1] . These lenses are essential for improving the precision with which we observe and study objects [2] . Through the refraction of light, converging lenses enable illuminated objects to project onto a screen, creating images that can be examined for various scientific purposes [3] . While several applications simulate image formation through these lenses, many do not fully capture the complex behavior of light rays. One such application, AR-GiOs, as analyzed in [4], has shown promising results in the academic field, particularly for learning about the formation of real and virtual images. However, despite its success in educational settings, AR-GiOs still struggles to accurately simulate the behavior of rays passing through optical systems. This gap highlights the limitations of current simulation tools in capturing the subtle details of ray behavior, which are fundamental to the study of physical optics. Several applications attempt to simulate image formation through lenses, but they often fail to accurately model the light rays involved in the process [4],[5] . These rays, referred to as principal, central, and focal rays, are essential for understanding key optical behaviors when light passes through lenses or mirrors. Accurate simulation of these rays is crucial because they dictate how images are formed and how optical systems function, yet many existing tools lack the necessary fidelity to simulate them effectively. No studies have been identified that apply reinforcement learning (RL) algorithms to model light refraction through lenses. RL methods like proximal policy optimization (PPO) [6], [7] and soft actor-critic (SAC) [8], [9] are extensively used in artificial intelligence (AI) and machine learning (ML) for decisionmaking tasks [10],[11] . These algorithms operate based on learning from interactions with their environment, where an agent makes decisions and is given feedback through rewards or

FAQs

What machine learning and reinforcement learning algorithms are compared in the study?

The study compares proximal policy optimization (PPO) and soft actor-critic (SAC) for controlling ray trajectories in a lens simulation.

What is the main goal of using RL in the biconvex converging lens simulation?

The goal is to guide rays through the lens so they converge at points corresponding to real or virtual image formation.

Which algorithm performs better and why?

PPO achieves superior ray convergence, outperforming SAC in stability and accuracy in the optical simulation context.

Aug 5, 2026

Upload by Evangeline