Holy Angel University

School of Engineering and Architecture

Computer Engineering Program

CPERESEARCH
PROGRESS REPORT No. 1

Project Study Title : AI-Enabled Smart Window with Privacy Detection and Self-
Cleaning
Project Study Group Name : Group 7

Group Members : Layug, Onaire

Perez, Matt Kerby
Yumang, Mikaela

Adviser : Engr. Niyo Kenn Jimenez

Date of Meeting : January 19, 2025
*Venue/Mode of Meeting : Holy Angel University

1. Introduction
1.1. Purpose

The purpose of this report is to document the progress made in the development of
our study titled AI-Enabled Smart Window with Privacy Detection and Self-Cleaning.
This report aims to highlight the accomplishments achieved thus far, the challenges
encountered, and the tasks that remain. It serves as a way to assess the current
state of the project and strategize for its timely completion. Furthermore, this report
ensures that all members are informed of the project’s overall direction and its
alignment with the original goals.

1.2. Background

Our project aims to design a smart window system that leverages artificial
intelligence to provide automated self-cleaning capabilities and privacy detection.
The integration of AI allows the system to adapt to environmental factors such as
dirt accumulation and privacy needs, offering innovative functionality for modern
households and workplaces. The scope of this project also includes a cost-effective
approach to ensure the feasibility of implementation. By addressing current
challenges in smart window optimization, this project contributes to sustainable and
intelligent window maintenance solutions.

1.3. Scope

This report will cover the tasks that have been completed, specifically the review of
related literature (RRL) and related studies (RRS), and the remaining tasks,
 including the methods section. It will also include an analysis of challenges
encountered during the research phase and strategies to address them moving
forward. However, it will not cover the final implementation details or performance
evaluation of the smart window system, as these are outside the scope of the
current progress. This report is focused on providing an update on the research and
design phases of the project.

2. Status
2.1. Tasks Completed
For each relevant task (repeat this pattern for each task or activity completed):
o Provide a description of the task.

The primary task completed was the review of related literature (RRL) and related
studies (RRS) to establish a strong theoretical foundation for the project. The focus was
on identifying recent studies that align with the AI-driven self-cleaning and privacy-
detection aspects of the smart window. This step was crucial for understanding existing
technologies and ensuring that our approach is innovative and feasible within current
technological standards.

o Describe the things that have been accomplished.

We successfully gathered a comprehensive set of references published between 2020

and 2025 to meet the criteria for up-to-date resources. These references provide insights
into smart sensor technology, AI integration, and self-cleaning systems, which are all
critical to our project. By compiling this information, we were able to develop a clear
understanding of the current state of technology and its application to our smart window
design.

o Describe how long it took to accomplish them.

It took three days to complete the review and compilation of RRLs and RRS. This
timeframe included searching through various databases, analyzing the relevance of
each study, and ensuring that all references met the requirement of being published
within the past five years. The time spent was essential to ensure the quality and
relevance of the gathered materials, which will serve as the backbone of our study.

o Describe the difficulties, if any, that were encountered.

One of the main challenges encountered was the limited availability of recent studies
specifically focused on smart windows. As smart window technology is not a widely
researched topic for theses anymore, finding studies that aligned with our criteria was
 time-consuming. Additionally, narrowing down resources that discussed both AI
integration and self-cleaning features required extensive filtering. Despite these
challenges, the team persisted and successfully completed this task.

2.2. Tasks Remaining

For each relevant task (repeat this pattern for each task or activity completed):
o Provide a description of the task.

The remaining tasks include completing the methods section and making minor updates
to the RRLs if new relevant studies become available. The methods section will focus on
detailing how AI will be integrated into the smart window system for both self-cleaning
and privacy detection. This section is vital for outlining the technical feasibility of the
project.

o Describe the things that still need to be accomplished.

We need to finalize the design and describe the specific mechanisms for self-cleaning
and privacy detection in the methods section. This involves detailing how the sensors, AI
algorithms, and mechanical components like DC motors and water pumps will interact.
Additionally, minor updates to the RRLs may be needed if new studies are found that
can further strengthen the theoretical framework of the project.

o Provide a timetable and strategy for completing the task.

The methods section is expected to be completed within 4-5 days, while any updates to
the RRLs can be completed in 1-2 days. Our strategy involves dividing tasks among
team members to ensure efficiency. For example, one member will focus on the
mechanical design, while another will work on documenting the AI integration process.
Regular meetings will be held to ensure progress is on track.

o Describe the risks and your approach for completing the task.

The primary risk is the potential delay in finalizing the methods section due to
unforeseen challenges in detailing the AI integration. To address this, we will allocate
additional time for review and consult available experts or advisors for guidance. Another
risk is the lack of new studies for the RRL updates, but we will mitigate this by ensuring
the current references are sufficient to support our project’s objectives.

3. Conclusion
o Summary: Provide an appraisal of the current status.
The project is nearing completion, particularly in terms of the RRL and RRS, which form
a critical foundation for the study. Despite challenges in finding relevant and recent
references, the team successfully met the requirement of using resources published
between 2020 and 2025. Additionally, adjustments were made to shift from tinting to
privacy detection due to the difficulty in sourcing cost-effective smart films, further
refining the project’s focus.

o Evaluation: Provide an assessment of the progress made to date.

Significant progress has been made, especially in the research phase. The RRL and
RRS are almost finalized, and their completion provides a solid basis for the next stages
of the project. Contributions from team member Mikaela Yumang were instrumental in
compiling and organizing the references, ensuring the project remained on schedule.
The team has demonstrated adaptability in addressing challenges and refining the
project scope as needed.

o Forecast: Provide a forecast for completing this project

With the RRL and RRS nearly complete, the remaining tasks are expected to be finished
within the next 7-10 days. The methods section, which is the primary focus now, will be
completed within 4-5 days, with additional time allocated for minor updates to the RRLs.
Overall, the project is on track to meet its goals, and the team is confident in its ability to
complete all tasks successfully.

------------------------
Layug, Onaire

------------------------
Perez, Matt Kerby

------------------------
Yumang, Mikaela
AI-Enabled Smart Window with Privacy Detection and Self-Cleaning

Smart glass technology has properties that can be affected by factors such as

light, heat, and electricity. This technology can be applied to control light, heat, and

privacy in homes and institutions. They offer a suitable alternative to manual solutions

(e.g. blinds, curtains, shades) as they have differentiated levels of control and ideal for

homeowners seeking privacy and modern aesthetics (International Franchise

Association, 2025).

A Polymer-Dispersed Liquid Crystal (PDLC) glass is utilized for privacy, security,

and energy efficiency, which can be applied as a plastic film onto an existing glass or

laminated into products such as smart windows. Allowing light transmission or the ability

to scatter light to make the glass opaque is what makes the PDLC glass a smart glass.

The scatter of light depends on the electrical voltage applied to it (36-110 Vac),

reorienting the liquid crystals. According to Phatak (2022), this type of glass is only as

smart as the control system drives it, ranging from a simple push-button switch to a

digital automation system. When parired with Arduino and light or temperature sensors,

this enables the PDLC glass respond to environmental changes. Based on sensor

inputs, the glass can adjust transparency to optimize light, heat, and privacy, resulting in

a more energy-efficient system.

Jabbar and Zaki (2021) proposed a system where users can control the intensity

of the PDLC film through thee use of their smartphones. Their study contained a PDLC
film that is controlled by a simple touch interface through an Android application,

connecting to Arduino microcontroller and Bluetooth. The user clicks the button in the

application, an identifier is sent through Bluetooth, then Arduino converts the identifier

into a signal. This action adjusts the electrical current of the film to increase or decrease

the transparency of the glass. The hardware is supplied a 12-volt DC source and a light

sensor module connected outside of the house.

According to Lederer (2024), artificial intelligence (AI) can play another role for

the window industry. This includes the creation of self-cleaning properties and regulate

light levels on windows. It is stated that through the use of smart sensor technology,

these windows can detect environmental factors such as dirt buildup to provide a

sustainable solution to window maintenance.

Artificial intelligence can be integrated in a self-cleaning process for windows

using the DC motor, Arduino controller and water pump (Autodesk Instructables, 2020).

By using AI to remove human intervention for cleaning, the project aims to use the DC

motor to pull the cleaning wipe while the water pump sprays the water as the cleaning

wipe moves downward. The motor driver module allows the operation of pulling the

strings done by the DC motor at variable speeds, with the help of AI integration to

optimize cleaning schedule. The AI setup for self-check, cleaning intensoity adaptation,

and fallback for unexpected weather conditions would be through the input of

environmental factors, such as dirt buildup, rain and water spots, and smog.
References:

Autodesk Instructables. (2020, July). Window Cleaner. Instructables.

https://www.instructables.com/Window-Cleaner-1/

International Franchise Association. (2025). What Are Smart Glass Windows? Glass

Doctor. https://glassdoctor.com/blog/what-are-smart-glass-windows

Jabbar, M. M., & Zaki, S. M. (2021). Smart-Glass Glazing using Arduino and Android

Application. International Journal of Interactive Mobile Technologies (IJIM),

15(05), 54–54. https://doi.org/10.3991/ijim.v15i05.19411

Lederer, P. (2024, September 4). Utilizing Artificial Intelligence in Window Design: How

AI is Revolutionizing the Industry. Dakowindows.com.

https://www.dakowindows.com/en/blog/utilizing-artificial-intelligence-in-window-

design-how-ai-is-revolutionizing-the-industry

Phatak, M. (2022, January 8). What Is PDLC Smart Glass? Smart Glass World.

https://www.smartglassworld.net/what-is-pdlc-smart-glass