illumiLOGIC: Redefining Illumination through Multi-Sensor Integration

Overview

Streetlights are essential to urban infrastructure and proper illumination but consume excessive energy regardless of actual usage, demonstrating the need for a more energy -efficient street lighting system through the use of external sensors and data. The hypothesis proposes that integrating passive infrared (PIR) and acoustic sensors into an adaptive lighting system will reduce energy consumption by at least 30% whilst remaining 75% accurate compared to traditional systems while maintaining effective illum ination. The adaptive streetlight prototype, featuring an HC -SR501 PIR sensor, an MAX4466 microphone, Arduino Uno Rev 3, and an ESP32 microcontroller, logged root mean squared (RMS) noise, peak noise, and PIR detection, along with time and weather data. Statistical analysis confirmed significant differences between the presence and absence of each variable, and a logistic regression model was implemented to predict presence based on sensor data, with an accuracy of 84.63%. The system uses this presence prediction to determine the brightness of the LED, which utilizes a rolling average for smooth transitions, a max brightness threshold, and a weather -adjusted safe minimum brightness level. The final model achieved a 50.29% reduction in energy usage compared t o a traditional streetlight, based on the controlled 67% presence rate testing scenario. This translates to $21.15 saved per streetlight annually— or $6.66 million annually if implemented in New York City. These findings validate the hypothesis, demonstrating substantial energy and cost savings while maintaining high detection accuracy and safety.

Competition history

• JSHS 2025 Category not listed

Resources

• National JSHS abstract book ↗

Related projects

Closest projects by meaning, across every fair and year in the corpus.