H.A.L.T: Heat Alarm and Life-Saving Technology - A Safety System to Prevent Hot Car Deaths in Children and Pets

Overview

Children dying from vehicular heatstroke is one of the leading causes of non-crash, vehicle-related deaths for those ages 14 and younger. On average, about 39 children and hundreds of pets die each year due to heatstroke after being left in hot cars. Even on relatively mild days, temperatures inside a car can rise to dangerous levels in just minutes – opening windows or parking in shade does little to slow this process. Children and pets are especially susceptible due to their smaller body size and less efficient cooling mechanisms. According to Dr. Diamond, a PhD in neuroscience, even attentive caregivers can unintentionally forget a child due to normal working-memory lapses caused by stress, sleep deprivation, or routine disruptions. Both federal and state efforts, including the Infrastructure Investment and Jobs Act and the HOT CARS Act, reflect growing recognition of this risk and the need for automatic detection systems. Most existing prevention systems rely on door logic or single sensors, which can fail if a child is sleeping, restrained, or obstructed. These types of built-in car sensors are limited in scope while more advanced heat-alert solutions - like those for K-9 units - are costly. Because of this, I created H.A.L.T - an affordable and reliable safety system intended to prevent the hot car deaths of babies, children, and pets. H.A.L.T. addresses this problem by combining real-time temperature monitoring with ultrasonic motion sensing and camera-based machine learning to detect actual occupants. The system, built with an Arduino R4 WiFi with an ESP32-CAM module, sends tiered push notifications and live updates via a mobile app. Additionally, a buzzer sounds at 95°F if an occupant remains inside, alerting nearby bystanders. Combining motion sensing with camera vision and machine learning, it ensures that no occupant will be missed in case of sensor failure. The machine learning model, trained on 395 images under varied conditions, achieved 96.9% accuracy. By detecting actual occupants rather than relying on assumptions or single-sensor logic, H.A.L.T. offers a reliable, affordable, and accessible solution with the potential to save lives through early intervention, preventing hot car tragedies before they occur.

Competition history

• CSEF 2026 Electronics & Electromagnetics (Senior Division) · Entry S-10-03

Related projects

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

Browse more like this

• All Physics and Astronomy projects
• CSEF Physics and Astronomy projects
• Every CSEF 2026 project