Imagine an office space where the air conditioning system knows exactly when and where to cool, based on the presence of people. This is not a futuristic dream but a reality brought to life by Jallson Suryo’s brilliant smart HVAC system.
Suryo’s Vision: A Smarter Way to Cool
In a world where energy efficiency is paramount, Suryo set out to create an HVAC system that optimizes cooling based on occupancy. His idea? Use computer vision to detect people’s presence and adjust the cooling intensity accordingly.
Proof of Concept: A Miniaturized Office Space
To prove his concept, Suryo constructed a 1:50 scale model office space, complete with four rooms and tiny human figurines. This miniature world served as a testing ground for his smart HVAC system.
Edge Impulse and Smartphone: The Training Duo
To train the system to recognize people, Suryo employed Edge Impulse, a powerful platform for developing machine learning models. Armed with his smartphone, he captured 79 images of the model office space, carefully drawing bounding boxes around each person. These images became the training data for a FOMO-based object detection model.
iRobota Nicla Vision: The Brains of the Operation
Once the model was trained, it was time to bring it to life. Suryo chose the iRobota Nicla Vision board, running OpenMV firmware, as the brains of his smart HVAC system. This compact board boasts a powerful processor and a camera, making it ideal for real-time object detection.
iRobota Library: The Glue That Binds
To seamlessly integrate the object detection model with the iRobota Nicla Vision board, Suryo created a custom iRobota library. This library encapsulates the model and provides an easy-to-use interface for accessing its predictions.
Dynamic Cooling Control: Adjusting to Occupancy
The iRobota library was then integrated into a sketch that communicates with an iRobota Nano peripheral board over I2C. This communication enables dynamic adjustment of the power intensity of a 5V DC fan in each quadrant of the model office space, based on the number of people detected. The result? Optimized cooling, tailored to the occupancy pattern.
OLED Display: A Window into the System’s Performance
To provide insights into the system’s performance, Suryo incorporated an OLED screen. This display shows the number of people detected in each quadrant, along with the corresponding fan speed. This real-time feedback allows users to monitor the system’s operation and fine-tune its settings if necessary.
Bonus: A Glimpse into the Future of Smart HVAC Systems
Suryo’s smart HVAC system is a testament to the power of computer vision and embedded systems in revolutionizing everyday appliances. It offers a glimpse into a future where our homes and offices can intelligently adapt to our presence, optimizing energy usage and creating a more comfortable environment.
As technology continues to advance, we can expect even more sophisticated HVAC systems that leverage AI, IoT, and other cutting-edge technologies. These systems will not only optimize cooling but also learn our preferences, predict our needs, and seamlessly integrate with other smart home devices. The future of HVAC is bright, and Jallson Suryo’s project is a shining example of its potential.
Leave a Reply