SKYGUARD AI

• 

  Image_6

• 

  Logo

• 

  Image_3

• 

  Image_2

• 

  Image_4

• 

  Image_1

• 

  Image_5

-Inspiration

To be really honest, everything started from a single YouTube video, which is about a "Navy CIWS Gun", that has an automated target lock system and I wanted to build something similar, that connects artificial intelligence with the physical world. AI can recognize objects in images, but I wanted to take it a step further by building a system that can detect an object through a webcam and physically respond to its movement. This idea inspired me to create SKYGUARD AI, a real-time object detection and tracking prototype.

-How I Built It

I built SKYGUARD AI using Python, YOLOv8, OpenCV, and an Arduino Uno and pre-trained AI model that helps to detect objects in the screen using any available camera. A camera captures live video, and YOLOv8 analyzes the frames to identify objects. Python processes the detection results and calculates the object's horizontal position. It then sends commands to the Arduino through serial communication.

The Arduino controls a servo motor that follows the object's horizontal movement. I also connected red and green LEDs to indicate whether an object is detected, along with a 16x2 LCD that displays the system's status.

-Challenges We Faced

One of my biggest challenges was getting all the hardware components to work together. Initially, my LCD displayed strange characters instead of readable text. I had to troubleshoot the wiring and pin configuration to get it working correctly. I also had to figure out how to communicate between Python and Arduino and make the servo respond smoothly to the object's position.

These challenges taught me that building an AI-powered hardware project involves much more than writing code. The software, electronics, and communication between components must all work together.

-What I Learned

Through this project, I gained hands-on experience with computer vision, object detection, image processing, serial communication, and microcontroller programming. I learned how to connect a machine learning model to a physical device and turn its predictions into real-world actions.

Most importantly, I learned how to debug problems across both software and hardware instead of giving up when something did not work as expected.

-What's Next

I want to improve SKYGUARD AI by making its tracking more accurate and responsive, handling multiple objects more effectively, and recording detection events for later analysis. This prototype is my first step toward exploring how computer vision and embedded systems can work together in real-world applications.

This is Mirakbar and my team and we do appreciate for your time and attention!

Built With

• ai
• air-defense-system
• arduino
• c++
• computervision
• electricalengineering
• hardwareintegration
• lcdscreen
• livefeedanalysis
• machine-learning
• objecttracking
• opencv
• pretrainedmodel
• pyserial
• python
• servomotor
• target-lock
• yolo