AI Quad-Cam Drone Detector with vibration aleart

I want to place four small cameras on a tabletop rig, point them outward, and have the system automatically notice when a drone passes through their vibrating a motor and lighting an LED for the direction the drone is in. The second it decides the object really is a drone, a vibration motor should buzz; if the object is only a bird in flight, the system must stay quiet. No extra screens, maps, or boxes—just the vibration alert and, for debugging, a console or log entry is fine.

The core of the job is the AI model that can draw information from all four feeds at once, recognise a drone at a variety of angles and distances, and confidently reject birds. You are not free to select the edge-device platform I have equipment this is concept testing. If something could save power of increase location time it will be taken into consideration. As long as everything runs locally with minimal latency. I expect you to handle camera calibration, multi-view fusion, model training (TensorFlow, PyTorch, or another proven framework), and the simple GPIO-driven vibration output.

Please deliver:
• The fully assembled tabletop demonstrator with four synchronised cameras and vibration module
• Source code, trained weights, and clear setup instructions
• A short video that shows a drone triggering the buzzer and a bird target being ignored

I will test by flying a small quadcopter and then releasing a bird decoy; the demonstrator must buzz only for the drone. If you can meet that benchmark reliably, I look forward to working together.

(Person Must Be In Alexandria Egypt)