An Arduino-based ultrasonic scanner can show the direction and approximate range of a nearby reflecting object, but it cannot identify the object from distance data alone. This paper describes a prototype that extends this type of scanner with webcam-based recognition using a YOLO model. An Arduino UNO rotates an HC-SR04 sensor on a servo motor through a nominal 0°–180° sweep, measures the echo duration, and sends angle–distance readings to a host computer over a serial connection. At the same time, a laptop webcam provides video frames to a Python application. OpenCV handles image capture, and the YOLO detector returns object labels, confidence scores, and bounding boxes. Processing presents the ultrasonic readings in a live radar-style view. A separate detection-records interface stores the object label, confidence, ultrasonic distance, servo angle, and threat level in CSV format. The alert logic uses configurable distance limits and detection conditions to assign Low, Medium, or High levels. Since no measured test dataset was provided, the paper does not claim numerical accuracy or response-time results. It explains the implementation, identifies the conditions needed to associate camera detections with ultrasonic returns, and sets out a procedure for evaluating the prototype. The device is an ultrasonic scanning and computer-vision system; it is not a Doppler radar or LiDAR system.

Maseeha Banu, Baladev DR, Animesh M, Manukrishna, Samyak, "AI-Enabled Arduino RADAR: YOLO-Based Object Detection" International Journal of Scientific Research and Engineering Development, V9(5): Page(2201-2204) September - October 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.