Edge-AI Vision Surveillance Robot: Real-Time Object Detection and IoT-Driven Autonomous Navigation

Abstract

Surveillance systems play a crucial role in security and monitoring applications, yet traditional solutions often lack real-time intelligence, autonomous navigation, and efficient object detection capabilities. Recent advancements in AI-driven surveillance have significantly improved object recognition; however, existing systems are limited in mobility, obstacle avoidance, and real-time tracking, restricting their effectiveness in dynamic environments. This study aims to bridge these gaps by developing an Edge-AI Vision Surveillance Robot, integrating YOLO models for real-time object detection and identification alongside Bluetoothcontrolled navigation, line tracking, and obstacle avoidance. The system comprises ESP32-CAM modules, servo motors, and an autonomous braking mechanism. Object detection is facilitated through multiple input sources. The results indicate significant improvements in real-time surveillance efficiency, reducing false detections while ensuring smooth IoT-driven autonomous navigation. The proposed system holds substantial implications for security applications, intelligent monitoring, and industrial automation, offering a scalable, low-power, and cost-effective AI-driven robotic solution for smart surveillance.