DefinePK

Abstract

This project introduces an AI-Powered Autonomous System for a Real-Time Blind Smart Shoe, aimed at enhancing mobility and independence for visually impaired individuals. The system integrates artificial intelligence, Internet of Things (IoT), and real-time computer vision to identify obstacles and navigate safely through various environments.At the core of the system is an ESP32 camera module, which continuously captures the surroundings in real time. These visual inputs are transmitted to a dedicated Android APK application via Wi-Fi. The application processes the incoming frames using pre-trained deep learning models to detect and classify objects or obstacles in the user’s path. Once an object is identified, the app immediately converts the detection result into voice output using Text-to-Speech (TTS) technology.To ensure that the voice feedback reaches the user clearly, a USB Bluetooth audio module is connected to a small speaker or headset worn by the user. This enables the system to provide spoken alerts, such as "Obstacle ahead", "Person", or "Vehicle", which helps the blind user make informed decisions while walking.Additionally, the system includes Google Maps integration within the app to support GPS-based route guidance. The user can input or select a destination, and the app will provide real-time navigation commands such as “Turn left in 20 meters” or “You have reached your destination”, ensuring smooth travel along predefined or dynamically generated routes.By combining real-time image recognition, auditory feedback, and navigation assistance, the Blind Smart Shoe system serves as a smart wearable solution that improves both environmental awareness and route navigation for blind and visually impaired individuals. It ultimately promotes safety, autonomy, and confidence in daily mobility