Smart Campus Navigation and Identifying Current Location through Android Device to Guide Blind People

Visual impairment represents one of the most significant barriers to independent mobility in educational environments. An estimated 2.2 billion people worldwide live with a visual impairment, of whom approximately 36 million are classified as blind (WHO, 2024). For blind and visually impaired (BVI) students enrolled in university campuses, independent navigation between classrooms, laboratories, libraries, and administrative offices represents a fundamental daily challenge. Existing solutions — including GPS-only smartphone applications, dedicated hardware devices, and trained guide dogs — each carry significant limitations in terms of indoor coverage, obstacle detection capability, cost, and campus-specific adaptability. This paper presents a comprehensive Android-based campus navigation system engineered specifically for BVI users at Shri Sai College of Engineering and Technology (SSCET), DBATU University, Chandrapur. The system implements a hybrid localisation architecture fusing GPS (outdoor), Bluetooth Low Energy beacon trilateration (indoor, ≤ 0.85 m accuracy), and Wi-Fi RSSI fingerprinting, coupled with Dijkstra shortest-path route planning over a custom campus map graph of 60 geo-tagged Points of Interest (POIs). Voice-based interaction via Android SpeechRecognizer and Text-to-Speech (TTS) APIs enables fully hands-free destination specification and turn-by-turn audio guidance. Ultrasonic distance sensing and camera-based depth estimation provide real-time obstacle detection with a 94% detection rate and 3.2% false alarm rate. A SQLite local database enables fully offline operation; Firebase Realtime Database provides administrative POI updates and cloud synchronisation. Haptic vibration patterns complement audio cues for noisy-environment reliability. Evaluation with 20 visually impaired participants across 100 navigation trials achieved 93.5% task completion, average outdoor localisation error of 3.2 m, BLE indoor error of 0.85 m, voice recognition accuracy of 96.5%, and user satisfaction scores of 4.2–4.4/5.0 across navigation confidence and ease-of-use dimensions. The results demonstrate that a smartphone-only, infrastructure-minimal approach can substantially improve campus mobility independence for BVI students.