• Nilima Prasanna Maybahte

Visually impaired individuals face significant challenges in perceiving their surroundings and avoiding obstacles during daily navigation. Traditional aids such as walking canes and guide dogs provide limited assistance and often fail to detect obstacles that are above ground level or moving. To overcome these limitations, this research proposes an AI and Machine Learning-based real-time navigation system that utilizes a Raspberry Pi and Pi Camera module to detect and identify obstacles in the user’s path. The main objective of the project is to design and develop an intelligent assistive device capable of capturing live video, processing it in real-time, and providing audio feedback or alerts to visually impaired users, ensuring safer and more independent navigation. The proposed system integrates computer vision and deep learning technologies to process the visual data captured by the Raspberry Pi Camera. A pre-trained object detection model—such as YOLO (You Only Look Once) or MobileNet SSD—is implemented to detect and classify obstacles like people, vehicles, walls, or furniture within the environment. The Raspberry Pi 4 Model B serves as the central processing unit, handling both image analysis and real-time decision-making. Once obstacles are detected, the system estimates their position and proximity to the user and delivers corresponding voice-based alerts through a connected speaker or earphones. This system emphasizes portability, affordability, and efficiency, offering a low-cost alternative to expensive navigation aids. The integration of the Raspberry Pi platform ensures compactness and energy efficiency, making it ideal for wearable or handheld applications. The project demonstrates how artificial intelligence and machine learning can be effectively applied to assistive technologies, enhancing accessibility and mobility for visually impaired individuals. Additionally, the proposed model can be extended to support GPS-based outdoor navigation, cloud-assisted object recognition, and edge AI processing for faster performance. Overall, this research aims to contribute toward building a smarter, safer, and more inclusive environment through the innovative use of AI, ML, and embedded systems technology.

Artificial Intelligence (AI) Machine Learning (ML) Real-time Navigation System Obstacle Detection Computer Vision Object Detection Raspberry Pi 4 Model B Raspberry Pi Camera Module Smart Glasses Assistive Technology Visually Impaired Users Deep Learning YOLO (You Only Look Once) MobileNet SSD Tiny-YOLO TensorFlow Lite OpenCV Image Processing Embedded Systems Wearable Technology Real-time Image Analysis Voice Feedback System Text-to-Speech (TTS) Accessibility Tools Edge AI Low-cost Navigation Aid Autonomous Guidance Sensor Integration Distance Estimation Object Classification Indoor and Outdoor Navigation Model Optimization Real-time Performance Human-Centered Design Smart Navigation Portable Assistive Device

"AI and ML-Based Real-Time Navigation System for Obstacle Detection Using Camera for Blind Users", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 12, page no.c20-c36, December-2025, Available :http://www.jetir.org/papers/JETIR2512204.pdf

"AI and ML-Based Real-Time Navigation System for Obstacle Detection Using Camera for Blind Users", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 12, page no. ppc20-c36, December-2025, Available at : http://www.jetir.org/papers/JETIR2512204.pdf