Related Experiment Video
Updated: Aug 21, 2026
A Standardized Obstacle Course for Assessment of Visual Function in Ultra Low Vision and Artificial Vision
Published on: February 11, 2014
Understanding AI-powered assistive technologies for visual impairment: a review of functional features
Vikas Singh Panwar1, Dhruv Kumbhar2, Vedant Kumbharkar2
1Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, India.
Purpose:
This review examines recent developments in AI-powered assistive technologies for individuals with visual impairment, focusing on object detection, obstacle avoidance, navigation, haptic feedback, and user-centered design. It aims to synthesize their capabilities, advantages, limitations, and gaps relevant to practical and rehabilitation-oriented applications.
Materials And Methods:
A structured literature review was conducted using Scopus, Web of Science, IEEE Xplore, PubMed, and Google Scholar. Literature published between 2017 and 2025 was considered. Studies were screened using predefined inclusion and exclusion criteria, and information on device type, AI and sensing techniques, navigation approaches, feedback mechanisms, and user-centered considerations was extracted and thematically analyzed. A total of 110 studies were included in the review.
Results:
The reviewed literature demonstrates substantial development of smart canes, wearable systems, smart glasses, smartphone-based applications, and other AI-enabled assistive devices. Vision-based and deep learning approaches support object recognition and environmental perception, while ultrasonic, LiDAR, and other sensing technologies contribute to obstacle detection and navigation. Haptic and auditory feedback provide alternative means of conveying environmental information. However, challenges remain in dynamic environments, including computational and energy requirements, false detections, environmental sensitivity, cognitive load, usability, affordability, and limited real-world validation. Emerging multimodal and edge-based approaches show potential for improving adaptability and accessibility.
Conclusion:
AI-powered assistive technologies have considerable potential to improve mobility, environmental awareness, independence, and quality of life. Future research should emphasize multimodal perception, advanced AI models, adaptive feedback, energy-efficient systems, and long-term user-centered evaluation, with greater attention to affordability, accessibility, and real-world validation.