Outdoor navigation has long been a wicked problem for the visually impaired. With the development of technologies, new designs with them are temping to deal with existing obstacles both from within and outside, but still leaving pain points to be solved. Driven by social considerations and user-centred perspectives, we propose an innovative and sustainable design framework for assistive devices catering to the visually impaired in this project. The core implementation layer utilizes Arduino, while the decision-making layer employs the YOLOv5 algorithm for object detection tasks, enabling hardware interaction in response to the detection results. The device successfully provides features such as environmental information feedback, tactile paving search and privacy intrusion warning, transforming necessary visual information into auditory and tactile feedback. By controlling it at a lower cost and simplifying interaction methods, our device offers essential mobility assistance to a broader range of visually impaired users.

Zihan Ma. Design Research on the Application of YOLOv5-based Road Condition Detection in Assistive Devices for the Visually Impaired. Academic Journal of Computing & Information Science (2023), Vol. 6, Issue 10: 106-112. https://doi.org/10.25236/AJCIS.2023.061016.

[1] Wu J. Introduction to convolutional neural networks [J]. National Key Lab for Novel Software Technology. Nanjing University. China, 2017, 5(23): 495.

[2] Redmon J, Divvala S, Girshick R, et al. You only look once: Unified, real-time object detection [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, Nevada, United States: IEEE Computer Society Conference Publishing Services, 2016: 779-788.

[3] Chen D, Gao Y, Song A, Liu J, Zeng H. Touchscreen interactive finger-mounted Braille reproduction system[J]. Chinese Journal of Scientific Instrument, 2022, 43(5): 199-208.

[4] Sáez Y, Montes H, Garcia A, et al. Indoor navigation technologies based on RFID systems to assist visually impaired people: A review and a proposal [J]. IEEE Latin America Transactions, 2021, 19(8): 1286-1298. 

[5] Du Y, Yuan X, Ma X. Design of an Intelligent Cane Based on Multi-sensor Fusion Technology [J]. Practical Electronics, 2021(07): 80-81+19.

[6] João Guerreiro, Daisuke Sato, Saki Asakawa, Huixu Dong, Kris M. Kitani, et al. CaBot: Designing and Evaluating an Autonomous Navigation Robot for Blind People[C]//Proceedings of the 21st International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS '19). New York, New York, United States: Association for Computing Machinery, 2019:68–82.

[7] Buchanan R. Design research and the new learning [J]. Design Issues, 2001, 17(4): 3-23.

[8] Brown T, Wyatt J. Design thinking for social innovation [J]. Development Outreach, 2010, 12(1): 29-43.