Abstract : This research developed an Internet of things (IoT)services system for cyclists to keep them safe when cycling –from terminal to back end – called the cycling safety servicessystem. The proposed system consists of wearable devices, aservice mobile app, and back-end services, primarily providingthree categories of services: (1) The cycling team services, (2)the physiological status services, and (3) the environmental informationservice, incorporating the technologies of deep learning,IoT end device development, mobile app programming, RESTfulAPI implementation, open data exploitation, etc. The proposedsystem aims at protecting the cyclists from being left out whencycling as a team, warning them when their physiological statusis going to be abnormal or when there is a possibility of a bicyclecrash, as well as proactively sending out urgent messages withlocation information when a crash occurs. Moreover, to enablethe mobile app to recognize crash events, this research traineda deep learning bicycle crash model of 87.8049% accuracy andimplemented a procedure in the mobile app based on this modelto detect bicycle crashes automatically. The proposed system alsoprovides a way for the cyclists to report any false crash alarm sothat the crash model can be re-trained to reduce its false alarmratio after the system is distributed to the consumers in thefuture. This research confirmed that the proposed system workswell, and suggests that the proposed system, the crash model, thedata collection method with its associated mobile apps, and theanonymous crash locations collected in the future can be valuableand contribute to the cycling society and relevant researchers ina positive way.
Index terms : AI, artificial intelligence, bicycle crash, cycling, cycling team, deep learning, Internet of things, IoT, off the team.