Design of Personal Mobility Safety System Using AI

Abstract

In this paper, we propose the implementation of a safety device that generates an alarm sound or braking operation to reduce the risk of accidents. It reduces the exposure of risks due to non-wearing by supplementing the function of the helmet for safety. For machine learning, the safety state is learned by using two types of sensing data, and when an abnormal helmet use or speed or drinking driving is detected, an alarm sound is generated and motion is broken to maintain the safe state. By measuring data using a gas sensor, alcohol is checked and this is used as abnormal data. Users form a habit of wearing safety equipment with continuous safety alarm sound and speed braking and proper driving habit by driving in a normal state without drinking alcohol. In addition, the proposed system enables real-time monitoring, thereby reducing risks by continuously maintaining safe driving and wearing protective equipment. The proposed system uses artificial intelligence to discriminate data related to helmet wearing, speed, and drinking in making an electric kickboard for safety, and triggers an alarm or operates the brake to prevent abnormal driving. If the design and function are supplemented, it will become a basic function that can be applied to various equipment of transportation.