Long Short Term Memory 기반의 운전자 행동 모니터링 시스템

Abstract

The driver behavior monitoring system is a system for observing unexpected maneuvers of a driver and alerting dangerous situation to prevent traffic accidents. In this paper, we propose the ‘Long Short Term Memory (LSTM)’ based driver behavior monitoring system. The LSTM is suitable for monitoring driver`s behavior because it is easy to process long term sequential data through selective memory algorithm using the partially storing method. The verification about the prediction simulation was conducted by comparing the actual driving data and the prediction results using the LSTM which implemented by computer language.