Abstract
In order to enrich the judgment index of the lane departure and avoid a sensitive system which is caused by missing vehicle signals, a method of detecting fatigue lane departure based on human-vehicle-road characteristics has been proposed. At first, an experiment about fatigue lane departure has been taken. And then, relevant parameters that can reveal the human-vehicle-road characteristics are collected and analyzed, compared with that under normal lane changing. At last fatigue lane departure recognition model is constructed based on Gaussian Mixture-Hidden Markov Model (GM-HMM). The recognition results show good performance under online and offline tests.
Highlights
The eSafety Forum (2005) attempted to calculate the benefit of lane changing monitoring systems used in Germany
The study estimates that implementation of the Lane Changing Assistance (LCA) system in vehicles would lead up to a 35% reduction in lane changing related accidents, which would equate to a 2.9% reduction in all accidents
In order to overcome these drawbacks, we introduce a method based on quantitative parameters and can be identified to distinguish between fatigue lane departure and normal lane changing
Summary
The eSafety Forum (2005) attempted to calculate the benefit of lane changing monitoring systems used in Germany. The study estimates that implementation of the Lane Changing Assistance (LCA) system in vehicles would lead up to a 35% reduction in lane changing related accidents (a significant proportion of which were a result of fatigue), which would equate to a 2.9% reduction in all accidents. Distinguished fatigue departure and normal lane changing has great important meaning, it helps increase reliability, reduce false alarm rates for LCA systems. Based on the Optimal Hidden Markov Model, Haijing (2013) has collected three parameters of head rotation, visual movement and vehicle movement. Based on the theory of Hidden Markov Model, Kuge et al (2000) have established the lane changing behavior identification model. Rogado et al (2009) based on Heart Rate Variability (HRV), steering-wheel grip pressure, as well as temperature difference between the inside and outside of the vehicle, make possible to estimate in an indirect way the driver’s fatigue level Different lane changing intention has different recognition rate, The recognition rate of correct lane changing intention by only collecting steering wheel angle is 85% and the recognition rate of lane keeping intention by collecting steering wheel rotation speed is 78.3%. Ahmed et al (2014) based on fatigue symptoms (Eye closure, yawning, head tilting) to detect driver fatigue. Rogado et al (2009) based on Heart Rate Variability (HRV), steering-wheel grip pressure, as well as temperature difference between the inside and outside of the vehicle, make possible to estimate in an indirect way the driver’s fatigue level
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
