Evaluation of Driver Distraction with Changes in Gaze Direction Based on a Vestibulo-Ocular Reflex Model

With the aim of improving parameter identification and, eventually, evaluating driver distraction with changes in gaze direction, we applied a genetic algorithm (GA) method to identify parameters for an existing vestibulo-ocular reflex (VOR) model. By changing the initial inputs to the GA and fixing two parameters pertaining to the horizontal direction, we achieved improved parameter identification with a lower mean-square error. The influence of driver distraction on eye movement with changes in gaze direction was evaluated from the difference between the predicted and observed VOR in the vertical axis. When a driver was given an additional mental workload, the mean-square error between the measured and simulated values was bigger than that in the absence of the mental workload. This confirmed the relationship between driver distraction and eye movement in the vertical direction. We hope that this method can be applied in evaluating driver distraction.

[1]  Flavius-Maxim Petcut,et al.  Solar Cell Parameter Identification Using Genetic Algorithms , 2010 .

[2]  J. Pearl.  Parameter Identiication: a New Perspective (second Draft) , 2001 .

[3]  Goro Obinata,et al.  On-line Method for Evaluating Driver Distraction of Memory-decision Workload Based on Dynamics of Vestibulo-ocular Reflex , 2008 .

[4]  F. Huang,et al.  APPLICATION OF GENETIC ALGORITHM TO STRUCTURAL DYNAMIC PARAMETER IDENTIFICATION , 2002 .

[5]  B. Hess,et al.  Low-frequency otolith and semicircular canal interactions after canal inactivation , 2000, Experimental Brain Research.

[6]  Goro Obinata,et al.  Mental Workloads Can Be Objectively Quantified in Real-time Using VOR (Vestibulo-Ocular Reflex) , 2008 .

[7]  D E Angelaki,et al.  Vestibular Discrimination of Gravity and Translational Acceleration , 2001, Annals of the New York Academy of Sciences.

[8]  D A Robinson,et al.  The use of control systems analysis in the neurophysiology of eye movements. , 1981, Annual review of neuroscience.

[9]  L. Zupan,et al.  Neural processing of gravitoinertial cues in humans. III. Modeling tilt and translation responses. , 2002, Journal of neurophysiology.

[10]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[11]  Goro Obinata,et al.  Quantitative Evaluation of Mental Workload by Using Model of Involuntary Eye Movement , 2009, HCI.

[12]  Vladimir Cretu,et al.  EFFICIENT GENERATION OF NEAR OPTIMAL INITIAL POPULATIONS TO ENHANCE GENETIC ALGORITHMS FOR JOB-SHOP SCHEDULING , 2010 .

[13]  Tatsuya Suzuki,et al.  Parameters optimization using genetic algorithm technique for Vestibulo-ocular reflex model , 2015 .

[14]  Dora E. Angelaki,et al.  Internal models and neural computation in the vestibular system , 2009, Experimental Brain Research.

[15]  Jarmo T. Alander,et al.  Effects of population size and relative elitism on optimization speed and reliability of genetic algorithms , 2006 .

[16]  D M Merfeld,et al.  Neural processing of gravito-inertial cues in humans. IV. Influence of visual rotational cues during roll optokinetic stimuli. , 2003, Journal of neurophysiology.