Study On Night-time Drivers Visual Search Mode On City Road

The major causation which triggers the traffic accident is human factors in driving; however the perception factors are dominated in human factors. The majority perception of information for drivers is vision, the peculiarity of nighttime traffic accidents possess with frequently and high mortality, therefore, this research is based on eye movement under the various environments at nighttime, which will conductive to reveal nighttime driving vision search rate of drivers. This research occur significant theoretical guidance and practical consequences to improve the standard design of traffic safety facilities and perfect measures of traffic control.Experiment with 7 eligibility testers with simulator, to complete the drive missions under traffic environmental of identical urban roads at day and night operation, use eye tracker to record the relevant eye movement data when the testers on the driving process. To combining China and abroad theoretical research of drivers’vision search rate, which determine basic parameters of characterization for accumulate fixation time, characteristic of angle of sight, fixation duration and angle of sight’s breadth search.Use statistics methods to processing and analysis gathering data. Firstly statistically analysis drivers’ fixation time, characteristics of angle of sight and the difference and regularity of visual breadth search at day and night operation. Secondly, use the dynamic clustering method to research view plane of testers’ right turn, left turn and straight drive through into intersection, according to result of fixation area partition, analysis the differences of testers’fixation area between day and night operation. Additionally, statistics the fixation objects and analysis frequencies of fixation when testers drive through signalized intersection, use the Markov one-step transition probability matrix to matrixing the fixation frequency of testers’ at the intersection, to find the fixation stationary distribution. The final section describe the research result apply to formulate traffic control policies and optimal design the traffic facilities.The research results provide that in between day and night operation will have an effect on testers’ visual search pattern.The major results were summarized as follows:1. The horizontal view angle and vertical view angle of testers exist significant differences along with day and night operation changes, however the visual breadth search and fixation duration does not reflect significant differences.2. Use dynamic cluster analysis to classify the vision suburbs of testers’ in different ways to drive through intersections in day and night operation, recognize classification suburbs will reflect the differences.3. Replay video of testers’ drive through intersections frame-by-frame, obtain the testers’ frequencies of fixation objects and visual points when drive through intersections in different ways, show the diversity between frequencies of fixation and visual points.4. One-step transition probability matrix and stationary distribution of testers’ in intersections reflect the diversity in day and night operation.