| With the continuous development of vehicle technology,the way of humancomputer interaction and interactive content are undergoing revolutionary changes.At present,the content of on-board human-computer interaction is gradually evolving from a single vehicle status information to complex information including vehicle status,assisted driving and entertainment.Head up Display(HUD)technology is also gradually applied to household cars.In order to pursue the improvement of driving assistance function and visual effect,the major automobile manufacturers constantly enrich the HUD display function.In a complex road environment,drivers not only need to deal with the complex external traffic environment,but also need to understand the meaning of HUD information,which leads to a sharp increase in the brain load of drivers,making it difficult for drivers to respond correctly to changes in the driving environment in a timely manner,which can easily lead to traffic accidents.In this paper,the visual attention allocation is used as a constraint to study the vehicle AR-HUD adaptive strategy considering the complexity of dynamic road environment.Firstly,this paper establishes a quantitative model of car-following dynamic road environment complexity.Based on the car-following behavior spectrum,three complexity quantitative indexes of modified collision margin reciprocal,lateral swing coefficient and velocity instability coefficient are selected to establish the model.Carfollowing scenes with different complexity were built on the driving simulator and participants were arranged to perform car-following driving tasks.Driving data such as front vehicle position,front vehicle speed,self-vehicle position and self-vehicle speed were collected.Calculate the specific value of the complexity quantitative index and use the entropy weight method to determine the index weight,and establish a quantitative model of the complexity of the dynamic road environment.The NASATLX scale was used to quantify the driver ’s driving load to verify the accuracy of the model.The results show that the quantitative results of the model are significantly correlated with the driver ’s driving load,which proves the effectiveness of the model construction.Secondly,the quantitative model of HUD interface complexity is established.According to the eye tracking model,the driver ’s eye movement process of recognizing HUD information is analyzed,and the scanning path time and average gaze time are selected as the eye movement model indicators.According to the eye movement indicators,the influencing factors of HUD interface complexity are analyzed.The existing HUD display information and the over-the-horizon information in the connected environment are analyzed.The HUD simulation is carried out on the driving simulator,and the eye movement behavior of the subjects when they recognize different HUD interfaces is collected.The law of eye movement index changing with influencing factors is analyzed and the relationship model between them is established.The eye movement time is used to represent the complexity of HUD interface,and the quantitative model of HUD interface complexity is established to realize the complex quantification of HUD interface for single element and multiple elements.Finally,the following adaptive HUD display strategy is developed.The TICC algorithm is used to segment and cluster the obtained driving data,divide the driving scene,and analyze the driving state and dynamic road environment complexity level of each scene.According to the characteristics of each scene,the corresponding HUD interface display content is designed,and the HUD information importance ranking and the HUD interface complexity limit value are determined.While ensuring that the driving information required for different scenes is satisfied,the total complexity of the vehicle information and the external environment of each scene is controlled.According to the display strategy,a car-following adaptive HUD display system is established.The system is evaluated by fuzzy comprehensive evaluation method,and the availability of the system is verified,which provides a theoretical basis for the research of vehicle human-computer interaction adaptive strategy. |
PDF Full Text Request