Injury Prediction Of Driver And Passenger In Small Overlap Collision And Design Of Notification Terminal

Vehicle collision safety technology can reduce or avoid the injury to passengers and pedestrians.Therefore,vehicle technicians have done a lot of research on the safety performance of vehicles.Advanced automatic crash notification(AACN)is a technical means of post-accident safety.Compared with the traditional vehicle accident call system,AACN system can predict the degree of injury of drivers and passengers on the basis of external call so that the rescue center can reasonably allocate medical resources and formulate targeted rescue plan according to the actual situation of the passengers in the vehicle.In traffic accidents,small overlap collision is a special type of frontal collision.When impacted by the impact force,the collision energy is directly transmitted to the passengers because it avoids the main energy absorbing parts of the front end of the car,which poses a great threat to the lives of the passengers.However,the existing AACN system does not take into account the small overlap collision situation,so it is of great practical significance to establish a small overlap driver and passenger injury prediction algorithm.In this paper,the prediction algorithm of driver injury degree was developed.First,the accident data of small overlap collision in the National Highway Traffic Safety Administration were counted,and then the factors in the accident information that will affect the driver injury degree were analyzed.After determining the relevant factors,the BP neural network was used to build the prediction model.In order to verify the reliability of the model and screen out the best cut-off value of the model,the model was analyzed by ROC.Finally,on the basis of verifying the effectiveness of the model,a prediction algorithm of driver injury degree in small overlap impact was established.The prediction algorithm of rear occupant injury degree was also developed.First,according to the occupant side small overlap crash test rules issued by China insurance vehicle safety index,the barrier and ground models were established by using finite element simulation software,and the dummy model was positioned and adjusted.Then,according to the output results of the simulation model,the reliability of the model was verified from three aspects: the collision energy curve of the model,the deformation of the whole vehicle in the simulation and test diagrams,and the calibration of the parameter curves of the dummy’s head and chest.On the basis of verifying the validity of the model,different initial speeds were set for the established simulation model to obtain the corresponding dummy head,neck and chest damage values,and the correlation model between the speed change and the damage value was constructed.Finally,according to the correlation model,a prediction algorithm for the damage degree of the rear passengers in small overlap impact was established.In the design of AACN system terminal,the first step was to design the hardware of the system terminal.In order to realize the functions of vehicle positioning and sending information,the terminal was equipped with display module,positioning module and communication module.Then,the software of the system was designed,and the pre calculation method of driver and passenger injury degree was integrated into the AACN system.Finally,the real case and simulation data were used to verify the reliability of the system terminal.The results of case study show that the AACN system terminal can recognize small overlap collision and predict the degree of driver’s and passenger’s injury,and the established algorithm is reliable.