Research On Driving Test System Based On VR

As social economy develops rapidly,and travel modes for people have undergone tremendous changes,automobiles have become one of the most important means of transportation in the modern era.Increases in the demand for automobiles has led to the development of traditional driving training.However,insufficiency in training resources including the lack of enough training cars and too much time cost has become increasingly evident.To solve problems mentioned above,there has already been a nonimmersive virtual driving training mode that is able to provide students with realistic three-dimensional driving scenarios and is simple,convenient,which proffers new perspectives to solve the discussed problem of driving training.Nevertheless,this method has got stuck in a dead end because of the lack of coaches who should have guided students in a timely manner.Moreover,if poor driving habits were difficult to be corrected in time,it would bring hidden troubles for people driving in the future.For the problems that arise in the non-immersive virtual driving training,this paper will propose some improvement measures aimed at increasing the efficiency of students acquiring driving skills and at the same time preparing them right driving habits.The main work and contributions of this paper are as follows: 1.This paper will present a virtual reality-based driving test training system.Solving the problems of non-immersive virtual driving training systems by combining a VR helmet,a Logitech G29 steering wheel,and a Unity3 D platform. The system can better assist students in learning driving skills,increasing learning efficiency,and correcting improper driving behaviors.2.The trajectory of parallel parking will be replanned by MATLAB and be mapped to the virtual driving test training system.In this paper,the trajectory of parallel parking is modeled,and after MATLAB simulation analysis,the trajectory will be replanned.It will also be mapped to the driving test training system to improve learning efficiency in the parallel parking process.3.A decision model for information fusion driving behavior based on machine learning has been proposed.The model can predict the driving behavior of the collected information and compare it with the student's driving behavior.According to the system judgment results,the trainees adjust their driving behaviors in time to form a good driving habit.Through systematic verification and analysis,the virtual reality-based driving test training system proposed in this paper can solve problems in the training process of non-immersive virtual driving tests,assist students in better driving learning,and can help students form safe and right driving habits.