A virtual reality-based personalized driving habit training program generation method

Abstract

The invention relates to a method for generating a personalized driving habit training program based on virtual reality, which belongs to the technical field of virtual reality. A personalized training program is automatically synthesized through an optimization algorithm. Specifically, the user's visual data is obtained through the FOVE virtual reality helmet, the user's driving data is obtained through the Logitech driving simulator, and other vehicles and pedestrians on the virtual street are controlled through scripts. We designed a pre-evaluation experiment to obtain users' bad driving habits, such as turning without looking at the rearview mirror, crossing an intersection without slowing down, etc. Then, according to the pre-evaluation data of each user, a targeted training program is automatically synthesized. After the users have finished training, we run an evaluation experiment and compare it with the pre-evaluation experiment. Experimental results show that the present invention has a good effect on improving bad driving habits of users.

Description

technical field [0001] The invention relates to a method for generating a personalized driving habit training program based on virtual reality, which belongs to the technical field of virtual reality. Background technique [0002] With the introduction of a large number of consumer-grade virtual reality devices, virtual reality technology can simulate the required real scenes at a lower cost and in a smaller space, and has significant advantages for some dangerous scenes, and virtual driving is one of them. application. One type of application of virtual driving technology is the performance evaluation of production vehicles by automobile manufacturers. This type of application relies on the simulation of high-precision vehicle motion. Through the simulation of vehicle motion, the performance of vehicles can be evaluated in extreme conditions of simulation. Another type of virtual driving application is to use virtual reality technology to help users conduct driving trainin...

AI Technical Summary

Problems solved by technology

There are three main problems in the current virtual driving training: (1) The current virtual driving training method has no difference in the training programs used by all users, and it is impossible to provide targeted guidance to users, that is, it is impossible to find bad driving habits of users and correct them. Correction, so the effect of its training is relatively limited

(2) Traditional virtual reality helmets and virtual drivers are mainly used for scene visualization and vehicle control. In this process, traditional methods do not pay attention to the user's driving-related data, such as user attention, vehicle control, etc. This makes the training items very limited, and it is impossible to judge whether the user's driving behavior is correct

(3) In the existing virtual driving training, the traffic events in the virtual city scene are relatively simple, and most of them use fixed animation to play, and it is impossible to truly interact with the vehicle driven by the user, which makes the training lack of authenticity

Images