Automatic driving network connection multi-vehicle testing method and system based on digital twinning

Abstract

The invention discloses an automatic driving network connection multi-vehicle testing method and system based on digital twinning. According to the invention, the digital twin environment is constructed in the simulator in a scene digitization mode, and the virtual vehicle and the virtual sensor are defined in the simulator, so that the virtual vehicle in the simulator can move according to the data of the sensor. Vehicles are divided into real vehicles and virtual vehicles, in the moving process of all the vehicles, specific data are uploaded to the digital twin cloud platform in a network TCP connection mode, due to the fact that the cloud platform regards all the vehicles as a scheduling node, the virtual vehicles or the real vehicles are subjected to consistent data management on the cloud platform part, and the data management efficiency is greatly improved. The cloud platform issues information of other vehicles to all connected vehicles through a certain period, so that all the vehicles can be communicated with one another, the virtual vehicle and the actual vehicle can obtain beyond-visual-range information of each other, and a basic form is provided for virtuality and reality combined automatic driving network connection multi-vehicle testing.

Description

technical field [0001] The invention belongs to the field of automatic driving testing methods, and in particular relates to a digital twin-based automatic driving network-connected multi-vehicle testing method and system. Background technique [0002] Autonomous vehicle testing is mainly divided into three steps, software-in-the-loop testing, hardware-in-the-loop testing, and vehicle-in-the-loop testing. The above test links are mainly defined for traditional single-vehicle intelligent autonomous driving. In today's environment where China is vigorously developing the autonomous driving of connected smart cars, many simulation software such as CARLA, AirSim, etc., are more focused on the autonomous driving of single-vehicle intelligence, that is to say, there is only one main car, The main car is equipped with a large number of sensors to test the performance of the vehicle in different cases, which cannot meet the current development trend of connected cars, and the test s...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem of high labor cost and time cost caused by multi-vehicle deployment in the process of multi-vehicle testing, the present invention builds a digital twin virtual environment and uses virtual vehicles to replace part of the actual vehicles, so that the virtual vehicles and the actual Vehicles coexist in the same scene through the digital twin environment, and a small number of actual vehicles can be used for networked multi-vehicle experiments, reducing the manpower and time costs in the process of multi-vehicle testing, improving the efficiency of multi-vehicle testing, and providing guarantee for actual road tests

Images