Method for building a virtual test platform for autonomous driving that maps real-world road conditions

Abstract

The invention discloses a method for building an automatic driving virtual test platform for mapping real-world road conditions. The steps are as follows: obtaining real-world road condition information; establishing a road environment module for mapping real-world road conditions; building an automatic driving vehicle module and an interactive vehicle module ; Integrate the road environment module, autonomous vehicle module and interactive vehicle module to realize information interconnection and build an autonomous driving virtual test platform that can map real-world road conditions. The present invention can quickly test the operating state of the automatic driving vehicle under real-world road conditions, and conveniently adjust the interactive vehicle driving trajectory, automatic driving perception sensor model and automatic driving function model, so as to meet the requirements of different automatic driving test conditions.

Description

technical field [0001] The invention belongs to the technical field of virtual testing of automatic driving, and relates to a method for building a virtual testing platform of automatic driving for mapping real-world road conditions. Background technique [0002] Due to the inexhaustible number of unknown test road traffic scenarios, compared with the virtual test method of autonomous driving, the test site or public road test method based on real vehicles has the disadvantages of long test period, uncontrollable test scene, limited manpower and material resources, and unguaranteed test safety. technical bottleneck. Therefore, virtual simulation testing has become an important substitute for real physical testing of autonomous driving under constrained economic conditions, and the virtual test platform is also an important tool for research related to autonomous driving. [0003] The road traffic system is a nonlinear complex system that is coupled among human-vehicle-road-...

AI Technical Summary

Problems solved by technology

[0004] Research on the existing virtual test platform for autonomous driving found that a single simulation platform cannot integrate complex road environment modules at the same time, including accurate road alignment description, adjustable road surface friction coefficient, unevenness and rich weather environment information; professional vehicles Dynamic models, physical model-based perceptual sensor models, vehicle models that can interact with autonomous vehicles, and autonomous driving function models. Existing co-simulation platforms cannot accurately build road environment modules that can map real-world road conditions

Images