Real driving scene simulation method and system, electronic equipment and storage medium

Abstract

The invention discloses a real driving scene simulation method and system, electronic equipment and a storage medium, and the method comprises the steps: building a three-dimensional space map of a real scene based on a multi-sensor data collection platform, carrying out labeling separation on dynamic and static traffic elements in the real driving scene, outputting local space positions and types of dynamic and static traffic elements in a real scene, generating a simulation scene file, reproducing a real driving scene, and storing the real driving scene as a simulation scene library file which can be imported into a Carla simulation software platform. According to the method, data sources and channels are diversified, the test coverage and authenticity of the simulation scene library are effectively ensured, and the accuracy of global scene mapping is ensured by adopting a data tight coupling fusion mode. According to the invention, a data labeling tool is used for supporting the joint labeling, automatic interpolation labeling and other functions of multi-sensor data; and the continuity of the labeling tracks of the dynamic traffic elements is ensured, and finally the high-reduction-degree building work of the whole scene is realized.

Description

technical field [0001] The present invention relates to a simulation method, system, electronic equipment and storage medium, in particular to a real driving scene simulation method, system, electronic equipment and storage medium. Background technique [0002] In recent years, with the rapid development of autopilot technology, the corresponding autopilot function testing requirements have also greatly increased. Conventional software testing methods and real vehicle testing are far from meeting the rapidly growing testing requirements. Therefore, simulation autopilot testing is gradually become the mainstream testing method. The usual simulated autonomous driving test method requires manual construction of the test environment and design of various driving behaviors in advance, which is inefficient and not realistic enough to fully test the autonomous driving function. Although the tool chain based on commercialized simulation scene construction can quickly build a test s...

AI Technical Summary

Problems solved by technology

[0002] In recent years, with the rapid development of autopilot technology, the corresponding autopilot function testing requirements have also greatly increased. Conventional software testing methods and real vehicle testing are far from meeting the rapidly growing testing requirements. Therefore, simulation autopilot testing is gradually Become a mainstream testing method

The usual simulated autonomous driving test method needs to manually construct the test environment and design various driving behaviors in advance. The test is inefficient and not realistic enough to fully test the autonomous driving function.

Although the tool chain based on commercialized simulation scene construction can quickly build a test simulation environment as realistic as possible, due to the characteristics of charging and authorization, it leads to high cost of use and is difficult to promote and apply on a large scale. Moreover, the applicability is relatively fixed and does not support users. Custom modification, it is difficult to adapt to the rapid development of autonomous driving test requirements

[0003] Currently existing similar technical solutions are implemented based on commercial simulation scene construction software tools such as VTD or SCANeR. The data processing process is similar, and they all first perform three major processes: separation of dynamic and static traffic elements, simulation 3D scene construction, and dynamic traffic flow generation. The main disadvantage of the existing technology is: the simulation scene creation tool used is paid, and the scalability is weak, and it does not support users to customize functions according to actual needs

Images