Method and system for optimizing automatic driving failure scene library, server and medium

Abstract

The invention provides a method and a system for optimizing an automatic driving failure scene library, a server and a storage medium. The method comprises the following steps: analyzing failure eventdata recorded in an automatic driving road test process; determining the failure weight ratio of each factor; comprehensively obtaining the complexity of each test scene by using an analytic hierarchy process; and determining the importance degrees of different scenes to a test function and the priorities of different test scene cases. Therefore, the optimization of a test scene library is realized. The method has important significance for the design, test implementation and evaluation after test of test scene cases. Through hierarchical classification of each factor, the failure weight ratio of each factor can be calculated more accurately.

Description

technical field [0001] The present invention relates to the technical field of failure testing in automatic driving, in particular to an optimization method and system, a server and a storage medium for an automatic driving failure scene library. Background technique [0002] Driving scene data is an important resource for the R&D and testing of intelligent connected vehicles, and is the key data basis for redefining the level of intelligent vehicles. The scenario library is used as a set of use cases for ICVs. Due to the continuity of the distribution of scene parameters and the diversity of permutations and combinations of scene factors, the number of test scenario use cases is also unlimited. Therefore, it is particularly important to evaluate the test scenarios, calibrate the complexity of each scenario, and determine the priority of testing. Contents of the invention [0003] In view of this, an embodiment of the present invention provides an optimization method, sys...

AI Technical Summary

Problems solved by technology

[0003] In view of this, an embodiment of the present invention provides an optimization method, system, server, and storage medium for an automatic driving failure scene library to solve the problem of using different arrangements and combinations of scene elements in the construction of existing automatic driving test scene use cases. The number of scenario use cases is increasing proportionally, and there is no corresponding system method and standard for the complexity assessment of each scenario use case, and technical issues such as the priority and importance of the scenario use cases cannot be clarified during the test

Images