Metamorphic test method for automatic driving system

Abstract

The invention discloses a metamorphic test method for an automatic driving system, which is characterized in that test cases are expanded on the basis of existing automatic driving system test cases according to metamorphic rules, newly generated test cases and original data are input into an automatic driving system together, and the performance of the automatic driving system is observed, so that the problem of insufficient test is solved. Three parts ofa metamorphic test case generation module, an automatic verification module and a test report generation module are included. The input is the original test picture and the test case which is expandedbased on the metamorphic rule, the original test picture and the test case are sequentially sent to the automatic driving system, the difference of output results is compared, and possible defects of the system are judged. The method has the beneficial effects that the problems of insufficient test cases of the automatic driving system and difficulty in test predication collection are solved, meanwhile, the test loss can be reduced, and the product research and development period is shortened.

Description

technical field [0001] The invention belongs to the field of automatic driving software testing, and in particular relates to the definition of automatic driving test criteria. According to the metamorphosis rules, the test cases are augmented based on the existing automatic driving system test cases, and then the newly generated test cases are input into the automatic driving system together with the original data, and the performance of the automatic driving system is observed to solve the problem of insufficient testing. Background technique [0002] A traditional car tests its performance, but a self-driving car will have other dimensions, because it is an intelligent body similar to a person. In self-driving cars, we call it the autonomy of driving, that is, the driving ability of the vehicle itself under certain external conditions or environments. Therefore, automatic driving system testing is an important part of ensuring the safety of autonomous driving vehicles. ...

AI Technical Summary

Problems solved by technology

However, autonomous driving testing has always been a very cumbersome task, so many researchers have devoted themselves to the related work of autonomous driving vehicle testing, trying to improve the robustness of autonomous driving systems by studying automated testing methods

However, most of the testing work needs to contain test oracles as a premise. In many real situations, there are no test oracles. How to automatically test such systems that lack test oracles has become a big challenge.

Images