Concolic Testing for High Test Coverage and Reduced Human Effort in Automotive Industry
The importance of automotive software has been rapidly increasing because software now controls many components in motor vehicles such as window controller, smart-key system, and tire pressure monitoring system. Consequently, the automotive industry spends a large amount of human effort testing automotive software and is interested in automated software testing techniques that can ensure high-quality automotive software with reduced human effort. In this paper, we report our industrial experience applying concolic testing to automotive software developed by Hyundai Mobis. We have developed an automated testing framework MAIST that automatically generates the test driver, stubs, and test inputs to a target task by applying concolic testing. As a result, MAIST has achieved 90.5% branch coverage and 77.8% MC/DC coverage on the integrated body unit (IBU) software. Furthermore, it reduced the cost of IBU coverage testing by reducing the manual testing effort for coverage testing by 53.3% which correlates to 0.5 million USD per year at Mobis India.