Thesis Project

Currently, I’m a PhD candidate on the domain of software testing. My work is based on the Regression Testing Problem (RTP). A regression test is used to verify if all functionalities of the software work after an introduction of a new feature. This type of test can detect regressions on the source code. However, they usually require more time to execute, so if there are a large number of regression tests, it is necessary to chose which ones will be executed in each version. To resolve the problem, we can find three big approaches in the literature: prioritization, selection, and minimization.

This thesis is a collaboration between Renault Group and the laboratory LAAS-CNRS (Laboratory for Analysis and Architecture of Systems). At LAAS, I’m part of the group TSF (Dependable Computing and Fault Tolerance). Fernand Cuesta is the enterprise supervisor and Hélène Waeselynck is the academic supervisor.

The context of my theses is an automotive project at Renault Software Factory. The project is the development of a system for connected vehicles. The system is divided into two parts: the infotainment and the communication. The infotainment is the main point of contact between the conductor and the vehicle. It is represented as the multimedia panel and all its functionalities, like the Bluetooth, the AC and the navigation. The communication is responsible for the connection between all the components of the system and also between the onboard and offboard components. For example, there is the communication between a smartphone and the vehicle, so the doors can be looked and unlocked.

My work is to find the best way to prioritize/select the regression tests that will be executed in each version of the system. The objective is to find the bugs as soon as possible, so they can be corrected earlier in the process.