COSMOS (EU Horizon 2020)
COSMOS (DevOps for Complex Cyber-physical Systems) is designing and developing novel DevOps methodologies, techniques, and tools that will enable effective, continuous development and evolution of cyber-physical systems.
For more information, please check out the official project website.
In the context of COSMOS, I focused on monitoring and testing of cyber-physical systems in virtual environments. As a cyber-physical system use case I experiment with self-driving cars in simulation and with small prototypes in the real world. I developed SDC-Scissor, an open-source tool that leverages regression testing concepts (e.g., test selection, prioritization) for testing self-driving cars more efficiently in simulation environments. If you are curious about the tool then you can checkout the following links:
- Documentation: https://sdc-scissor.readthedocs.io/en/latest/
- Repository: https://github.com/ChristianBirchler/sdc-scissor
- Discussion: https://github.com/ChristianBirchler/sdc-scissor/discussions
For future work, SDC-Scissor will be extended and maintained. The goal is to achieve an impact for practioners in industry and researchers who are interested in efficient testing of self-driving cars.
SDC-Scissor is a tool that let you test self-driving cars more efficiently in simulation. It uses a machine-learning approach to select only relevant test scenarios so that the testing process is faster. Furthermore, the selected tests are diverse and try to challenge the car with corner cases.
For more details, take a look at the following sites:
This application is meant to collect all words in function and method names of Python and Java repositories on GitHub. Reactant’s implementation is based on a producer-consumer architecture with different docker containers. Furthermore, the use of Docker Compose makes the deployment of the tool trivial.
Ticket Tagger Analysis
Issue labeling on GitHub is usually done manually by the developers. In order to automate this process a tool name Ticket Tagger was developed. It classifies the issues on GitHub by a fasttest classifier. Ticket Tagger is a machine learning-driven issue classification bot. It was written by Rafael Kallis in the scope of a project similar to this one. Once installed in a GitHub repository, Ticket Tagger offers the benefit of automatic issue classification. Small repositories may not gain much value from it, but larger ones do since they receive more issues per time unit.
Santorini Board Game
We implemented the Santorini Board Game as a Web app with dedicated repositories for the frontend (React) and backend (Springboot).