Applying and Evaluating Automated Testing for Triple-Ended Systems

The increasing complexity of Internet of Things ecosystems, especially in triple-ended systems (systems involving a backend, frontend, and a third end such as Bluetooth), calls for efficient and scalable automated testing approaches. This thesis investigates the application of the Patrol testing framework for Flutter crossplatform mobile applications to automate functional and integration tests in such systems, with a focus on Bluetooth-driven Firmware Over-the-Air updates. The research evaluates Patrol’s effectiveness in automating regression tests, reducing manual effort, and enhancing test reliability across interconnected components. Key findings include the successful automation of 70% of Thule’s regression test suite, significant time savings, and improved validation consistency. The study also provides practical insights into applying test automation in real-world industrial environments and discusses limitations related to hardware interactions and external dependencies. The results contribute to more efficient testing practices for cross-platform and IoT applications, offering a foundation for further research in automated testing of interconnected systems.