Infotainment Validation Optimization for Global Automotive Projects

Automotive infotainment systems are becoming more and more complex, due to the increasing number of functionalities provided for the user. Indeed, the number of ECUs and electronic components has increased over the past decades, with a consequent radical change for both customer’s experience and HMI. Validation and testing are key aspects to improve the reliability of the infotainment ecosystem, stressing the components and observing their behaviour in those situations, which must be compliant according to the norms’ specifications. The testing operation can be accomplished in two ways: manually or automatically. Human testing is sometimes preferred, but for intense and repetitive testing, automated testing is well-suited. The tests can run on a larger time-frame without the need of a human. The present work concentrates in optimizing the ATB Editor tool powered by Stellantis, improving its OCR functionalities. More specifically, combining PP-OCR and Tesseract OCR to perform text detection and recognition instead of the current pattern matching. This will improve the software capabilities, exploiting deep neural networks and giving more flexibility during the test-writing phase.