Grant: Alliance Grants

The quality of flight data is a fundamental element in developing mathematical models for designing high-fidelity flight simulators. Unfortunately, several aspects can alter these data during flight tests, making them unusable or leading engineers to erroneous analyses. The main goal of the MAIDAY project is to create innovative data-testing tools by combining knowledge of flight physics and Artificial Intelligence (AI) techniques. These tools will help detect and identify any anomalies or inconsistencies in flight test data at an early stage, thus ensuring its accuracy and validity. From a scientific point of view, this problem can be considered a more complex time series anomaly detection problem involving many constraints and subjective criteria. Therefore, the studies conducted in this project will advance knowledge in this field by introducing new methods that use AI and flight physics to convert this big data problem into a "smart data" solution. MAIDAY project will contribute to the training of Highly Qualified Personnel (HQP) at all levels (MSc, Ph.D., and post-doctorate) and initiate undergraduate interns to research projects in one in one of the most important sectors of the aeronautics industry, as well as in the field of artificial intelligence, which now has significant importance in industrial and academic aerospace research. The methodologies and tools developed will allow industrial partners to considerably reduce the time and costs allocated to flight data analysis. When fully developed, MAIDAY will decrease the time needed to create a new performance simulator model by 10% and reduce the required flight-testing hours. In addition, developing solutions to reduce the number of flight tests is also an excellent way for industrial partners in Canada to demonstrate their efforts to reduce greenhouse gas emissions in the aviation sector.