Grant: Discovery Grants Program - Individual
University of Waterloo
$50,000.00
Pilot performance assessment using both data-driven and theory-driven computational models
EF 2024-25
Résumé
Aviation is an important mode of transportation for Canadians, but the aviation industry is threatened by a pilot shortage (CBC News, 2023), due to the long hours of training and the competitive global market. Thousands of flight hours are needed before a pilot can qualify for airline jobs based on current training methods that do not use much of information technologies. Recent advancements in machine learning and cognitive modelling provide an opportunity to innovate and transform pilot training, but previous research has not integrated the two approaches to gain their complementary benefits, and there is no pilot performance database for researchers. Data-driven machine learning models are better for practical applications but lack explainability for people to understand the models' reasoning and logic. Theory-driven cognitive models can explain the knowledge and skills that cause differences in human performance but require more work from researchers to build and validate the models. Taking the advantages of both types of models, the long-term objective of this project is to develop and validate an automatic and objective approach for pilot performance assessment using flight data, integrated computational models, and skill diagnosis tools. A database will be built and accumulate data over time to store pilot performance data with instructor rating labels. This long-term goal is supported by the PI's unique strength in combing the two complementary modelling approaches. The short-term objectives are to design an integrated approach, collect data, and examine the benefits of the new approach through the modelling of representative flight tasks from general aviation training. Obj 1: Collect about 300 hours of flight data, which is needed to provide a large enough sample size with variations in factors such as pilot experience, flight task, and weather condition, which is essential for machine learning and pilot skill analysis. Obj 2: Train machine learning models and examine their predictive power. Obj 3: Build and validate cognitive architecture models for pilot performance. Obj 4: Generate synthetic pilot data using cognitive models and test it as a new method of data augmentation for machine learning. Obj 5: Support pilot skill diagnosis and model explainability by testing both machine learning and cognitive models built to represent how performance changes with experience. This project will create a database for the research community and aviation industry to train and validate models. The current work will enable new pilot training technologies to provide more objective and equitable training for the next generation of aviation professionals and make training more effective and efficient, addressing the pilot shortage issue. Students will receive training on aeronautics knowledge, modelling techniques, and inclusive practice. They will be highly qualified for research and development jobs in the aviation sector of Canada.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
RGPIN-2024-04808
Name nom
Cao, Shi
Department département
Systems Design Engineering
Organization id
33
Institution établissement
University of Waterloo
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
50000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1509
Committee name en
Civil, Industrial and Systems Engineering
Committee name fr
Génie civil, industriel et des systèmes
Area of application code
700
Area of application group en
Transportation systems and services
Area of application group fr
Services et systèmes de transport
Area of application en
Transportation systems and services
Area of application fr
Services et systèmes de transport
Research subject code
1605
Research subject group en
Industrial engineering
Research subject group fr
Génie industriel
Research subject en
Human factors engineering
Research subject fr
Génie des facteurs humains
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Pilot performance assessment using both data-driven and theory-driven computational models
Keywords
human factors; human performance modelling; pilot training; aviation human factors; flight simulator; cognitive models; pilot performance models
Application summary
Aviation is an important mode of transportation for Canadians, but the aviation industry is threatened by a pilot shortage (CBC News, 2023), due to the long hours of training and the competitive global market. Thousands of flight hours are needed before a pilot can qualify for airline jobs based on current training methods that do not use much of information technologies. Recent advancements in machine learning and cognitive modelling provide an opportunity to innovate and transform pilot training, but previous research has not integrated the two approaches to gain their complementary benefits, and there is no pilot performance database for researchers. Data-driven machine learning models are better for practical applications but lack explainability for people to understand the models' reasoning and logic. Theory-driven cognitive models can explain the knowledge and skills that cause differences in human performance but require more work from researchers to build and validate the models. Taking the advantages of both types of models, the long-term objective of this project is to develop and validate an automatic and objective approach for pilot performance assessment using flight data, integrated computational models, and skill diagnosis tools. A database will be built and accumulate data over time to store pilot performance data with instructor rating labels. This long-term goal is supported by the PI's unique strength in combing the two complementary modelling approaches. The short-term objectives are to design an integrated approach, collect data, and examine the benefits of the new approach through the modelling of representative flight tasks from general aviation training. Obj 1: Collect about 300 hours of flight data, which is needed to provide a large enough sample size with variations in factors such as pilot experience, flight task, and weather condition, which is essential for machine learning and pilot skill analysis. Obj 2: Train machine learning models and examine their predictive power. Obj 3: Build and validate cognitive architecture models for pilot performance. Obj 4: Generate synthetic pilot data using cognitive models and test it as a new method of data augmentation for machine learning. Obj 5: Support pilot skill diagnosis and model explainability by testing both machine learning and cognitive models built to represent how performance changes with experience. This project will create a database for the research community and aviation industry to train and validate models. The current work will enable new pilot training technologies to provide more objective and equitable training for the next generation of aviation professionals and make training more effective and efficient, addressing the pilot shortage issue. Students will receive training on aeronautics knowledge, modelling techniques, and inclusive practice. They will be highly qualified for research and development jobs in the aviation sector of Canada.