Deep Learning for Situation Awareness in Airport Operations

This project that is funded by Northrop Grumman Corporation investigates the patterns of ground vehicles and aircraft trajectories, combining the analysis of their spatial movement behaviours and contextual information. We utilise state-of-the-art artificial intelligence and big data analytics techniques to clean, preprocess, and analyse sizeable on-ground aircraft GPS data. We also establish a context-aware system to predict the delay time of each aircraft by developing a novel Airport Traffic Complexity (ATC) model.

The outcome leads to multiple economic and security benefits for airline passengers, air traffic controllers, and airport managers. Our work provides a situational awareness map to the airport traffic controller, which reveals the potential factors that affect the traffic flow at the airport, as well as points out the hot spots and anomalous cases. Additionally, our proposed solution could inform airport managers and traffic controllers regarding the estimated delay time of flights on the tarmac at the specific airport.

Main Participants

Flora Salim
Associate Professor
Wei Shao

His interest research area focused on data mining, spatio-temporal data analysis and device-free activity recognition.