| Summary: | Today, Air Traffic Controllers (ATCO) plays an important role in aviation safety. The nature of Air Traffic Control (ATC) is fast paced and requires high concentration where lapses in concentration can result in fatal accidents. By measuring Human Factors Variables such as Workload and Situation Awareness (SA) of an ATCO, preventive measures can be taken before errant human mental states lead to accidents. In order to do this, however, detection and measurement of human factors has to be on a real-time basis which cannot be done by existing methods. Currently, the existing methods to measure Workload and SA either require ATCOs to perform secondary tasks that can be intrusive to their work or are questionnaires- based done on a post-activity basis. Physiological approach such as using Electroencephalography (EEG), on the other hand, is non-intrusive and can measure human factors variables on a real-time basis. Despite the huge potential of EEG to overcome the short-comings of current methods, currently there is only little research done on topics related to implementing EEG for evaluating SA. The aim of this project was therefore to develop an algorithm to assess real-time SA using EEG data in order to address the limitations of current traditional methods.
This project first proved the hypothesis that there is a negative relationship between Workload and SA. This project then made use of Machine-Learning to develop an algorithm to predict SA of ATCO using EEG data and it was found to be as reliable as traditional methods after validating with the proven hypothesis.
|
|---|