E-ploits: A system to predict hard landing during the approach phase of commercial flight

MCA

Over fifty percent of all airline accidents may have been avoided by doing a goaround. Making a timely choice to conduct a go-around manoeuvre has the potential of reducing the overall accident rate in the aviation sector. In this research, we propose a cockpit6 deployable machine learning system for assisting flight crew go-around making choices based on a hard landing event prediction. This paper describes a hybrid technique for hard landing prediction that employs features describing temporal relationships of aircraft characteristics as neural network inputs. Based on a huge dataset of 58177 airline flights, our technique offers an average sensitivity of 85% and a specificity of 74% at the go-around point. As a result, our method is a cockpit-deployable recommendation system that outperforms previous methods

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.