PILOTS: A SYSTEM TO PREDICT HARD LANDING DURING THE APPROACH PHASE OF COMMERCIAL FLIGHTS

Abstract

By performing a go-around, more than half of all business aeroplane operation errors may have been avoided. Making a prompt choice to do a go-around manoeuvre may help to lower the overall accident rate in the aviation industry. In this paper, we define a cockpit-deployable equipment learning system to support flight staff decision-making for a go-around based on the forecast of a difficult touchdown event. In order to forecast challenging touchdowns, this work offers a hybrid approach that uses attributes that model the temporal dependencies of aircraft data as inputs to a semantic network. Based on a large dataset of 58177 commercial flights, the findings indicate that our technique has an average level of sensitivity and uniqueness at the go-around point of 85% and 74%, respectively. It follows that our strategy outperforms other approaches and can be deployed in the cockpit.