Collision Risk Research Based On Stochastic Differential Equation Of Brownian Motion In Free Flight

Free Flight is a flight mode to effectively solve the problem of air traffic congestion. In free flight, the flight path no longer bound by fixed air route(airline), the pilot independently select optimal flight path and is responsible for maintaining safety separation between aircraft.In order to ensure the safety of free flight, collision risk in free flight was demanded to assessment.Collision risk assessment based on stochastic differential equation of Brownian motion in free flight was studied in this thesis, the main work was as follows:1. The characteristics of free flight, collision risk and influence factors, target level of safety and collision risk assessment process were analyzed.2. In view of pilots being free to choose the flight path in free flight, the straight-line routes would increase, collision risk based on straight-line routes in free flight was studied.According to the position relationship between two straight-line routes in the same plane,the method based on stochastic differential equation of Brownian motion was used to establish the collision risk evaluation model respectively under three cases of the same route, parallel routes and cross routes.3. In view that the aircraft position errors caused by CNS factors in free flight was random, which was in line with Brown motion characteristics, and with CNS factor error distribution of 3-dimensional correlation concerned, the collision risk model of the aircraft movement stochastic differential equation in free flight was established. The combined method of Milstein method and finite difference method was used to solve the model.4. According to the occurrence of aircraft collision accident in free flight was the result of the interaction of aircraft conflict event, collision avoidance system warning failure event and pilot intervention failure event, the event tree model of aircraft collision was established.Firstly, stochastic differential equation method was applied to analyze the impact of CNS performance on conflict risk; Secondly, the set pair analysis method was applied to analyze the impact of collision avoidance system reliability on collision risk; Finally, the Bayesian network was applied to analyze the impact of pilot reliability on collision risk.