Abstract:

With the increasing of air traffic flow, high accuracy and realtime shortterm conflict detection is needed to ensure flight efficiency and safety. A probabilistic conflict detection algorithm is proposed based on hybrid system and position space discretization. The state-dependent stochastic linear hybrid system is used for trajectory prediction based on track and intent information of surveillance information. The predicted trajectory is a variable of multiple Gaussian distribution. To calculate the conflict probability between aircrafts, a computational efficient algorithm is developed by the relative aircraft position space discretization. Simulation shows that the algorithm is much more efficient than the cumulative distribution function approximation and Monte Carlo method. It meets the requirement of real-time conflict detection for aircrafts in high flight density airspace.