基于概率模型的ATC系统冲突目标生成算法

Probability-Based Method of Generating the Conflict Trajectories for ATC System

  • 摘要: 通常用于测试空中交通管制(Air Traffic Control, ATC)自动化系统的飞行冲突告警功能的方法主要有放宽系统告警值和向系统输入模拟的飞行冲突目标的雷达数据。前一种方法存在不可靠性,第二种方法由于只产生简单的确定目标轨迹数据,因此只能简单地测试系统能否告警,无法对系统的飞行冲突预测能力作出评价。为了使用于测试系统的模拟雷达数据更符合实际飞行情况,并检测系统预测飞行冲突的技术水平,本文提出了一种基于飞行冲突概率模型的航迹模拟方法,通过对不同目标对的冲突概率建立数学模型,反推出存在不同冲突概率时目标的运动参数。并通过仿真实验产生了一系列具有不同冲突概率值的目标对的航迹,利用这些数据能对ATC自动化系统进行更有效的冲突探测和告警能力测试。

     

    Abstract: For testing the capability of short term conflict alerting of air traffic control system, two methods are usually used. The former is to set a higher threshold, use the real data testing whether the system can alert when distance between two flights gets lower than the threshold. However, this method is not reliable. The second method is simulating flights which will conflict and obtain their trajectory from calculating, and then send these data to ATC system to see its reaction. This method is usually too simple to test whether the system can pre-detect a conflict effectively. To solve these problems, a probabilistic approach is used in this paper to simulate aircrafts with given probability of conflicting. Firstly, we derived the conflict probability of turing flights from Prandaini’s method of conflict probability estimation for linear flight. Then using reverse derivation we got the motion parameters of two targets whose conflict probability was pre-setted. At last, we simulated this pair of targets’ track and anlysised their conflict probability. The simulation results show that the targets’ probability of conflict was in line with the previous assumption. The trajectories generated by this algorithm are more realistic then a more effective conclusion of ATC system’s capability of short term conflict alerting and pre-detecting will be provided.

     

/

返回文章
返回