论文部分内容阅读
随着低空飞行密度不断增加,低空航行安全已引起广泛关注,由于低空环境复杂,低空飞行受地面障碍物和天气影响比商用航空显著,传统的空中交通警戒与防撞系统(TCAS,Traffic Alert and Collision Avoidance System)和其他冲突探测方法并不适用于低空密集飞行环境。针对传统探测方法计算量大、适用性差的不足,引入支持向量机(SVM,Support Vector Machine)的二元分类方法,通过对本机和周边飞机航迹归一化处理,采用智能优化算法对关键参数进行优化,利用模拟数据对分类器进行预先训练,实现了适用于低空飞行的高效冲突探测。最后以大量的仿造数据对算法有效性进行了测试验证,结果表明漏警率和误警率分别控制在0.1%和6%左右,克服了传统确定型方法与概率型方法难以兼顾效率与适用性的缺陷。
With the increasing density of low-level flight, low-level navigation and safety has drawn much attention. Due to the complexity of low-altitude environment, low altitude flight is more affected by ground obstructions and weather than commercial aviation. The traditional traffic alert and collision avoidance system (TCAS) Collision Avoidance System and other conflict detection methods do not apply to low altitude dense flight environments. Aiming at the shortcomings of large computation and poor applicability of traditional detection methods, a binary classification method based on Support Vector Machine (SVM) is introduced. Through the normalization of trajectories of local and surrounding aircraft, intelligent optimization algorithm is applied to the key parameters The optimization is carried out, and the classifier is pre-trained by using the simulation data to realize the efficient collision detection suitable for low-level flight. Finally, a large number of imitation data are used to test the validity of the algorithm. The results show that the false alarm rate and false alarm rate are controlled at about 0.1% and 6%, respectively, which overcomes the traditional deterministic and probabilistic methods that make it difficult to balance both efficiency and applicability Defects.