基于 Bayes 判别方法的舰载机着舰安全分析

摘要

The aircraft-deck landing process is more complex and dangerous than land-based aircraft lan-ding,and the landing safety is a significant challenge to carrier pilots,aircraft commanders and engineers.The multi-dimensional state space model is extended.The normality test on F-18A′s aircraft-deck landing data for CVN-65 aircraft is conducted,and the safety state discrimination method of aircraft-deck landing is proposed with Bayes discriminant analysis.And then based on the F-18A′s landing data,this method is applied to dis-criminate the safety states of aircraft-deck landing.The discrimination result is basically consistent with the ac-tual situation,with the correct rate higher than 80%,which demonstrates the effectiveness of the method.Al-so,the discrimination result is compared with the result of Fisher discriminant analysis,showing that the pro-posed method has stronger practicality in the condition of a few parameters.%舰载机着舰过程比陆基飞机着陆过程更具复杂性和危险性，舰载机着舰安全是对舰载机飞行员、航母指挥人员和机舰系统工程师的重要挑战。本文扩展了多维安全状态空间模型的内容，并在此基础上针对F-18A舰载机在 CVN-65“企业号”航母上的真实着舰数据进行正态性检验，结合 Bayes 判别分析，提出舰载机着舰的安全状态判别方法。通过真实着舰数据检验，证明舰载机着舰安全状态判别结果正确率达到80％以上，验证了方法的有效性。实验进一步将判别结果与 Fisher 判别结果进行了对比，证明在较少参数情况下本方法具有更强的健壮性。