论文部分内容阅读
飞行器高超声速飞行时,多普勒频移增大,GNSS捕获跟踪易丢失,CNS也会造成光线传播误差,使天体跟踪出现问题。提出了一种高超声速飞行器残差χ2-Fuzzy ARTMAP快速神经网络INS/GNSS/CNS组合导航故障检测方法,给出了高超声速飞行器INS/GNSS/CNS组合导航系统故障诊断原理,推导了残差χ2检验法的检测函数公式和Fuzzy ARTMAP快速神经网络原理算法,研究了残差χ2-Fuzzy ARTMAP快速神经网络组合导航故障检测实现方法。仿真结果表明该检测方法比传统残差χ2检验法更准确及时有效实现故障检测和隔离,并且得到的姿态、位置和速度信息精度更高,提高了临近空间高超声速飞行器INS/GNSS/CNS组合导航体系的精度和可靠性。
When the aircraft is flying at high speed, Doppler frequency shift increases, GNSS acquisition and tracking are easy to lose, and CNS also causes light propagation error, which causes problems in celestial tracking. A novel INS / GNSS / CNS integrated navigation fault detection method based on χ2-Fuzzy ARTMAP residual neural network for hypersonic vehicles is proposed. The fault diagnosis principle of INS / GNSS / CNS integrated navigation system is given. The residual χ2 The detection function formula of the test method and the fuzzy ARTMAP fast neural network algorithm, the implementation of the combined χ2-Fuzzy ARTMAP fast neural network navigation algorithm is studied. The simulation results show that the proposed method is more accurate and effective than the traditional χ2 test in fault detection and isolation. The attitude, position and velocity information obtained are more accurate and the INS / GNSS / CNS integrated navigation of nearby space hypersonic vehicles is improved System accuracy and reliability.