| Inertial/stellar integrated navigation sysytem is widely used in military and civil field because of its autonomous navigation,high accuracy and anti-interference capability.Aiming at the application under strong background noise and taking overcoming the obstacle of daytime stellar detection as the key point,this paper concentrate on the star images enhancement and inertial/stellar integration error suppression.The concrete content is as follows:(1)The attitude correlated frames adding approach(ACFA)is proposed to improve the detection capability of dim stars under strong noise.It utilizes the information of the inertial navigation system(INS)as an assitantant to improve the signal-to-noise ratio(SNR)of star images.The simulation and experimental results proves that the SNR of n frames processed by ACFA approach increases to n ? 2 n times.It also improves the star centroid accuracy.The impact of gyroscope error with navigation or tactical grade,and fixed angle error with several hundreds arc-seconds scale can be ignored when only tens of images are correlated and added.(2)The advantage of the short wave infrared imaging system for daytime stellar detection is analysed.According to the noise character of the actual infrared star images,several algorithms are selected and combined to preprocess these images.As a result,the stripe non-uniformity is eliminated.Besides the defective pixels can be successfully distinguished and compensated.(3)A novel inertial/stellar framework based on deeply-coupled algorithm is proposed to deepen the inertial/stellar information fusion.It is proved by simulation and experiment that when the navigation stars are less than 3 with a small filed of view the deeply-coupled method has higher accuracy on attitude,position and gyroscope bias estimation. |
PDF Full Text Request