Strap-down inertial navigation system (SINS) is a kind of advanced inertial navigation system, which put gyroscopes and accelerometers solid to connect directly with carrier to replace the traditional inertial navigation system and have series of advantages such as small volume and weight,easy installation and maintenance, so that SINS is the main research direction in recent years. Initial alignment technology has a direct impact on the work performance of the SINS, and initial alignment error is one of the important error sources of the SINS, so the research of initial alignment is particularly important.Alignment accuracy and alignment time are two important technical indexes when initial alignment proceeds on inertial navigation system. Initial alignment accuracy affects the performance of inertial navigation system, initial alignment time marks the rapid response capability, so request the initial alignment accuracy is high and time is short. In order to meet the requirement, the gyro and accelerometer must have high accuracy and stability, the system robustness is better and is not sensitive on the outside of interference. Initial alignment is one of the most important key technical and the research hotspot of scholars at home and abroad in recent years.Recently, kalman filtering method in the study of the SINS is widely used and also the most mature method. It use kalman filtering algorithms to estimate misalignment horns thus to correct attitude matrix through building state space model of initial alignment to get the state equation and observed equation then setting the initial value.The research and application field of the dynamic base initial alignment of the SINS is wider. In modern warfare, shortening the reaction time of weapon system and improving its hit precision is very important, so that dynamic base transfer alignment become one of the key topics in recent studies.In the study of the dynamic base initial alignment, based on dynamic base transfer alignment, the velocity error model, posture error model and inertial components error model are derived. Based on different parameters matching method, the corresponding state equation and observed equation are built, so that using kalman filtering method to estimate the influence of misalignment horns of different parameters matching method, simulate corresponding graph and draw a conclusion through comparing them.In the study of the big misalignment horns initial alignment, the posture error model and velocity error model of the SINS based on additive quaternion error are derived. This paper introduces the extended kalman filtering method and the unscented kalman filtering method and builds the filtering model and uses different filter method respectively to simulate it thus to compare the simulation results of the two methods. |