基于支持向量机的惯导初始对准系统

针对平台式惯导系统的初始对准 ,提出了基于支持向量机 (SVM)网络的滤波器 ,并将其用于惯导初始对准中 ,代替初始对准系统中的闭环Kalman滤波器 ,可以确保系统的误差状态始终为小量 ,实现了惯导初始对准中的滤波与校正功能 ,并将其与神经网络滤波器、闭环Kalman滤波器进行对比。仿真结果表明 ,采用SVM网络滤波器简化了系统运算的代数结构 ,提高了系统状态估值运算的实时性 ,而对准系统的精度又与采用闭环Kalman滤波器的精度相当 ,并且SVM网络克服了神经网络的不足。

Initial Alignment System of Inertial Navigation Based on Support Vector Machine

Based on the support vector machine (SVM) a filter is developed for the inertialnavigation's initial alignment. SVM network instead of the closed loop Kalman filter in the initial alignment can keep the error small, and implement the function of estimation and alignment in the inertial navigation. Simulation results show that it is simpler for the system's algebraic structure and more attractive for real time than classical filters, and its precision is similar to that of the kalman filter. SVM 's performance is better than neural network's.