A Combination Of Navigation Techniques Based On Real-time Multitasking System

In the INS/GPS integrated navigation system, Kalman filtering is an information fusion method used to combine information of two positioning systems to achieve the best result of integrated navigation. This paper mainly introduces the application of Embedded Linux in integrated navigation computer. Embedding Linux system, real-time serial communicating in Linux, INS/GPS integrated navigation system and multi-thread have been researched in this dissertation. The major researches can be divided into three parts:In the first part, based on the requirement of vehicle integrated navigation system, embedding navigation computer and embedding Linux by means of recompiling kernel is researched. And presents some methods of developing program of integrated navigation. In the second part, the method of serial communication in embedded Linux is researched. And realizes real-time communicating in the environment of multi-task. In the third part, we build the Kalman filtering model of INS/GPS integrated navigation system, and realize the filtering model by multi-thread programming in Linux. In the last part of this dissertation, tests prove that the program developed in embedded Linux can completely arrive at the real-time demand of the Kalman filtering in the integrated navigation system.