【作者】 廖鹤；

【导师】 林晓辉；

【作者基本信息】 哈尔滨工业大学 ， 飞行器设计， 2007， 硕士

【摘要】 近年来,交会对接、空间攻防、星际探测成为航天研究的热点,长时间、大范围的轨道机动成为必然。因此,实现空间机动目标的自主导航,并要求其自主性强、抗干扰、可靠性高,成本低是面向我国航天控制发展的实际提出的挑战性新课题。SINS导航系统和GPS导航系统各有优缺点,但在误差传播特性上正好是互补的,前者长期稳定性差,但短期稳定性好。所以采用组合导航技术将这两个子导航系统有机地组合起来,克服各自的缺点,互相取长补短,能够实现轨道机动航天器的实时、高精度的导航定位。对于轨道机动航天器而言,以地理坐标系或地固坐标系为导航坐标系的SINS系统模型并不适用,其导航参数应该是在地心惯性坐标系下的。针对这种情况,本文采用基于地心惯性坐标系的SINS系统,并建立了系统在地心惯性坐标系下的位置、速度、姿态误差传播方程。将误差传播方程与C/A码GPS接收机的误差方程组合,建立起轨道机动航天器的SINS/GPS组合导航系统模型,利用集中式卡尔曼滤波进行优化处理,并通过数学仿真验证该组合导航算法的可行性。建立的组合导航模型的状态变量为48维,计算速度慢。利用奇异值分解法进行可观测性/可观测度分析,降维,并对降维后的组合导航系统进行数学仿真,与降维前的仿真结果相比较来验证可观测性/可观测度结果的正确性以及降维后的组合导航系统方案的可行性。更多还原

【Abstract】 Recent years, Rendezvous and Docking, space attack and defense, space detection is the focus of astronautics research. It results in space vehicle doing long-time and extensive orbit transfer. Autonomous navigation of orbital transfer vehicle which has the characters of independence, high reliability and low cost is an attractive and challenged subject of Chinese astronautics.The SINS navigation system and GPS system each has its own advantage and disadvantage. But the error propagation property is complementary. Long time stability of fist is bad, but its short time stability is good. So the combined navigation system which can compensate each other can realize real-time and high accuracy of position fixing for orbit transfer vehicle.For orbit transfer vehicle, it’s not suitable that using geographic coordinate system or terrestrial coordinate system as navigation coordinate system. Its navigation parameters should be in geometric inertial coordinate system. In this study, geometric inertial coordinate system is navigation coordinate system. And the position, velocity and attitude error propagation equation is established in geometric inertial coordinate system.SINS/GPS navigation system of orbit transfer vehicle is established by making the equation combination with error equation of C/A code GPS receiver. To validate the feasibility of this algorithm through arithmetic simulation by using centralized Kalman filter.The dimension of state equation is 48. It could cause the lowering of calculated speed. So it’s adopted that using singular value decomposition method to analyze the observability and degree and descend the dimensions of combined navigation system. And make the arithmetic simulation of combined navigation system which the dimensions is descended. Then contrast with undescended navigation system to test and verify the validity of observability and its degree results and the feasibility of combined navigation system which is descended.更多还原