RJMCMC粒子滤波方法在月球探测器自主天文导航中的应用  

An Autonomous Celestial Navigation Method for Lunar Probe Based on RJMCMC Algorithm Paticle Filter

作　　者：宁晓琳[1] 房建成[1] 

机构地区：[1]北京航空航天大学第五研究室,北京100083

出　　处：《宇航学报》2005年第B10期39-44,共6页Journal of Astronautics

基　　金：国家自然科学基金(:60174031);"863-703"(2003AA735101)"

摘　　要：天文导航系统是典型的非线性和噪声非高斯分布的系统。针对传统的扩展卡尔曼滤波不适于非线性和噪声非高斯分布的系统,和一般粒子滤波存在的粒子退化等问题,提出了一种将RJMCMC(可逆跳转马尔可夫链蒙特卡罗)算法应用于月球探测器自主天文导航粒子滤波器中的新方法。计算机仿真结果显示了该方法在加快收敛速度、提高导航定位精度和自适应调整粒子个数方面的有效性和可行性。Autonomous celestial navigation system is a typical nonlinear, non-Gaussian dynamic system. Because the extended kalman filter method is not good at dealing with nonlinear and/or non-Gaussian problems and the conventional particle filter method exists particle degeneration problem. A new autonomous celestial navigation method for lunar probe based on RJMCMC（Reversible - Jump Markov Chain Monte Carlo）method is presented in this paper. A simulation result demonstrated the validity and feasibility of this new method in quickening up the speed of convergence, increasing the precision of position determination and adjusting the sample size automatically.

关 键 词：月球探测器 自主天文导航 粒子滤波 RJMCMC 

分 类 号：V448.13[航空宇航科学与技术—飞行器设计]