Lambert转移中途修正的全局概率最优策略  

Optimal correction strategy during Lambert transfer from view of probability

作　　者：徐明[1] 谭田[2] 李志武[2] 徐世杰[1] 

机构地区：[1]北京航空航天大学宇航学院,北京100191 [2]航天东方红卫星有限公司研发中心,北京100094

出　　处：《北京航空航天大学学报》2012年第5期574-578,共5页Journal of Beijing University of Aeronautics and Astronautics

基　　金：国家自然科学基金资助项目(11172020);工信部"唯实"人才培育基金资助项目;北京航空航天大学"蓝天新秀"专项基金资助项目

摘　　要：应用Monte-Carlo法和遗传算法的联合仿真求解Lambert转移中途修正的全局概率最优策略.首先推广限制性三体问题中求解周期性特解的微分修正算法构造出考虑J2项摄动下的Lambert转移轨道并以此作为参考轨迹,则中途修正策略仅需针对导航误差、初始偏差修正的控制偏差等进行补偿.应用微分修正算法导出的单值矩阵,设计出3类线性和非线性中途修正策略,以适应不同的精度需要.随后应用Monte-Carlo和遗传算法的联合仿真,可以得到实现代价函数(落点误差最小)在概率意义下的最优解.与直接利用优化算法寻优需要已知各种误差量不同,得到的最优修正策略更具有普适性.Monte-Carlo simulation and genetic algorithm were employed to achieve the optimal correction strategy during Lambert transfer from the view of probability.The Lambert trajectory associated J2 perturbation was generated from differential correction algorithm that was used to yield the periodic orbit in circular restricted three body problem（CR3BP）.So the correction maneuvers dealt with just the residual errors from navigation and controller.Then the linear and nonlinear strategies were derived from the monodromy matrix of differential correction algorithm,and then the Monte-Carlo simulation and genetic algorithm were employed to minimize the rendezvous errors from the view of probability.Quite different from the traditional strategies,the correction developed is universal for its robustness to measure errors.

关 键 词：Lambert转移 微分修正 中途修正 MONTE-CARLO法 遗传算法 

分 类 号：V421.41[航空宇航科学与技术—飞行器设计]