【作者】 谭丽芬；

【导师】 唐国金；

【作者基本信息】 国防科学技术大学 ， 航空宇航科学与技术， 2011， 博士

【摘要】 轨道机动飞行器具有较强机动能力,通过空间交会,它可对空间目标执行观测、捕获、吸附、伴飞及动能碰撞拦截等空间任务。赤道大椭圆轨道机动飞行器是指停泊在倾角为零的大椭圆轨道上可实施机动的航天器。本文以赤道大椭圆轨道机动飞行器与中、高轨目标飞行器的空间交会问题为背景,研究了交会过程设计、远程最优交会轨道规划与制导方法、近程最优交会轨道规划与制导方法和末制导导引律。全文主要研究成果如下:提出了赤道大椭圆停泊轨道的倾角长期漂移解析计算与保持方法。(1)分析了赤道大椭圆轨道机动飞行器与任意轨道目标飞行器的交会过程,定义、划分了各交会阶段,建立了各交会阶段的动力学模型。(2)推导了小倾角大椭圆轨道的倾角长期漂移解析计算公式,通过仿真验证了解析计算公式的正确性。提出了赤道椭圆交会轨道的远程最优交会轨道规划方法。(1)给出了椭圆固定时间单脉冲多圈Lambert最优交会轨道的快速确定方法,只用比较两条轨道就可确定最优交会轨道。(2)结合固定时间单脉冲多圈Lambert交会轨道,应用非最优主矢量理论、非最优轨道改进方法与非线性规划方法,提出了远程多脉冲最优交会策略。(3)研究了赤道大椭圆轨道机动飞行器与多个目标飞行器的固定时间最优交会问题,给出了省能量目标飞行器交会序列的生成方法。提出了考虑J2摄动的赤道椭圆交会轨道的远程最优交会轨道规划与制导方法。(1)考虑地球非球形J2项摄动,提出了一种脱靶量满足精度要求,且省燃料的固定时间远程脉冲交会轨道规划方法。(2)应用Legendre伪谱法,研究并得到了固定时间远程有限推力最优交会制导方案;针对制导优化变量多的特点,构造了优化变量初值生成器;仿真验证了制导方案与求解策略的有效性。(3)考虑地球非球形J2项摄动,采用协方差分析描述函数法与Monte Carlo仿真方法,分析了初始状态误差对远程交会的影响。(4)给出了一种基于J2摄动轨道状态转移矩阵的制导修正策略,在远程交会的惯性滑行段进行轨道修正。提出了赤道椭圆交会轨道的近程最优交会轨道规划与制导方法。(1)对相对动力学模型的几种模型误差进行了比较,分析了导航误差与控制误差对交会终端偏差的影响,确定椭圆线性相对运动方程为近程交会相对动力学模型;给出了椭圆线性最优交会轨道非线性规划模型。(2)应用主矢量理论、非最优主矢量理论与非线性规划方法,提出了固定时间多脉冲近程最优交会策略。(3)基于椭圆线性相对运动方程,考虑飞行器质量变化,结合最优控制混合方法与多重打靶法提出了近程有限推力最优交会制导混合优化策略。研究了赤道大椭圆轨道机动飞行器搭载的智能撞击器末制导导引律。(1)利用线性二次型导引提供神经网络训练数据,采用遗传算法对神经网络结构与初始连接权值、阈值进行优化设计,给出了满足脱靶量要求且省能量的神经网络末制导导引律,仿真结果表明,给出的导引律对不同目标机动方式具有优良的适用性与鲁棒性。(2)采用模糊控制逻辑取代符号切换函数,提出了模糊Terminal滑模变结构导引律,利用Lyapunov稳定性理论对模糊Terminal滑模变结构控制器的稳定性进行了分析与证明。仿真表明,给出的导引律在消除滑模变结构导引固有抖振现象的同时,对外界干扰仍具有强鲁棒性能。论文研究了椭圆最优交会轨道规划与制导方法,拓展了最优交会理论,具有一定理论意义。论文提出的椭圆远程、近程最优交会策略和末制导导引律可为轨道机动飞行器空间交会方案设计与轨道控制提供技术支撑。更多还原

【Abstract】 Orbit maneuver vehicle (OMV) has high maneuver ability. It is able to perform space missions using space rendezvous technology, including the space oberservation, capture, absorption, flyby and kinetic interception to some space spacecraft. The OMV studied in the paper is specified to that parking on the high eccentric orbit in equator plane. With the background of the rendezvous missions between the OMV and the target spacecraft on medium or high earth orbit, the dissertation studies the rendezvous process design, long-distance rendezvous orbital planning and guidance approaches, near-distance rendezvous orbital planning and guidance approaches, and the terminal guidance laws. The main results achieved in this dissertation are summarized as follows.The analytic formulae for the long-duration drift of inclination of the elliptical parking orbit in equator plane are deduced. (1) The rendezvous process between OMV and target spacecraft on un-limited orbit is analyzed. The rendezvous process is divided into several phases, and the dynamic model of each phase is established. (2) Aiming at the orbit with small inclination and high eccentricity, the analytic formulae for the long-duration drift of inclination are deduced and validated by simulation.The optimal planning approaches for the long-distance elliptical rendezvous trajectory in equator plane are proposed. (1) An optimal multiple-revolution Lambert rendezvous algorithm for elliptical trajectory is provided, and the optimal rendezvous trajectory can be obtained by solving only two Lambert orbits. (2) With the single-impulse multiple-revolution Lambert rendezvous trajectory, the optimal long-distance multiple-impulse rendezvous strategy is proposed by applying the non-optimal primer vector theory, the methods improving non-optimal trajectory and nonlinear programming approach. (3) The fixed-time rendezvous between the OMV and multiple target spacecraft is also studied, and the method to obtain the fuel-optimal visiting sequence of targets is provided.The optimal planning and guidance approaches with considering J2 perturbation for the long-distance elliptical rendezvous trajectory in equator plane are proposed. (1) Considering the J2 perturbation, the fuel-optimal fixed-time long-distance impulsive rendezvous strategy with high precision is proposed. (2) The optimal guidance plan of finite-thrust rendezvous is proposed by using Legendre Pseudospectral Method (LPM), and an initial guess generator is provided for a large number of design variables. The proposed guidance plan and the initial guess generator is validated by simulation. (3) Considering the J2 perturbation, the nonlinear error analysis for the long-distance rendezvous trajectory with initial state errors is performed by using covariance analysis description equation technique (CADET) and Monte Carlo simulation. (4) A guidance correct strategy based on state transfer matrix with considering J2 perturbation is proposed to reduce the effect of errors on the rendezvous trajectory.The optimal planning and guidance approaches for the near-distance elliptical rendezvous trajectory in equator plane are proposed. (1) The modeling errors of relative dynamic models are analyzed and compared, and the influences of navigation error and control error are analyzed. The linearized elliptical relative equations are chosen to describe the relative movement during the near-distance rendezvous phase. The nonlinear planning model for the elliptical linear rendezvous planning is provided. (2) The fixed-time near-distance rendezvous optimization approach is proposed by using the non-optimal primer vector theory, the methods improving non-optimal trajectory and nonlinear programming method. (3) Considering the mass variation and the constant thrust with adjustable direction, the hybrid optimization strategy for near-distance finite-thrust optimal rendezvous is proposed by hybrid optimal control method and multiple-shooting technique.The guidance laws for the rendzvous terminal phase of intelligent impactor carried by OMV are studied. (1) Considering environment disturbation, the neural network guidance (NNG) law is proposed and trained with data provided by LQG guidance, and the network structure, initial connect parameters and bias parameters are optimized by genetic algorithm. Simulation results show that the NNG guidance law satisfies miss-distance requirement and is roubust to the target’s various kind of maneuver. (2) Replacing the sign function with fuzzy control logic, the fuzzy terminal variable structure guidance law is proposed. The stability of fuzzy controller is analyzed and proved by Lyapunov stability theory. Simulation results show that the provided guidance law doesn’t only have strong robustness, but also eliminates the chattering phenomenon which is inherent of a sliding mode control.The dissertation studies optimal planning and guidance approaches for elliptical rendezvous trajectory, and expands the optimal rendezvous theory which have some theoretical significance. The proposed optimal rendezvous strategies both for long-distance elliptical rendezvous trajectory and near-distance elliptical rendezvous trajectory and terminal guidance laws will provide technique support for the space rendezvous scheme design and orbit control of OMV.更多还原