【作者】 李九人；

【导师】 唐国金； 李海阳；

【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】 自主交会对接技术是进行在轨维修、空间组装、燃料补给、空间碎片清理和深空探测等任务的支撑技术。本文对基于光学相机的自主交会相对导航、闭环控制协方差分析、多目标最优交会轨迹设计和近距离交会制导与控制等技术进行了研究。论文的主要研究成果如下:建立了仅测角相对导航及其可观测度模型,提出了应用潜在距离信息和主动机动提高仅测角相对导航可观测度的方法。1)根据航天器相对轨道动力学方程,建立了以光学相机为唯一相对测量设备的仅测角相对导航模型;2)采用牛顿迭代方法,推导了相对导航的可观测度模型,实现了对相对导航系统可观测性能的定量分析;3)研究了追踪器在轨道机动时的潜在距离信息及其误差模型,并应用该信息提高仅测角相对导航的可观测度;4)以可观测度为优化指标,对提高仅测角相对导航可观测性能的主动机动策略进行了设计。分析了基于仅测角相对导航的自主交会闭环控制的偏差演化规律。1)基于脉冲轨道机动假设,在仅测角相对导航条件下,建立了自主交会闭环控制的偏差预测、测量更新和控制修正等过程的线性偏差递推模型,获得了相对导航及相对状态控制的协方差表达式;2)根据线性协方差分析方法,对自主交会闭环控制过程的相对导航和相对状态控制的协方差特性进行了分析,并采用Monte Carlo方法验证了结果的正确性。以燃料消耗、相对位置鲁棒性和相对速度鲁棒性为优化指标,提出了基于闭环控制的自主交会多目标最优轨迹设计方法。1)采用线性协方差分析方法,对仅测角条件下的自主交会闭环控制偏差进行了分析,并在此基础上构建了燃料消耗、相对位置鲁棒性和相对速度鲁棒性等优化指标;2)分别采用NSGA-II和物理规划方法,对考虑上述优化指标的多目标最优交会轨迹进行了设计,并采用Monte Carlo方法对优化结果的有效性进行了验证。提出了近距离交会接近、绕飞和逼近段制导与控制的新方法。1)建立了视线制导控制模型,并采用该制导模型对考虑导航和控制误差情况下的接近段轨道进行了控制;2)推导了轨道面内和水平面内快速绕飞控制的速度增量计算模型,研究了快速绕飞的模糊控制;3)设计了用于逼近轨道控制的模糊控制器、PID控制器和模糊/PID混合控制器,并对三种控制器的控制稳定性和燃料消耗进行了对比分析;4)建立了适用于任意偏心率的相对运动轨道和姿态动力学模型,对追踪器逼近旋转目标的参考轨迹和参考姿态进行了分析和设计,推导了用于六自由度逼近控制的自适应滑模控制器,通过仿真验证了参考轨迹和参考姿态设计的合理性和自适应滑模控制器的稳定性。论文深入探讨了基于光学相机的自主交会相对导航与控制问题,发展了仅测角相对导航、闭环控制协方差分析、多目标最优交会轨迹设计和近距离交会制导与控制等技术,获得了一些关于自主交会导航和控制问题研究的新方法和有用结论,研究成果对自主交会任务设计具有较好的参考价值。更多还原

【Abstract】 The autonomous rendezvous and docking technology is crucial for on-orbit service, space-assembly, propellant supply, space debris clearance, and deep-spaced exploration. Based on measurements of optical camera, the methods of relative navigation, closed-loop covariance analysis, optimal multi-objective rendezvous trajectory design, and guidance and control of close range rendezvous are researched. The main results achieved in this dissertation are summarized as follows.The angles-only relative navigation and its degree of observability (DOO) models are formulated, and the latent range information and the active maneuvering strategy are employed to enhance the DOO of angles-only relative navigation. 1) According to relative dynamics equations, angles-only relative navigation model is formulated under the condition that optical camera is the only available sensor for relative measurement. 2) DOO model of angles-only relative navigation is obtained based on the Newton iterative method, and then the observability of navigation is quantificational. 3) The latent range information of orbital maneuver is analyzed, and is employed to enhance the DOO of angles-only relative navigation. 4) According to the performance index of DOO, the optimal maneuvering strategy is designed for chaser to enhance the observability of angles-only relative navigation.The covariance of autonomous rendezvous under closed-loop control is analyzed on basis of angles-only relative navigation. 1) Based on the assumptions of impulsive control and angles-only relative navigation, error dispersion equations during the process of propagation, update and correction are derived, and the closed-loop control covariance of relative navigation and relative trajectory control are obtained. 2) According to the linear covariance analysis method, the closed-loop control covariance of angles-only relative navigation and trajectory control are calculated, and the effectiveness of the covariance result is validated by Monte Carlo simulations.According to the performance indexes of propellant consumption, relative position robustness and relative velocity robustness, the method of optimal multi-objective trajectory design based on closed-loop control is proposed. 1) According to linear covariance analysis method, the closed-loop control covariance of autonomous rendezvous with angles-only relative navigation is analyzed, and expressions of propellant consumption, relative position robustness and relative velocity robustness are formulated. 2) Considering these performance indexes, the multi-objective optimization algorithms of NSGA-II and Physical Programming are employed to solve the optimal multi-objective rendezvous problem respectively, and the effectiveness of the optimization results are testified by Monte Carlo simulations.New guidance and control algorithms are developed for closing, flyaround and approach during the close range rendezvous. 1) Line-of-sight guidance law is formulated, which is proposed for the closing trajectory control perturbed by measurement and control noise. 2) The propellant consumptions of the in-plane and horizontal fast flyaround are analyzed, and the fuzzy controller is proposed for the flyaround trajectory control. 3) Fuzzy, PID and Fuzzy/PID hybrid controllers are designed for final approach trajectory control respectively, and their robustness and propellant consumptions are analyzed and compared. 4) The relative position and attitude dynamics equations are formulated, which are suitable for spacecraft with arbitrary eccentricity. The reference trajectory and attitude for the chase approaching the uncontrolled rotating target are designed, and adaptive sliding mode controller is proposed for the six degree of freedom position and attitude control. The simulation results show that the reference trajectory and attitude are reasonable and the adaptive sliding mode controller is robust.The relative navigation and control based on optical camera for autonomous rendezvous are researched in this dissertation, and the technologies such as the angles-only relative navigation, the close-looped covariance analysis, the optimal multi-objective rendezvous trajectory design, and the guidance and control of close range rendezvous are developed. Some new approaches and significant conclusions of the relative navigation and control are obtained, which are useful for autonomous rendezvous mission design.更多还原