Modeling and analysis of solar Doppler difference bias with arbitrary rotation axis

The solar rotation causes the solar Doppler difference bias, which leads to the decline of the velocity measurement accuracy. Modeling and compensation are an effective solution. The limited model with specific geometric direction, where the solar rotation axis is perpendicular to the plane through the Sun, the Earth and Mars, was established. However, in fact, the geometric relationship among the Sun, Mars and the spacecraft is not fully in line with the hypothesis of the model due to the spacecraft orbital angle and the solar rotation axis drift. Thus, this model is not consistent with the fact. In order to solve this problem, a universal solar Doppler difference bias model, which provides the expression with arbitrary rotation axis, is established in this paper. In this method, for any point at the solar surface, four variables including the direction of the solar rotation linear velocity at this point, the distance from this point to the rotation axis, the vector from this point to Mars, and the vector from this point to the spacecraft are calculated. Based on these four variables, the solar Doppler difference bias corresponding to this point is obtained. The theoretical analysis and simulation results demonstrate that the solar Doppler difference bias model with the actual rotation axis is different from that with one of the specific rotation axes. Therefore, it is indispensable to build the proposed model for compensation. Besides, the direction of the solar rotation axis, the spacecraft-Mars-Sun angle and the spacecraft-to-Mars distance are important impact factors for the proposed model.     

Concurrent image-based visual servoing with adaptive zooming for non-cooperative rendezvous maneuvers

An image-based servo controller for the guidance of a spacecraft during non-cooperative rendezvous is presented in this paper. The controller directly utilizes the visual features from image frames of a target spacecraft for computing both attitude and orbital maneuvers concurrently. The utilization of adaptive optics, such as zooming cameras, is also addressed through developing an invariant-image servo controller. The controller allows for performing rendezvous maneuvers independently from the adjustments of the camera focal length, improving the performance and versatility of maneuvers. The stability of the proposed control scheme is proven analytically in the invariant space, and its viability is explored through numerical simulations.     

Deep reinforcement learning for six degree-of-freedom planetary landing

This work develops a deep reinforcement learning based approach for Six Degree-of-Freedom (DOF) planetary powered descent and landing. Future Mars missions will require advanced guidance, navigation, and control algorithms for the powered descent phase to target specific surface locations and achieve pinpoint accuracy (landing error ellipse <5 m radius). This requires both a navigation system capable of estimating the lander’s state in real-time and a guidance and control system that can map the estimated lander state to a commanded thrust for each lander engine. In this paper, we present a novel integrated guidance and control algorithm designed by applying the principles of reinforcement learning theory. The latter is used to learn a policy mapping the lander’s estimated state directly to a commanded thrust for each engine, resulting in accurate and almost fuel-optimal trajectories over a realistic deployment ellipse. Specifically, we use proximal policy optimization, a policy gradient method, to learn the policy. Another contribution of this paper is the use of different discount rates for terminal and shaping rewards, which significantly enhances optimization performance. We present simulation results demonstrating the guidance and control system’s performance in a 6-DOF simulation environment and demonstrate robustness to noise and system parameter uncertainty.     

基于预测模型迭代优化的改进比例导引律

为了提高对加速度时变机动目标的制导精度,将预测决策理论与传统比例导引(PN)相结合,提出了一种通过泰勒级数预测模型迭代控制加速度修正项的改进比例导引律。首先,以预设的低阶泰勒级数预测模型预测特定时间的位移,并计算位移预测值与测量值的差值;然后,通过迭代方法逐阶增加泰勒级数预测模型阶数,直至满足精度要求;最后,计算泰勒级数预测模型的二阶导数,修正比例导引律的加速度指令。仿真结果表明,传统PN和APN的脱靶量分别约为195 m和95 m,提出的改进比例导引律的脱靶量约为8.3 m,极大地提高了制导精度。     

UKF稳定性研究及其在相对导航中的应用

讨论了UKF应用于状态方程为线性,测量方程为非线性对象时的稳定性问题。文中 给出了一个UKF保持稳定的充分条件,指出适当增加过程噪声和测量噪声方差阵的设置值, 有利于保证UKF的稳定性。本文还将UKF应用于航天器相对导航滤波器设计,通过数学仿真和 理论分析证实了文中有关UKF稳定性结论的正确性。本文所得到的结论为将UKF应用于实际工 程,特别是航天器相对导航滤波器设计,提供了理论参考。
     

一种单子样旋转矢量姿态算法

旋转矢量姿态算法可有效抑制高动态环境下捷联惯导系统(SINS)产生的圆锥误差。当 直接应用多子样旋转矢量姿态算法时,会降低系统姿态更新频率;若要保持姿态更新频率, 则需要提高采样频率,从而增加了导航计算机的硬件负担,并导致量化误差突出。针对上述 不足,提出了一种利用当前及前N个姿态更新周期角增量的单子样旋转矢量姿态算 法,并在典型圆锥运动条件下推导了算法补偿项系数。此外,由于陀螺输出经过数字滤波处 理后其幅频特性的改变会影响圆锥误差的补偿效果,根据滤波器特性推导了单子样旋 转矢量姿态算法的修正算法,以便于在工程中推广应用。该算法在不降低姿态更新频率 的同时,可获得较高的解算精度,适于高动态环境应用,实验结果验证了上述算法的正确性 和有效性。
     

基于比例分配的过驱动碟形飞行器滑模控制

为了避免一些执行机构提前出现饱和,针对具有位置约束的过驱动碟形飞行器,按照执行机构的约束范围,提出了一种比例分配策略.基于该策略,可以使双输入系统简化成单输入系统,并避免有执行机构提前出现饱和.考虑执行机构具有一阶动态特性和系统气动参数的不确定性,设计了滑模控制律.通过比例分配和伪逆分配的比较,得出了一种确定伪逆分配权矩阵的方法.仿真结果表明了该方法的正确性和有效性.     

四元数UKF超紧密组合导航滤波方法

针对GPS/SINS(Global Positioning System/Strapdown Inertial Navigation Sys-tem)超紧密组合导航系统线性化滤波在高动态、弱GPS信号等环境下性能下降严重的问题,提出了一种以四元数UKF(Unscented Kalman Filter)为基础的GPS/SINS超紧密组合导航系统非线性滤波方法.通过分析GPS与SINS在超紧密组合导航系统中的关系,提出了一种利用SINS信息来估计、修正GPS载波信号的方法,在此基础上建立了组合导航滤波的状态方程和量测方程;以四元数乘性误差模型为基础,提出了一种可应用于UKF滤波的求解四元数加权均值的方法;提出了基于四元数UKF的GPS/SINS超紧密组合导航滤波方法.仿真实验表明:在高动态环境下,四元数UKF滤波方法增强了GPS/SINS超紧密组合导航系统的定位精度和抗干扰能力.     

用于多模卫星导航信号接收的半球形阿基米德螺旋天线

平面阿基米德螺旋天线具有双向辐射特性,在实现单向辐射时会降低效率 。将平面阿基米德螺旋天线共形到局部半球上可以得到一种新型结构的半球形阿基米德 螺旋天线,分析了这种天线的结构特点,进行了仿真与实验,研究结果表明,在覆盖多模卫 星导航信号工作频率的1.1-1.7 GHz频段内,反射损耗小于-10 dB,增益大于3.9 dB,前后 比大于3.6 dB,相位中心稳定度小于2 mm,可满足宽频带圆极化多模卫星导航共用接收天线 的设计要求。
     

一种改进RLS算法及其在SINS快速对准中的应用

在传统递推最小二乘算法(RLS)中,人为设置的递推初始值将导致状态估计的有偏性,也就丧失了最优性,当量测数据次数较小时尤为严重。摒弃了传统RLS算法"新估计值=旧估计值+修正值"的递推结构,提出了借助中间量进行递推,再由中间量直接作状态估计的改进算法。改进RLS算法状态估计结果与批处理LS算法完全一致,且无需初始状态的任何信息。将改进RLS算法应用于捷联惯导系统(SINS)初始对准。对于一定的初始对准精度要求,理论上改进RLS算法所需的初始对准时间是最短的。最后,SINS初始对准数值仿真结果验证了所提算法的正确性。