一种同步自旋卫星姿态控制优化方法

根据同步自旋卫星姿态的运动规律以及自旋卫星姿态的长期实测数据,建立了姿态变化规律方程,采用最小二乘方法求解姿态变化方程,实现对自旋卫星姿态预测,预测误差小于0.05..针对目前工程应用中姿态保持周期短的问题,结合姿态及轨道倾角变化规律,建立目标姿态选取的约束方程,并利用投影梯度法求解约束方程,寻求姿态保持时间长的姿态控制目标.实际应用结果表明：该方法可以延长姿态控制周期,使自旋卫星姿态保持周期达到1 a.     

空间机械臂视觉测量技术研究

空间机械臂技术是维护空间站、执行指定任务、保障航天员出舱作业的安全等必不可少的关键技术。视觉测量技术则是保证空间机械臂顺利完成空间遥操作任务的前提。基于此前提,首先对空间机械臂视觉测量技术进行了概述,并分别对手眼关系标定、标志器识别以及相对三维位姿测量等关键技术进行了阐述;然后,以加拿大机械臂为例提出了一种基于边缘特征的标志器识别算法,给出了具体的识别流程,并采用一种基于非迭代的相对位姿测量算法实现了位姿求解;最后,针对标志器识别算法和位姿测量算法给出了基于仿真图像的实验结果和机械臂原理样机集成实验结果。最终实验结果表明:提出的空间机械臂视觉测量方法合理可行,满足预期的技术指标要求,具有较强的工程应用价值。     

室外大气折射率结构常数实时测量研究

与传统的射频链路相比,自由空间光通信FSOC（Free Space Optical Communication）系统是对现有无线通信技术的有力补充,为了抑制大气湍流对信号光束带来的相位噪声和强度波动,有必要对通信信道状况进行实时监测。首先介绍了基于光斑质心漂移的大气折射率结构常数测量方法,利用大口径接收靶面和窄带滤光片,结合实时图像处理算法,实现了大气折射率结构常数的全天候实时监测。将大气折射率结构常数和光强闪烁指数进行了比较分析,测量结果表明它们的包络之间具有很好的线性关系,进一步验证了实时测量系统的可靠性。     

基于高斯过程的航天器自适应滑模姿态控制

针对存在模型不确定性和外界干扰的刚性航天器,提出了一种基于高斯过程回归(GPR)的新型自适应滑模姿态控制算法。该算法具有自学习能力,在不同的姿态控制任务下都能够实现高精度、强鲁棒和高效率的姿态跟踪。首先,在航天器的四元数标称系统动态模型基础上,应用在线稀疏高斯过程回归(SOGP)方法学习系统的未知动态;其次,结合高斯过程的预测均值设计滑模控制算法,利用高斯过程的预测方差自适应调节控制增益,并应用李雅普诺夫方法严格证明闭环系统的稳定性,保证了航天器姿态跟踪误差的渐进收敛性;最后,通过数值仿真验证了所设计控制器的有效性。结果表明,该自学习控制算法与自适应滑模控制(ASMC)与神经网络自适应控制等算法相比,具有更快的收敛速度、更高的跟踪精度以及更低的控制成本。     

Receiver-channel based adaptive blind equalization approach for GPS dynamic multipath mitigation

Aiming at mitigating multipath effect in dynamic global positioning system (GPS) satellite navigation applications, an approach based on channel blind equalization and real-time recursive least square (RLS) algorithm is proposed, which is an application of the wireless communication channel equalization theory to GPS receiver tracking loops. The blind equalization mechanism builds upon the detection of the correlation distortion due to multipath channels; therefore an increase in the number of correlator channels is required compared with conventional GPS receivers. An adaptive estimator based on the real-time RLS algorithm is designed for dynamic estimation of multipath channel response. Then, the code and carrier phase receiver tracking errors are compensated by removing the estimated multipath components from the correlators’ outputs. To demonstrate the capabilities of the proposed approach, this technique is integrated into a GPS software receiver connected to a navigation satellite signal simulator, thus simulations under controlled dynamic multipath scenarios can be carried out. Simulation results show that in a dynamic and fairly severe multipath environment, the proposed approach achieves simultaneously instantaneous accurate multipath channel estimation and significant multipath tracking errors reduction in both code delay and carrier phase.     

复杂环境下多无人机轨迹姿态协同控制

考虑多无人机在实际飞行过程中的避障需求，在建立面向控制的四旋翼无人机轨迹姿态模型基础上，为克服传统多无人机为每架无人机规划轨迹的不足，研究了基于多无人机中心点的轨迹设计策略。进一步，考虑避障约束，基于半定规划进行迭代区域扩张完成了多无人机的安全飞行区域及队形设计。在此基础上，基于干扰补偿策略为每架无人机设计了协同控制器，最终确保了多无人机在多障碍环境下的安全飞行。     

一种用于同步静止卫星监测的微型VLBI网

针对地球静止轨道（GEO）卫星全天时全天候高精度的监测需求，考虑传统甚长基线干涉（VLBI）测站高成本、高投入和GEO卫星专用观测时段有限等制约条件，研发了简易型VLBI观测系统，并组建了包括上海、都匀和乌鲁木齐三站的微型VLBI网（micro VLBI network，MVN），开展了并置站测试以及对GEO卫星亚太6C的连续监测，并评估了当前MVN的观测能力。结果表明MVN扣除系统差后的单站接收精度为2ns，各基线观测时延拟后残差约几纳秒，GEO目标实测位置精度为百米级（内外符精度分别约100m和400m）。不同于传统VLBI和其他GEO监测手段，MVN还具备全天时、全天候、低造价、易布设及易推广等特点，充分表明了其在GEO卫星监测领域的应用价值。     

Evaluation on real-time dynamic performance of BDS in PPP,RTK, and INS tightly aided modes

Since China’s BeiDou satellite navigation system (BDS) began to provide regional navigation service for Asia-Pacific region after 2012, more new generation BDS satellites have been launched to further expand BDS’s coverage to be global. In this contribution, precise positioning models based on BDS and the corresponding mathematical algorithms are presented in detail. Then, an evaluation on BDS’s real-time dynamic positioning and navigation performance is presented in Precise Point Positioning (PPP), Real-time Kinematic (RTK), Inertial Navigation System (INS) tightly aided PPP and RTK modes by processing a set of land-borne vehicle experiment data. Results indicate that BDS positioning Root Mean Square (RMS) in north, east, and vertical components are 2.0, 2.7, and 7.6?cm in RTK mode and 7.8, 14.7, and 24.8?cm in PPP mode, which are close to GPS positioning accuracy. Meanwhile, with the help of INS, about 38.8%, 67.5%, and 66.5% improvements can be obtained by using PPP/INS tight-integration mode. Such enhancements in RTK/INS tight-integration mode are 14.1%, 34.0%, and 41.9%. Moreover, the accuracy of velocimetry and attitude determination can be improved to be better than 1?cm/s and 0.1°, respectively. Besides, the continuity and reliability of BDS in both PPP and RTK modes can also be ameliorated significantly by INS during satellite signal missing periods.     

Tumbling object deorbiting using spaceborne laser engagement – A CubeSat case study

This paper demonstrates active space debris removal using spaceborne laser systems. The laser beam and the surface of the target are discretised into multiple rays and finite elements, respectively, for laser-target interaction modelling, in which the laser ablation process is investigated. A high-fidelity attitude/orbit propagator tool is developed to account for both the linear impulse and angular impulse induced by the laser engagement and other perturbations. The laser system is activated only when three switch criteria are satisfied. In numerical simulations, laser pulses from international space station are generated to deorbit a 3U CubeSat with initially tumbling modes. The results validate the effectiveness of deorbiting tumbling CubeSats using spaceborne laser engagement, with the perigee height lowered by approximately $2.4\mathrm{km}$ in around $30\min$ after $2\mathrm{h}$ propagation. It is also found that the laser engagement becomes more effective for an initially faster rotating object.     

Adjustable box-wing model for solar radiation pressure impacting GPS satellites

One of the major uncertainty sources affecting Global Positioning System (GPS) satellite orbits is the direct solar radiation pressure. In this paper a new model for the solar radiation pressure on GPS satellites is presented that is based on a box-wing satellite model, and assumes nominal attitude. The box-wing model is based on the physical interaction between solar radiation and satellite surfaces, and can be adjusted to fit the GPS tracking data.