带落角落速约束的导弹虚拟期望落角末制导律

为解决导弹末制导阶段同时考虑落角和落速约束时带来的过载需求大、落速散布广的问题,提出一种基于虚拟期望落角的末制导律。首先,提出虚拟期望落角的概念,设计过渡函数降低末制导初期过载需求；然后,分析过渡函数各参数对落角、落速影响,设计预测-校正算法计算期望参数；为了提高预测效率与精度,使用深度神经网络离线训练弹道数据集。实际飞行中,基于扩展卡尔曼滤波在线辨识气动参数摄动,提高算法的适应性。蒙特卡洛仿真结果表明,所提出的算法能够降低末制导初期过载需求。在满足落角约束与位置精度的前提下,落速控制精度在±15 m/s以内。     

基于视惯融合的大型空间碎片质心位置辨识

大型自由翻滚碎片的质心是在轨操作基坐标系下的不动点，也是碎片连体基下动力学参数向卫星坐标系转换的基准，对其精确识别是提高碎片动力学参数辨识精度的关键。提出基于惯性单元测量数据与双目视觉定位数据融合的大型空间碎片质心位置识别方法。基于无力矩欧拉方程，获取附着到空间碎片表面的惯性单元间转换关系，利用该转换关系对惯性单元冗余测量数据优化，再优化求解惯性单元到质心点距离；利用双目视觉获取惯性单元上标记点动态坐标，再利用惯性单元到质心点距离，基于三点定位原理识别大型空间碎片的质心位置。以加入高斯白噪声的惯性单元与双目视觉测量数据进行仿真，结果表明优化解算后惯性单元实时测量数据的误差降低到1%以下，解算的质心位置三轴误差小于0.47mm；开展了地面试验，结果表明，解算的质心位置三轴误差小于0.49mm。仿真和试验证明，该方法能够为大型空间碎片的消旋、捕获任务提供准确的数据基准。     

星载控制软件在轨动态重构技术研究

为使星载控制软件可在轨动态重构，提出一种基于量子编程框架、无须操作系统支持、可实现多版本切换的星载控制软件在轨动态重构方法。在分析影响在轨动态重构关键技术基础上，从量子框架的面向对象运行机制出发来寻求软件框架对动态重构的支持；通过划分函数边界，将函数归类为内部函数和公共函数，避免了模块间的循环依赖；给出了函数向量表维护策略，并以版本号为导向实现了向量表切换。该方法在BM3803星载处理器平台进行了充分测试，结果表明：所提出的在轨重构方法系统无须停机、版本可回退且更新过程可靠。本方法占用内存小、平台依赖性弱、代码可复用性强，可推广应用至硬件资源有限的星载控制器终端。     

空天飞行器星敏感器安装误差在线标定方法

空天飞行器高动态、长航时的运动特性可能导致一体化安装的惯性/天文组合导航系统中星敏感器与惯导间产生安装误差角。设计了一种星敏感器安装误差角动态辨识方法,建立了星敏感器安装误差角模型,设计了基于天文角度观测的星敏感器安装误差角动态辨识方案,分析了不同机动飞行方式下星敏感器安装误差角的可观测度。仿真结果表明,所设计的基于卡尔曼滤波的动态辨识方法能够在飞行器机动过程中快速地对星敏感器安装误差角进行在线标定,对安装误差角的标定值可以达到实际误差值的85%以上,有效地提高了组合导航系统的精度。     

增强神经网络辨识模型泛化能力的研究

神经网络(Artificial Neural Network,ANN)辨识模型的泛化能力是其最主要的性能之一,增强ANN模型的泛化能力也是近年来国内外有关专家学者研究的重点问题。大量研究表明,ANN模型泛化能力的改善与很多因素相关联,其中恰当的性能指标函数设计是一个重要影响因素。文中在分析常见的基于均方误差最小原则的性能指标函数基础上,通过加入ANN辨识模型权值间的延迟信息,进而获得一种改进型性能指标函数。通过仿真,验证了所设计的改进型性能指标函数对增强ANN辨识模型的泛化能力是有效的。     

Transfer learning: A new aerodynamic force identification network based on adaptive EMD and soft thresholding in hypersonic wind tunnel

The aerodynamic test in the pulse combustion wind tunnel is very important for the design, evaluation and optimization of aerodynamic characteristics of the hypersonic aircraft. The test accuracy even affects the success or failure of hypersonic aircraft development. In the aerodynamic test of pulse combustion wind tunnel, the aerodynamic signal is disturbed by the inertial force signal, which seriously affects the test accuracy of aerodynamic force. Aiming at the above problems, this paper innovatively proposes an aerodynamic intelligent identification method, that is the transfer learning network based on adaptive Empirical Modal Decomposition (EMD) and Soft Thresholding (TLN-AE&ST). Compared with the existing aerodynamic intelligent identification model based on deep learning technology, this study introduces the transfer learning idea into the aerodynamic intelligent identification model for the first time. The TLN-AE&ST effectively alleviates the problem of scarcity of training samples for intelligent models due to the high cost of wind tunnel tests, and provides a new idea for further implementation of deep learning technology in the field of wind tunnel aerodynamic testing. And this study designed residual attention block with soft threshold and dense block with adaptive EMD in TLN-AE&ST model. Residual attention block with soft threshold module can more effectively suppress the influence of instrument noise signal on model training effect. Dense block with adaptive EMD makes the deep learning model no longer a black box to a certain extent, and has certain physical significance. Finally, a series of wind tunnel tests were carried out in the Φ = 2.4 m pulse combustion wind tunnel of China Aerodynamic Research and Development Center to verify the effectiveness of TLN-AE&ST.     

Workspace,stiffness analysis and design optimization of coupled active-passive multilink cable-driven space robots for on-orbit services

The use of space robots (SRs) for on-orbit services (OOSs) has been a hot research topic in recent years. However, the space unstructured environment (i.e.: confined spaces, multiple obstacles, and strong radiation interference) has greatly restricted the application of SRs. The coupled active-passive multilink cable-driven space robot (CAP-MCDSR) has the characteristics of slim body, flexible movement, and electromechanical separation, which is very suitable for extreme space environments. However, the dynamic and stiffness modeling of CAP-MCDSRs is challenging, due to the complex coupling among the active cables, passive cables, joints, and the end-effector. To deal with these problems, this paper proposes a workspace, stiffness analysis and design optimization method for such type of MCDSRs. Firstly, the multi-coupling kinematics relationships among the joint, cables and the end-effector are established. Based on hybrid series-parallel characteristics, the improved coupled active–passive (CAP) dynamic equation is derived. Then, the maximum workspace, the maximum stiffness, and the minimum cable tension are resolved, among them, the overall stiffness is the superposition of the stiffness produced by the active and the passive cable. Furthermore, the workspace, the stiffness, and the cable tension are analyzed by using the nonlinear optimization method (NOPM). Finally, an 8-DOF CAP-MCDSR experiment system is built to verify the proposed modeling and trajectory tracking methods. The proposed modeling and analysis results are very useful for practical space applications, such as designing a new CAP-MCDSR, or utilizing an existing CAP-MCDSR system.     

A hybrid deep neural network based on multi-time window convolutional bidirectional LSTM for civil aircraft APU hazard identification

Safety is one of the important topics in the field of civil aviation. Auxiliary Power Unit(APU) is one of important components in aircraft, which provides electrical power and compressed air for aircraft. The hazards in APU are prone to cause economic losses and even casualties. So,actively identifying the hazards in APU before an accident occurs is necessary. In this paper, a Hybrid Deep Neural Network(HDNN) based on multi-time window convolutional neural network-Bidirectional Long Short-Term M...     

A model-free method for attitude estimation and inertial parameter identification of a noncooperative target

Determining the attitude and inertial parameters of a noncooperative target is essential in an on-orbit servicing mission. Various methods based on machine vision have been proposed, but most of them require the 3D model of the target. This paper proposes a model-free method through sequentially registering point clouds captured by a depth camera. Our main contributions are the avoidance of the ambiguity in registration, and the combination of the multiplicative extended Kalman filter and the pose graph optimization to reduce the effect of measurement noise and drift error. A hardware experiment was performed to generate the sequence of point clouds of a three-axis free-floating target and validate our method. The result shows that the proposed method outperforms existing methods and effectively identifies the inertial parameters, including the normalized principal moments of inertia and the orientation of principal axes.     

Quick identification of guidance law for an incoming missile using multiple-model mechanism

A guidance law parameter identification model based on Gated Recurrent Unit (GRU) neural network is established. The scenario of the model is that an incoming missile (called missile) attacks a target aircraft (called aircraft) using Proportional Navigation (PN) guidance law. The parameter identification is viewed as a regression problem in this paper rather than a classification problem, which means the assumption that the parameter is in a finite set of possible results is discarded. To increase the training speed of the neural network and obtain the nonlinear mapping relationship between kinematic information and the guidance law parameter of the incoming missile, an output processing method called Multiple-Model Mechanism (MMM) is proposed. Compared with a conventional GRU neural network, the model established in this paper can deal with data of any length through an encoding layer in front of the input layer. The effectiveness of the proposed Multiple-Model Mechanism and the performance of the guidance law parameter identification model are demonstrated using numerical simulation.