As artificial intelligence technology developing rapidly, the intelligence level of unmanned systems is much increasing. Specially, intelligent reconnaissance technology is more mature and widely used. To solve the above problems, an adversarial patch attack based camouflage and deception method is proposed. Convolutional neural network is used to build a classifier as the attack object, and a novel patch generation method and loss function are designed to attack target samples, which effectively maps the attacked target samples to the specified wrong target category. A directed evaluation method and wealthy experiments are provided to verify the advancement and effectiveness of this method.

At present, with the complex and changeable game environment, deep learning models such as deep convolutional neural networks are introduced to assist in improving personnel's cognition and decision-making level of the game situation. However, when deep learning is introduced into game situation understanding, it also introduces data uncertainty and cognitive uncertainty in artificial intelligence, which leads to problems such as divergence of artificial intelligence prediction results. Key elements of uncertainty in the measurement process of game situation understanding are decomposed, extracted and measurement modeling constructed based on the measurement uncertainty evaluation method. The experimental results show that the physical measurement method based on GUM can effectively measure and evaluate the cognitive uncertainty of game situation accurately and efficiently. Finally, based on Monte Carlo method, the proposed new qualitative measurement method of game situation cognition uncertainty is verified, which shows the accuracy and applicability of the proposed method.

This article takes the real-time intelligent reconnaissance of the American "Black Swift" hypersonic intelligent aircraft as an example to analyze the demand for strong real-time, high-energy efficiency intelligent computing of typical intelligent hypersonic vehicles. On this basis, it analyzes in detail how to build a strong real-time, high-energy efficiency intelligent computing system for intelligent hypersonic vehicles from three levels: intelligent model lightweight, software and hardware collaborative compilation and optimization, and ultra-heterogeneous integrated computing hardware. Furthermore, a system integration example is provided to illustrate the practical application of these principles. In the future, with the loading of the strong real-time, high-energy efficiency intelligent computing system, intelligent hypersonic aircraft will be more autonomous, reliable and capable of group collaboration, and will drive the intelligent upgrade of aerospace cross-domain flight and global rapid transit.

To address the limitations of existing methods for underwater unmanned vehicle (UUV) motor fault diagnosis, which rely on manual feature extraction and do not fully leverage the potential of intelligent diagnosis, a two-stream CNN-LSTM fault diagnosis model is proposed. The model employs convolutional neural networks as feature extractor, which can learn the low frequency trend and high frequency detail features of the original signal without complex pre-processing steps, making real-time motor status monitoring possible. Afterwards, the classifier based on the long short-term memory network uses these features to explore temporal dependencies and identify motor faults. Experiments are conducted on a self-constructed UUV motor fault simulation platform, and the performance of the model is validated by setting multiple speeds and load conditions. The results show that this method can efficiently diagnose six typical states in UUV motors and achieve an average diagnostic accuracy of 97.22%. These findings demonstrate the model's effectiveness and robustness in UUV motor fault diagnosis.

Aiming at the problem of difficulty in quickly and accurately estimating the pose of non-cooperative structures of satellites during the operation of space robotic arms, a neural radiation field based method for estimating and tracking the pose of non-cooperative key structures of satellites is proposed. This method first obtains the scene point cloud online through an RGBD camera, segments the point cloud to obtain satellite key structures, and then uses neural radiation fields to automatically establish a three-dimensional model of the key structures. Finally, based on the initial pose generation network and pose evaluation network, accurate pose estimation is obtained. An experimental platform consisting of an RGBD depth camera, a robotic arm, and a satellite model is constructed to conduct pose estimation experiments on key structures of satellites with different poses. The experimental results show that the algorithm proposed can automatically construct a 3D model of non-cooperative targets online without the need for human preparation of target data in advance. At the same time, it can effectively deal with target object occlusion and motion situations, thus achieving true non cooperative target pose estimation and tracking in spatial operations.

With the increasing frequency of human space activities, there are surviving debris of re-entry spacecraft and debris that impact the Earth's surface, causing typical damage cases and affecting the future technological development of spacecraft. Firstly typical re-entry damage cases and disposal methods of foreign spacecraft that have attracted significant social attention are analyzeds. Secondly the main international norms and standards for re-entry casualty risk are introduced. Then those assessment methods of re-entry casualty risk are studied together with re-entry risk analysis tools of NASA and ESA. Finally it's looked forward to the furture development and existing problems of major space powers in the technical directions of active deorbit, assisted deorbit, large-scale reentry, and reusable spacecraft. Research shows that although the re-entry casualty risk of spacecraft and debris is still difficult to predicte very accurately, it is expected to be moderately controlled by the design of risk-reduction measures along with technological development.

Numerical investigation on the characteristics of a two-dimensional, mixed-compression inlet with various leading edge bluntness is presented. Effects of leading edge bluntness on the self-starting ability, aerodynamic performance of the inlet at design and off design operations are acquired. Results indicate that, with the increase of blunted radius, the self-starting ability and mass flow capture of the inlet are deteriorated, the backpressure tolerance capability and critical total pressure recovery coefficient goes down, while the drag coefficient rises obviously. 5% decrease of mass flow ratio is observed while the backpressure tolerance and critical total pressure recovery coefficient drop off at least 8.5% for the design operating point. Complex shock wave interference pattern forms due to the oblique shock waves from the external compression of the inlet intersecting the bow shock wave produced in front of the leading edge with the variation of freestream Mach number and the angle of attack. Diminution of flow separation in the inner side of the inlet lip is observed with the increase of blunted radius at a high Mach number condition. As the angle of attack rises, the influence of the bow shock wave induced by the blunted leading edge on the performance of the inlet is found to be weaken.

Composite materials used in engineering design often exhibit different properties in tension and compression in addition to anisotropy. The tensile and compressive anisotropy of constitutive behavior can affect the stress and strain distribution of composite structures under operating conditions, further consideration of the difference between tension strength and compression strength may have an impact on the strength analysis results of the structure. This article takes the carbon-carbon composite material air rudder shaft as the research object, implements secondary development through ABAQUS' USDFLD interface, compares and analyzes the changes in rudder shaft stress before and after considering tensile and compressive anisotropy. Then, considering the tensile, compressive and in-plane shear strength of the material, corresponding stiffness reduction models are introduced to analyze the failure process of the rudder shaft under bending and shear loads. The analysis results indicate that different properties in tension and compression will have a significant impact on the stress distribution of composite structures, and this analysis method can achieve more accurate simulation of the failure process of the rudder shaft.

In response to the current reality that cryogenic liquid launch vehicles have become the mainstay of space missions and the urgent need to enhance personnel competency, this study proactively addresses the training requirements of launch sites for the upcoming universal ground-based testing, launch, and control systems. By fully considering scenarios such as normal testing and launch procedures, troubleshooting under abnormal conditions, and emergency response operations, research, design, and development of a simulation and training system for cryogenic liquid launch vehicle testing and launch procedures are conducted. This system thoroughly analyzes the universal characteristics of the product and aligns with the training needs of launch sites for the soon-to-be-deployed universal ground-based testing, launch, and control systems. It develops hardware equipment and testing software consistent with actual products, covering all phases of testing operations across launch site systems. It supports comprehensive, full-process operational training for all systems and positions, fully meeting the training needs of personnel involved in cryogenic launch vehicle testing and launch operations. The system is capable of fostering a more technically proficient testing and launch team, thereby comprehensively supporting the foundational capacity building of a leading spacefaring nation.

In order to respond to the booming development of the commercial satellite launch service market, while effectively utilizing the launching capacity of domestic satellite launch missions and the available envelope space within the payload fairing, and to provide more cost-effective launch services for satellite users, the Long March-2D launch vehicle has developed a main satellite-disk mounting configuration and a tandem-parallel-side hybrid configuration for micro satellite piggyback and multi-satellite rideshare launch missions. This has been applied in the launch missions of the Chinese Hα Solar Explorer (CHASE) and Qilu-2/3 satellites. The feasibility and efficiency of the configuration layout design have been validated through flight tests, achieving effective integration and flexible deployment of satellites with different installation methods. As a mature liquid launch vehicle of conventional propellant, the application of rideshare mode further enhances the mission adaptability of the Long March-2D launch vehicle and can reduce the launch cost of a single satellite.