Real-time UAV path planning based on LSTM network

To address the shortcomings of single-step decision making in the existing deep reinforcement learning based unmanned aerial vehicle(UAV)real-time path planning problem,a real-time UAV path planning algorithm based on long shortterm memory(RPP-LSTM)network is proposed,which combines the memory characteristics of recurrent neural network(RNN)and the deep reinforcement learning algorithm.LSTM networks are used in this algorithm as Q-value networks for the deep Q network(DQN)algorithm,which makes the decision of the Q-value network has some memory.Thanks to LSTM network,the Q-value network can use the previous environmental information and action information which effectively avoids the problem of single-step decision considering only the current environment.Besides,the algorithm proposes a hierarchical reward and punishment function for the specific problem of UAV real-time path planning,so that the UAV can more reasonably perform path planning.Simulation verification shows that compared with the traditional feed-forward neural network(FNN)based UAV autonomous path planning algorithm,the RPP-LSTM proposed in this paper can adapt to more complex environments and has significantly improved robustness and accuracy when performing UAV real-time path planning.

Mechanical Properties and Thermal Shock Resistance of SrAl_(2)Si_(2)O_(8) Reinforced BN Ceramic Composites

Hexagonal boron nitride(h-BN)ceramics have become exceptional materials for heat-resistant components in hypersonic vehicles,owing to their superior thermal stability and excellent dielectric properties.However,their densification during sintering still poses challenges for researchers,and their mechanical properties are rather unsatisfactory.In this study,SrAl_(2)Si_(2)O_(8)(SAS),with low melting point and high strength,was introduced into the h-BN ceramics to facilitate the sintering and reinforce the strength and toughness.Then,BN-SAS ceramic composites were fabricated via hot press sintering using h-BN,SrCO_(3),Al_(2)O_(3),and SiO_(2) as raw materials,and effects of sintering pressure on their microstructure,mechanical property,and thermal property were investigated.The thermal shock resistance of BN-SAS ceramic composites was evaluated.Results show that phases of as-preparedBN-SAS ceramic composites are h-BN and h-SrAl_(2)Si_(2)O_(8).With the increase of sintering pressure,the composites’densities increase,and the mechanical properties shew a rising trend followed by a slight decline.At a sintering pressure of 20 MPa,their bending strength and fracture toughness are(138±4)MPa and(1.84±0.05)MPa·m^(1/2),respectively.Composites sintered at 10 MPa exhibit a low coefficient of thermal expansion,with an average of 2.96×10^(-6) K^(-1) in the temperature range from 200 to 1200℃.The BN-SAS ceramic composites prepared at 20 MPa display higher thermal conductivity from 12.42 to 28.42 W·m^(-1)·K^(-1) within the temperature range from room temperature to 1000℃.Notably,BN-SAS composites exhibit remarkable thermal shock resistance,with residual bending strength peaking and subsequently declining sharply under a thermal shock temperature difference ranging from 600 to 1400℃.The maximum residual bending strength is recorded at a temperature difference of 800℃,with a residual strength retention rate of 101%.As the thermal shock temperature difference increase,the degree of oxidation on the ceramic surface and cracks due to thermal stress are also increased gradually.

Situational continuity-based air combat autonomous maneuvering decision-making

In order to improve the performance of UAV's autonomous maneuvering decision-making,this paper proposes a decision-making method based on situational continuity.The algorithm in this paper designs a situation evaluation function with strong guidance,then trains the Long Short-Term Memory(LSTM)under the framework of Deep Q Network(DQN)for air combat maneuvering decision-making.Considering the continuity between adjacent situations,the method takes multiple consecutive situations as one input of the neural network.To reflect the difference between adjacent situations,the method takes the difference of situation evaluation value as the reward of reinforcement learning.In different scenarios,the algorithm proposed in this paper is compared with the algorithm based on the Fully Neural Network(FNN)and the algorithm based on statistical principles respectively.The results show that,compared with the FNN algorithm,the algorithm proposed in this paper is more accurate and forwardlooking.Compared with the algorithm based on the statistical principles,the decision-making of the algorithm proposed in this paper is more efficient and its real-time performance is better.

Crushing Mechanism of Spherical Tungsten Alloy Fragments Penetrate Thick Steel Plate Target

Against protection requirements for high-speed fragments on the ground weapons,we carried out the research work of crushing mechanism at different impact speeds ofφ8.7 mm spherical tungsten alloy,the penetration to 603 armor steel was completed by 20 mm ballistic gun,and the ANSYS/LS-DYNA software was used to complete the numerical calculation of the penetration.We find that there are different crushing mechanisms of spherical tungsten alloy with different speeds and low speed,the crushing mechanism of fragment is mainly controlled by overall plastic deformation,shearing stripping,and squeezing at a high pressure and a high speed.The crushing mechanism will have a spallation phenomenon in addition to the crushing mechanism under high pressure.

Anti-interference self-alignment algorithm by attitude optimization estimation for SINS on a rocking base

The performance of a strapdown inertial navigation system(SINS)largely depends on the accuracy and rapidness of the initial alignment.A novel anti-interference self-alignment algorithm by attitude optimization estimation for SINS on a rocking base is presented in this paper.The algorithm transforms the initial alignment into the initial attitude determination problem by using infinite vector observations to remove the angular motions,the SINS alignment is heuristically established as an optimiza-tion problem of finding the minimum eigenvector.In order to further improve the alignment precision,an adaptive recursive weighted least squares(ARWLS)curve fitting algorithm is used to fit the translational motion interference-contaminated reference vectors according to their time domain characteristics.Simulation studies and experimental results favorably demonstrate its rapidness,accuracy and robustness.

Experimental Simulation for Fracture of Gun Propellant Charge Bed

The simulation of compression and fracture of charge bed in chamber is one of the key problems in the study of launch safety of gun propellant charge. A new kind of experimental device that can be used for simulation is given. Its structure and operational principle are introduced. Using a semi-closed vessel as a source of compression force, the device can simulate any kind of dynamic environment in a gun propellant charge. Using the low temperature inert gas （N2） as the compression medium, the device can not only ensure that the simulation is real, but also protect the fragmentized propellant from combustion after experiment. Using the device, many simulation experiments have been accomplished, and dynamic environment of propellant fracture is acquired. With the experiments, fragmentized propellant for the compression and fracture of charge bed is obtained. Results of experiments show that the new device can be used to study the principle of the compression and fracture of charge bed.

In situ synthesis and hardness of TiC/Ti_5Si_3 composites on Ti-5Al-2.5Sn substrates by gas tungsten arc welding

TiC/TisSi3 composites were fabricated on Ti-5A1-2.5Sn substrates by gas tungsten arc welding （GTAW）. Identification of the phases was performed using X-ray diffraction （XRD）. The microstructures were analyzed using scanning electron microscopy （SEM） combined with energy-dispersive X-ray spectrometry （EDS） and optical microscopy （OM）. The Vickers hardness was measured with a micro-hardness tester. The TiC/TisSi3 composites were obtained in a double-layer track, and the Vickers hardness of the track increased by two to three times compared with the Ti-5A1-2.5Sn substrate.

Effects of ZrO_2 Nanoparticles on the Microstructure and Thermal-protective Properties of PEO Coating on Al-12.5%Si Alloy

PEO ceramic coatings including ZrO_2-Al_2O_3-SiO_2 in three phases were prepared on an Al-12.5%Si alloy in electrolyte solutions containing ZrO_2 nanoparticles. The microstructures and phases of the coatings were analyzed by SEM and XRD, and the heat insulation performance and the thermal shock resistance of the coatings were investigated. The compactness of the coating increased significantly and the hindrance of the Si element on plasma electrolytic oxidation process was effectively weakened. The growth rate of the coating was improved substantially with the addition of ZrO_2 nanoparticles. The PEO ceramic coatings are primarily composed of SiO_2 and high temperature steady phases such as a-Al_2O_3 and c-ZrO_2. Both the content of c-ZrO_2 and the heat-insulating property of the coating increased significantly. The ceramic coatings with special microstructure and composition formed in the solutions containing ZrO_2 nanoparticles possess excellent heat insulation performance and thermal shock resistance.

Morphology and tribological properties of Ni/n-SiO_2 composite coatings by pulse-reverse current brush-plating

Ni/n-SiO2 composite coating was electrodeposited by brush-plating with pulse-reverse current（RC）. The morphology, hardness, and tribological properties of the coating were investigated and compared with those of Ni and composite coatings electrodeposited with direct current（DC）. The results indicate that Ni/n-SiO2 composite coating electrodeposited by RC, because of RC and the nano powders, has denser coating, finer crystal grains, higher hardness（HV650.0, nearly 1.5 times higher than that of Ni coating electrodeposited by DC） and lower friction coefficient（nearly 0.62）, as a result, in the wearing experiment, the Ni/n-SiO2 composite coating electrodeposited by RC has the least worn loss. So this kind of coating has better wear resistance. And RC electro brush-plating can be used as a new technology of brush-plating in the area of wear resistance.

Wireless Relay Communication on High Speed Railway：Full Duplex or Half Duplex？

In this paper,we introduce one full-duplex(FD) relaying transmission scheme for high speed railway and compare its ergodic capacity and outage performance with half-duplex(HD) relaying scheme. Both decode-and-forward(DF) amplify-and-forward(AF) relay modes are considered. Moreover,the carriage penetration loss(CPL) and the self-interference(SI) cancellation ratio are investigated. We derive the closed-form expressions for the outage probability and ergodic capacity of both HD and FD relay transmission schemes. It is shown that when CPL is larger than a certain level,the FD relay can achieve better performance in terms of capacity and keep the outage probability in a low level. It is also found that three factors: AF or DF modes,CPL and SI cancellation ratio,can have impact on capacity performance,and that CPL is one decisive factor. Our results can provide theoretical supports for development and deployment of future wireless communication systems on high speed railways.

Traffic Anomaly DetectionMethod Based on Improved GRU and EFMS-Kmeans Clustering

In recent years,with the continuous development of information technology and the rapid growth of network scale,network monitoring and management become more and more important.Network traffic is an important part of network state.In order to ensure the normal operation of the network,improve the availability of the network,find network faults in time and deal with network attacks;it is necessary to detect the abnormal traffic in the network.Abnormal traffic detection is of great significance in the actual network management.Therefore,in order to improve the accuracy and efficiency of network traffic anomaly detection,this paper proposes a comprehensive anomaly detection method based on improved GRU traffic prediction and improved K-means clustering,and cascade the traffic prediction and clustering to achieve the purpose of anomaly detection.Firstly,an improved highway-GRU algorithm HS-GRU(An improved Gate Recurrent Unit neural network based on Highway network and STL algorithm,HS-GRU)is proposed,which combines STL decomposition algorithm with highway GRU neural network and uses this improved algorithm to predict traffic.And then,we proposed the EFMS-Kmeans algorithm(An improved clustering algorithmthat combined Mean Shift algorithmbased on electrostatic force with K-means clustering)to solve the shortcoming of the traditional K-means clustering which cannot automatically determine the number of clustering.The sum of the squared errors(SSE)method and the contour coefficient method were used to double test the clustering effect.After determining the clustering center,the potential energy gradient was directly used for anomaly detection by using the threshold method,which considered the local characteristics of the data and ensured the accuracy of anomaly detection.The simulation results show that the anomaly detection algorithm based on HS-GRU and EFMS-Kmeans clustering proposed in this paper can effectively improve the accuracy of flow anomaly detection and has important application value.

Research on the Application of Explosive Network in the Shaped Charge Warhead

With respect to the problems of that the shaped charge warhead currently uses a cover method to improve the penetration power, a method using an explosive network technology as the detonation mode of shaped charge warhead is proposed. In the context of some shaped charge warhead, a synchronous explosive network prototype is designed according to some charge structure parameters, such as the liner and main grain. From the performance comparison test, it can be known that the explosive network not only stably detonates, but also largely improves the penetration power and stability. Experimental results show that explosive network technology is an effective method for improving the penetration power. The results lay a solid foundation for the engineering application of the technology in the shaped charge warhead.

Online hierarchical recognition method for target tactical intention in beyond-visual-range air combat

Online accurate recognition of target tactical intention in beyond-visual-range (BVR) air combat is an important basis for deep situational awareness and autonomous air combat decision-making, which can create pre-emptive tactical opportunities for the fighter to gain air superiority. The existing methods to solve this problem have some defects such as dependence on empirical knowledge, difficulty in interpreting the recognition results, and inability to meet the requirements of actual air combat. So an online hierarchical recognition method for target tactical intention in BVR air combat based on cascaded support vector machine (CSVM) is proposed in this study. Through the mechanism analysis of BVR air combat, the instantaneous and cumulative feature information of target trajectory and relative situation information are introduced successively using online automatic decomposition of target trajectory and hierarchical progression. Then the hierarchical recognition model from target maneuver element, tactical maneuver to tactical intention is constructed. The CSVM algorithm is designed for solving this model, and the computational complexity is decomposed by the cascaded structure to overcome the problems of convergence and timeliness when the dimensions and number of training samples are large. Meanwhile, the recognition result of each layer can be used to support the composition analysis and interpretation of target tactical intention. The simulation results show that the proposed method can effectively realize multi-dimensional online accurate recognition of target tactical intention in BVR air combat.

A Virtual Assembly System for Aero-engines Based on VR

In this paper,a synthesized technique is explored through design of a virtual experiment system for main aero-engine based on VR,which attempts to integrate some VR techniques,such as 3D stereo simulation and modeling,3D scene management and human-computer virtual assembly based on data glove and collision detection.The geometry modeling and instantiation technique,virtual hand modeling and data glove interaction technique are discussed chiefly.Finally,taking aim at assembling and dismantling experiment system of a certain aero-engine,interface technology is expounded also.

Biodegradability and Tribological Properties of Mineral Base Oil Enhanced by Caprylic Methyl Diethanolamine Phosphate Ester

The influence of synthetic caprylic methyl diethanolamine phosphate ester (abbreviated as MDEACP) on biodegradability and tribological properties of 400SN mineral base oil was studied. The biodegradability of the neat base oil and the oil doped with MDEACP was determined on a biodegradation tester. The tribological properties of the neat base oil and the oil doped with MDEACP were evaluated on a four-ball tester. Moreover, the worn surfaces were investigated by scanning electron microscopy (SEM) and X-ray photoelectron spectroscopy (XPS). The results revealed that MDEACP significantly promoted the biodegradation of the mineral base oil. The improvement in biodegradability was attributed to the enhanced growth and quantity of microbes by MDEACP. Furthermore, MDEACP enhanced the anti-wear properties, the friction-reducing properties, and the extreme pressure properties of the base oil. It was mainly attributed to the formation of the complex boundary lubrication film resulted from the adsorption and tribochemical reactions of MDEACP on the friction surface.

Description method of conceptual model based on the ontology

The conceptual model is the first time abstraction ceptual model description method based on the UML and the of the real world. The traditional con- text usually has the semantic ambiguity problem. The formal description method lacks the mechanism of integrity and consistency check. Therefore both of them cannot support the simulation scenario description and the model compos- ability. A simulation conceptual description method based on ontology is put forward. According to the OWL language which was put forward by the W3C organization, a conceptual model description language is established. The OWL language semantic element in the description function in the mili- tary conceptual model is introduced. Then two layers of the military conceptual model framework are built up based on the general ontology and the domain ontology. In the general ontology, the class and property definition are given. While in the domain ontology, they are extended and a relat- ed example is given. The technique advantages of our method include the mechanism of conceptual model consistency and integrity check, the support of composable simulation models, and automatic creation of simulation scenario.

Analytical measurement method for situation elements' dynamic characteristics

This paper gives an analysis of the dynamic characteristics of situation elements(SEs) in situation awareness(SA)research. The purpose of the discussion is to understand the factors that influence SA and to help in designing the training systems to improve operators’ SA. The status function of SEs is defined and the derivative of the function represents trends of the status of SEs at each moment. Then, Fourier transform(FT) is used to give the frequency-domain function in terms of the time-domain status function. In frequency domain, the bandwidth of the status function is used as a criterion to characterize the notion of "fast" and"slow" of the change of SE’s status, which represents the dynamic characteristic of SEs. The criterion constitutes the first analytical measurement of the dynamic characteristic of SEs, which is one of the important factors that influence the SA process.

Repairable Configuration Optimization Model under Two-Echelon Inventory

Spares inventory configuration optimization is an effective way to improve readiness and reduce life cycle cost of equipment.Through analyzing two-echelon spares support system,the METRIC model basic theory was used.An inventory configuration optimization model of two-echelon spares support system was proposed which took the spares expected shortfall as the object and made the minimum repairable parts expected shortfall instead of the maximum spares supportability as the objective function.Marginal efficiency analysis algorithm was applied to optimizing the spares configuration and generating a rational spares inventory configuration.Finally,several examples are given to verify the model.

Semi-physical Simulation Platform of a Parafoil Nonlinear Dynamic System

Focusing on the problems in the process of simulation and experiment on a parafoil nonlinear dynamic system,such as limited methods,high cost and low efficiency,we present a semi-physical simulation platform.It is designed by connecting parts of physical objects to a computer,and remedies the defect that a computer simulation is divorced from a real environment absolutely.The main components of the platform and its functions,as well as simulation flows,are introduced.The feasibility and validity are verified through a simulation experiment.The experimental results show that the platform has significance for improving the quality of the parafoil fixed-point airdrop system,shortening the development cycle and saving cost.

Speech detection method based on a multi-window analysis

Aiming at the poor performance of speech signal detection at low signal-to-noise ratio(SNR),a method is proposed to detect active speech frames based on multi-window time-frequency(T-F)diagrams.First,the T-F diagram of the signal is calculated based on a multi-window T-F analysis,and a speech test statistic is constructed based on the characteristic difference between the signal and background noise.Second,the dynamic double-threshold processing is used for preliminary detection,and then the global double-threshold value is obtained using K-means clustering.Finally,the detection results are obtained by sequential decision.The experimental results show that the overall performance of the method is better than that of traditional methods under various SNR conditions and background noises.This method also has the advantages of low complexity,strong robustness,and adaptability to multi-national languages.