中国科技核心期刊
中国指挥与控制学会会刊
军事装备类重点期刊
中国指挥与控制学会会刊
军事装备类重点期刊
中国指挥与控制学会会刊
军事装备类重点期刊
With the success of AlphaGo, deep learning received widely attention. Deep learning has excellent feature extraction, multi-layer learning and expression ability. It provides a new idea for the research and development of military intelligent decision support systems (MIDSS). The features and key technologies of deep learning is presented and the development course of MIDSS is reviewed. Also, this paper describes the development of military intelligent decision assisted technology based on deep learning and prospects some of future work.
With the Joint operations command and control system as the research object, the topology structure of the command and control system was posed. In the Complex point of view, abstracted the information acquisition, information fusion, command decision and task execution unit and command information flow to the vertexes and edges in the system. A network model of command and control system composed of information network, command and control network and fire network was constructed. The concept of basic command chain and generalized command chain of Joint operations command and control system was put forward, and the effectiveness parameters of complex network were expanded on this basis. The evaluation method of network efficiency, such as command information efficiency, command reorganization capability and command node betweenness was put forward. Then compared and analyzed the measure of the three models. It was a useful exploration of performing efficiency Joint operations command and control system.
The research on the operational mechanism of the operation system for knowledge graph is an important part in the research of operation system. It is the basis of the knowledge graph building, the operation and reasoning of the operation system. In this paper, starting from the concept model of operation system, according to semantic, process, mission, and uncertainty aspects of the mechanism, we provide the ontology description of operational mechanism of system operation, rule driven mechanism description method and task driven mechanism description method.
The knowledge of armament strongly supports the demonstration of weapon and equipment and military operation. The obtainment, storage and research of the knowledge of armament tend to involving large volume of data, forming complex structure and becoming less intuitive. The essay illustrates the forming method of the knowledge graph of armament based on open source data and establishes the whole knowledge graph’s building framework which includes two main parts: the armament information extraction and the graphing of knowledge graph. The detailed steps are also demonstrated specifically.
In this paper, a new metric, i.e. loop betweenness, is proposed to evaluate the node importance of complex operational network by referring to the definition of the betweenness. We use the program to find all the loops in the operational network and calculate the loop betweenness of the operational nodes, and analyzes several kinds of special operational networks. Through comparison with other evaluation methods, proves the effectiveness of the method for complex operational network.
Aiming at the problem caused by the environment changes of complex software system, a concept model about self-refactoring of complex software system is proposed. Technology of software self-refactoring is taken into complex software system field, software system can take appropriate strategies of self-refactoring when inside or outside environment of system changes. Meanwhile a method of software execute based on redundant backups is presented to make sure critical task running without any break.
Multi-layer missile defense system works through the whole process of the ballistic missile. Compared with midcourse defense and terminal defense, boost-phase defense is the first layer. The characteristic of boost phase ATBM and requirement for the early warning system are analyzed firstly. Furthermore, the effectiveness and protecting range of BPI are calculated. At the end, the combating process of airborne boost-phase interception weapon system is analyzed. A whole scheme is achieved for the airborne boost-phase interception. According to the elementary analysis of the boost phase interception flight sequence scheme of the airborne boost phase interceptor is proposed.
Through analysising and comparing the characteristic of earthquake tsunami, the non-linear shallow water equation of earthquake tsunami motion is improved firstly. Secondly, the calculate equation of earthquake tsunami initial position is choosed and proved. Finally, the model of movement in the earthquake tsunami based on various kinds state-model is established, and the genealregulation when submarine encounteringearthquake tsunami through simulating four typical movement states of submarine is got. All these researches canprovidetheoretical support for commanding-office making replytsunami plan.
The intelligence data accumulated by modern intelligence system has the characteristics of rapid accumulation, various types and complicated relations. It is of great significance to fully exploit and effectively utilize the potential value of these data in the military field. This paper presents a general software framework of data analysis for massive data, and analyzes the key technologies to be solved. These can be used to provide technical support for further improving processing capability of Naval intelligence.
At present, the threaten from low-slow-small UAV acting on key points is much bigger than before. And it is difficult to detect and intercept this kind of target. This paper summarizes the methods of detection and defense measures. Taking the advantages and disadvantages of each facility into consideration, we propose a scheme of Anti-UAV system for key point defense and present a brief introduction for the key technologies. The result of demonstration of some facilities in the system is also briefly introduced in this paper. At last, we make a analysis and conclusion of the situation and mission in Anti-UAV fight.
Based on the differential evolution algorithm (DE) and quantum particle swarm optimization algorithm (QPSO), this paper proposes an improved quantum particle swarm optimization algorithm, which is successfully applied to AUV route planning in the marine environment. Simulation results show that in different battlefield environment (different launch points, different target points, different detection threats), the improved quantum particle swarm optimization algorithm has the advantages of strong searching efficiency, fast convergence speed and good algorithm stability, which can be well adapted to AUV route planning.
A hierarchical optimization strategy based on path planning and trajectory planning process is presented in this paper to solve the trajectory planning problem of multiple unmanned aerial vehicles(UAV). Firstly, a model of path planning considering obstacle constrains is established, and a modified Sparse A * Algorithm (SAS) is proposed to shortcomings of fixed step adaptability to environmental changes. Secondly, The trajectory planning strategy is established based on Decentralized Model Predictive Control(DMPC). Lastly, the hierarchic optimization strategy is simulated with MATLAB. Compared with the traditional trajectory optimization methods, the strategy achieves a balance between better solution and real-time performance with stronger adaptability.
Radar image and electronic chart are the important sources of information for ship sailing. Now for the problems that radar image and electronic chart match not accurately because of the noise, isolated points, blocked targets. The matching algorithm of partial Hausdorff distance is improved, which is applied to the matching of the edge feature of the extracted coastline from radar image and the electronic chart. Through the setting of the two parameters, the accuracy of the matching is greatly improved, to achieve the maritime security in real-time accurate positioning.
It is difficult to realize effective detection of distant targets due to the limited range of the active sonar caused by the attenuation of the sound signal during underwater transmission. Passive sonar detection has the disadvantage that it can only obtain the target position and can not obtain the target distance which seriously affects the positioning of the underwater target.The underwater target localization algorithm based on the explosive sound source uses the explosive sound as the sound source to provide the target echo signal for the passive sonar, such as the position of the explosive sound source, the target echo azimuth and the echo time interval. Target location information can be calculated by the triangulation method. The algorithm has high positioning accuracy and can provide accurate target position information for the operations of long-range anti-submarine weapons in anti-submarine weapon systems.
The key to the integrated control of multi-ship air missile weapon system lies in the organization, resource scheduling and target allocation of fire channel of multi-ship air missile. This paper discusses the type of ship to air missile firepower channel more dynamic organization and the method of resource scheduling by giving an example of two different types of ship to air missile. According to the features and performance of these two kinds of missiles, the paper solved the optimal scheme of firepower channel target distribution with discrete artificial firefly algorithm in terms of considering the rule of fire distribution and giving the most combat effectiveness of the premise.
As time goes on, the accuracy of Monte Carlo algorithm will get higher and higher, which is also the case when calculating the probability of finding the wire-guided torpedo. However, on the battlefield, it is often necessary to obtain a certain degree of detection probability of a wire-guided torpedo in a short period of time so as to seize the opportunity of enemy control. In this paper, a fast Monte Carlo-based method for discovering probability of wire-guided torpedoes is proposed. The new method uses the Monte Carlo algorithm to simulate the time distribution of each calculation module when calculating the torpedo discovery probability, and optimize the most time-consuming algorithm to achieve the purpose of shortening the time-use. The simulation results show that the fast solution method can shorten the solving time of the Monte-Carlo algorithm for the discovering probability of wire-guided torpedoes under the premise of ensuring the accuracy of the results.
With the improvement of intelligent torpedoe abilities in detecting and tracking, traditional weapons have difficulty getting rid of the attack from coming torpedoes. The world navy has devoted to researching and developing anti-torpedo torpedoe(ATT)as a new killing technology. This paper focuses the fighting style and operating flow of ATT under typical modes of normal and emergency. Then analyses will be conducted on head-on interception and direct interception about application conditions and efficiency. In the end, simulations of ATT fighting method are carried based on system efficiency. Preliminary conclusion shows that: it can intercept coming torpedoes on head or directly (first choose direct interception) in sonar active and active-passive composite detection due to the high target motion elements resolution precision based on comprehensive target information. However, weapon system can only draw estimations of motion elements from target azimuth information conducts intercepting method of the position line.While finding a coming torpedoe urgently at close range,it intercepts coming target at current position.
As the maritime experiment task of the seeker simulation system becomes more and more important, the research on its effectiveness evaluation is also imminent. In this paper, the synthesis method of effectiveness evaluation based on WBS modeling, ADC evaluation and AHP weight calculation is presented. The time function curve of system effectiveness can be obtained by calculation. Finally, a simulation example of effectiveness evaluation of the seeker simulation system for a typical maritime experiment is given. An intuitionistic quantitative analysis can be given before the task begun, that can help decision and maintenance.
For operational experiment application is a process of energy accumulation and release, to effectively understand the winning mechanism and solve the fog of war, A Space-time Logic wargaming is proposed, which is focused on the process of battle space perception, energy savings, transfer and transformation in various stages of the war to potential deduction spectrum model, optimization of disposition to realize the battlefield. Considering the war can perceive and battlefield forecast demand, proposed method of constructing prediction model in spatial domain based on convolutional neural network, to sensing energy and organic combination of historical data, accurately grasp the current state of other operations, provide support for the decision of one’s own next step; using reinforcement learning method to construct the war energy spectrum and decision network, then the decision model, value model of organic combination, designed the parallel Monte Carlo tree search algorithm for optimization plan of action. Finally, an example of a submarine’s "red blue" rivalry is taken as an example to verify the effectiveness of the method.
This paper presents a design scheme based on Unity3D engine for real-time maritime battlefield 3D situation display system. First of all, the paper briefly analyzes the features and advantages of display system based on Unity3D engine, elaborates the method of building a display system with Unity3D,and then analyzes the key technologies of optimization of database connection, timing driving and attitude adjustment in detail. At last,an example system is build and tested. The test results show that the system can achieve the purpose of real-time display of three-dimensional situation.
In order to improve the performance of sound pressure vertical array, optimize the hydrophone array structure, the time-reversal passive location technology with vector vertical array is researched and three methods are given in this paper. We compared the vector vertical array and the sound pressure vertical array in processing gain of TRM through computer simulation, and analyzed the relationship between positioning performance and parameters such as SNR, number of array elements and spacing between elements. The simulation results show that, when the detection probability is 0.8, the detection threshold of vector vertical array is 1dB-3 dB better than that of sound pressure vertical array. For the wideband signal of 2kHz-4kHz, the vector vertical array can obtain better time-reversal passive location effect by using four elements array with 2 meters element equal spacing.
Aimming at the problem of target damage estimation on shipborne anti-missile simulation of gun, this paper analyses the damage factors of shipborne anti-missile gun based on the lackage of estimation approaches on hitten parts, proposes the method of target damage estimation from anti-missile simulation of gun based on target practice data, sets up the relative models and offers the emulational procedure. Simulation practice shows that this method can accurately simulate target damange effect.
Traditional corner point identification methods have some disadvantages, such as high computational complexity and high error detection rate. For these problems, this paper proposes an improved corner detection algorithm based on Hessian matrix. The algorithm adopts adaptive threshold method to identify angular point location and circular template symmetric templates to eliminate pseudo angular point. It improves the traditional fixed threshold detection method for missing pseudo Angle point. Experiments show that the proposed algorithm has better flexibility and adaptability, and can still maintain a high focus recognition accuracy under different working conditions.
The sidelobe effect of synthetic aperture radar (SAR) image affects the resolution and dynamic range of the image. Classical windowing technique can cause the mainlobe broadening and reduce the image resolution. For this reason, the apodization filtering technique is used to suppress the sidelobe of the image, noninteger Nyquist sidelobe apodization filtering method is improved. Simulation results and measured data processing results show that the proposed method can well suppressed the sidelobes and improve the resolution and dynamic range of the image.
The stacker is the core component of the automatic intelligent storage system, and the data of the stacker is an important part of the large data of the power intelligent operation and maintenance warehouse. With the rapid intelligence of the warehouse equipment, the collection, processing and analysis of the stacker's state data are of great significance to the normal operation of the automatic intelligent storage system. In order to dig out the abnormal data from the stacker state data, warning of stacking in actual engineering machine, to prevent major failure of stacker. Firstly, using the mean to smooth the data smoothing, and a pre-processing model for smoothing data is constructed. Secondly, using box-plot method after pre-processing to dig out abnormal screening interval, and a interval automatic identification model is established. Finally, the algorithm is verified by using the the stacker operation data obtained from automatic intelligent storage system, and then the correctness and effectiveness of the algorithm are proved.
For the DS communication system, DS spread spectrum communication within the allocated bandwidth means that the service information to be transmitted is extended to the carrier frequency corresponding to the carrier frequency band allocated by the satellite transponder to the shipborne station by the pseudo-random code, and then modulated and frequency-converted Sent to the wireless channel, the receiver at the carrier station by DS despreading processing, the input broadband useful signal spectrum is restored to the original signal bandwidth, while the wider band interference signal spectrum is further expanded, Narrowband filter after DS despreading module can filter most interference signals [1, 2], so as to achieve the suppression of interference by DSSS. However, the interference effects of DSSS systems with different interference patterns are not the same, and the concrete analysis is performed according to the specific interference signal patterns. In this paper, the main analysis for the suppression of coercion: point frequency interference, multi-frequency interference, part of the band interference, full-band interference on the DSSS technology interference analysis.
This paper construct a model of altitude solving and output for measuring initial altitude of vehicle in battery in moving base. The realtime altitude model of vehicle is described with some constrainst include the ship platform compass, the assembly of launching set and the target azimuth. The geometric firing model aimed at fixed geographic coordinates point is presented and the lead-angle is calculated. The mechanical gyroacope measuring model is discussed based on the firing altitude, constant-height flight altitude of vehicle and the assembly in vehicle body of mechanical gyroacope, the output of fire control director mode is presented adapted to control structure of vehicle. Finally, the digit Trajectory computation in several conditions showed that the output ploarity and amplitude of this model is correct and validity.
