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
In order to sort out the difficulties in the application of deep learning technology in the military field and clarify the direction of deep learning in the application of public relations in the military field, this paper firstly summarizes the application status of deep learning technology in the military field from three aspects: target recognition, situation awareness and command decision-making, and then analyzes the difficulties and challenges in the application of deep learning technology in the military field. In the field of target recognition: deep learning technology for scarce cognitive samples, deep learning technology under uncertain information, its real-time and deep learning based on unmanned platform need to be broken through. In the field of situation awareness: the representation and mining technology of big data characteristics of battlefield situation based on deep learning, and the technology of battlefield situation understanding need to be broken through. In the field of command and decision-making: the solvability of deep learning needs to be improved, and the multi-entity collaborative decision-making technology and reasoning decision-making technology need to be improved. The research results can provide reference direction for the innovation and development of deep learning technology in the military field and engineering research.
The paper discusses the concept, characteristics and development trend of big data, and analyzes the significance of big data technology for electronic equipment test. Then based on the general framework of big data, combined with the status quo and characteristics of electronic equipment test, the basic framework of the big data system of electronic equipment test is constructed. At last, according to the main lines of data base planning, data warehouse construction, online analysis, data mining and intelligent decision-making, the preliminary discussion on the construction and application of big data in the field of electronic equipment test is carried out.
Based on the game between pedestrians and terrorists in terrorist attacks, the social force model of pedestrian-terrorist system and the social force model of main factor terrorist attack system are established.Cellular automata simulation of cellular network is used to realize the visualization the process of terrorist attack,and the simulation shows pedestrians are the distribution of column, anchor and shield,the pedestrian victim effect and the terrorist goalkeeper effect are obtained.The relationship between pedestrian evacuation time, escape strategy, the number of terrorists and the number of pedestrian casualties is analyzed.The conclusions show that the average combat effectiveness of terrorists is the main factor in the number of pedestrian casualties. The best escape strategy in terrorist attacks of pedestrians is to stay away, take appropriate risks and run resolutely.
According to practical demand for effective modeling of behavior rules of combat entities, this paper analyzes the possibility of describing behavior rules of combat entities based on decision tree, puts forward a formal description method for behavior rules of combat entities based on decision tree, designs a framework of the modeling tools for behavior rules of combat entities based on decision tree, and provides theoretical guidance for carrying out further research and engineering practice.
Combining the surface ships strength in anti-submarine operations, this paper creates a model of attack intention inference based on the intuitionistic fuzzy inference method. Qualitative reaction is used to describe the uncertain events, and the inference progress is developed through the operation progress. The assumed case has proved that this model can describe uncertain events in battlefield effectively, reduce the fuzziness. This model is in accordance with the commander thought process; it can be adopted to infer military intention, and has some important significance in assistant decision.
The large warship fleet HF frequency hopping network is researched. According to the complex characteristics of the HF frequency hopping, the large warship fleet HF frequency hopping, combining the tree type, star type and quasi grid network, are built. In the network, the point-to-point frequency hopping is taken in the medium and remote distance communication, which uses the real-time AFH technology based on the asymmetric frequency table technology. In the short range HF frequency hopping communication, high-density orthogonal networking technology is adopted based on the symmetric crossed frequency table technology. The research results have important theoretical research value in promoting the large warship fleet HF communication network's anti-reconnaissance and anti-jamming abilities.
In order to study the applicability of particle swarm optimization (PSO) in cooperative detection task allocation, the multi-sensor cooperative detection task allocation is described as a typical multi-constrained integer programming problem. A particle encoding scheme based on the feasible region of tracking and detection of detection equipment is constructed to reduce the solution space of the planning problem and improve the efficiency of the algorithm. An improved PSO algorithm based on diversity control strategy is adopted to solve the premature phenomenon in PSO and get rid of the limitation of local optimal solution. Taking the cooperative target detection task of "low slow and small (LSS)" UAV as an example, its detection efficiency and constraints is discussed, and its task assignment is simulated and analyzed. The result shows that the algorithm can speed up particle search and avoid premature phenomena. It can provide reference for multisensory cooperative detection task allocation.
Aiming at the problem of target detection rate and low accuracy in the high-altitude remote sensing image intelligence support, this paper proposes an improved YOLOv3-RE algorithm based on the YOLOv3 framework. The algorithm uses res2block to improve the original feature extraction network, realize multi-scale extraction of target features, and overcome the problem of missing features due to fewer pixels. Secondly, for the problem of high-altitude remote sensing small target, the fourth detection scale is added to the detection network. Tests are carried out using standard data sets. The results show that the accuracy of YOLOv3-RE is nearly 5 percentage points higher than that of YOLOv3, which is an important contribution to the high-altitude remote sensing image intelligence support business.
Aiming at the low correlation accuracy for classical track association algorithm in target-intensive, track-crossing or bifurcation environment,a method of multi-sensor track association based on set pair analysis theory and double threshold is presented. The algorithm is divided into two stages. Firstly,the coarse correlation is performed by the wave-gate method to obtain the preliminary association set. Then the track information of the two sensors is treated as a set pair in the form of set pair based on the SPA closeness,and the expression of connection degree between tracks is built. Finally, in order to improve the association accuracy rate, the set pair closeness matrix is established at multiple times,and the double threshold algorithm is used to judge the closeness matrix of multiple moments to obtain the correlation result. The simulation results show that the algorithm has higher correlation accuracy than the nearest neighbor trajectory association algorithm and fuzzy double threshold association algorithm.
In this paper, the accuracy and robustness of monocular camera positioning in low-texture environment or camera rotation are presented. A monocular visual inertial odometry (VIO) method with integrated line features is presented. The Dog-Leg method is used instead of the Levenberg-Marquard method for nonlinear solution, which reduces the amount of computation. The experimental results based on the measured data also show that the positioning method can achieve higher positioning accuracy than the VINS-Mono algorithm and the OKVIS-Mono algorithm.
In order to carry out reliability prediction of software-hardware integrated C4ISR system, the complexity of each component of the C4ISR system needs to be determined. According to the composition characteristics of C4ISR system, The methods to determine the complexity of hardware unit, software unit and hardware-software combination unit are given respectively. A method to determine the complexity of component counting method for hardware unit is presented, The arithmetic is proposed to control the scoring error; The main factors affecting the complexity of software unit are presented, and the quantification method of the main factors affecting the complexity of software unit is given. On this basis, the complexity of software unit is synthetically given. According to the composition characteristics of hard and soft elements, a method to determine the weighted complexity of hard and soft elements is proposed.
In the unmanned traction mode of carrier-based aircraft, it is necessary to automatically generate carrier flight deck transit routes to assist decision-making. This paper is devoted to the study of collision avoidance and transshipment routes of carrier aircraft on flight deck. Firstly, the plane polygon method is used to describe the ship-borne aircraft on the flight deck. At the same time, a new polygon collision detection method is proposed. Secondly, the mathematical model of collision avoidance and transshipment of carrier aircraft on flight deck is established. Then, the new collision detection method is applied to the mathematical model, and the particle swarm optimization algorithm is used to optimize the solution. The simulation results show that this method can solve this kind of collision avoidance problem well, and the optimal path meets the actual needs.
Kinematics analysis of multi-joint robot play an important part in robotics research. Because of the nonuniqueness of its inverse solution, robot kinematics with more than 6 freedom degrees is still the focuses of the research. For the inverse kinematics problem, some optimization methods,like neural network, are mostly used. These methods have many problems such as heavy calculation and low accuracy. In this paper, a new inverse kinematics algorithm based on geometric and analytical methods is proposed and verified by simulation. The D-H model is established according to the robot structure and the structure parameters of the robot are determined. The positive and negative solutions of the robot are analyzed and calculated. Finally, Matlab is used as the simulation tool to verify the correctness and accuracy of the results. It provides a new idea for the improvement, design and application of the robot with more than 6 axises.
Effectiveness evaluation has important guiding significance for the construction, development and operational application of weapon system of systems. Aiming at the structural complexity, task diversity and functional emergence of weapon system of systems, an operator-based effectiveness evaluation framework for weapon system of systems is proposed. The framework conforms to the flexible evaluation modeling concepts such as the rapidly reconfigurable evaluation index system, the comprehensive evaluation method and the reconfigurable evaluation process based on the means of simulation. Then, the key technologies of operator-based effectiveness evaluation for weapon system of systems are analyzed, the operator-based effectiveness evaluation tool is designed and implemented, which lays a solid foundation for the effectiveness evaluation for weapon system of systems under the background of joint operations.
In order to solve the problem of selecting the aiming point of ranged weapons in urban anti-terrorism operation, a kill rate evaluation model applied to urban anti-terrorism operation is established. Firstly, the killing sources of human in the explosion area are classified, and overpressure, thermal radiation and building debris are used as the killing sources. The overpressure, impulse, average mass of debris, heat flux and thermal dose are selected as the evaluation parameters of explosion power, and according to these parameters, the explosion kill model is established. Finally, aiming at the explosion scene of a British film, the kill rate of civilians and terrorists in the targeted area is evaluated by using the model based on Matlab. The simulation results show that the kill rate of the missile to the person is related to the distance between the aiming point and the persons. The kill rate evaluation model can reflect the kill probability of the aiming point of ranged weapons in the explosion area. It is of great significance to reduce the damage of missile to people in urban anti-terrorism operation and to increase the damage of missile to terrorists.
Result is the most direct and intuitive reflection of training and competition effect. As a matter of fact, scores are more important than single numbers, but a data set composed of many indicators and factors. It is necessary to adopt scientific methods and tools to analyze the data, so as to grasp and apply it more scientifically and effectively. In this paper, the results of a battlefield rescue combat course are taken as samples, and the SPSS is used to describe the results from multiple dimensions. It is important to analyze and verify the influencing factors of the results, so as to apply this research method to the other military activities.
Aiming at the problems of more subjective factors and less quantitative evaluation in the evaluation of combat capability of troops, on the basis of establishing the index system for evaluating the combat capability of armored units, the importance of indicators of command and control ability, mobile impact ability, target damage ability, comprehensive protection ability and rear-loading support ability is screened. Then the weights of index system are determined by combining subjective and objective weights with AHP method and entropy weight method. Finally, weighted TOPSIS comprehensive evaluation method is used to evaluate and rank the combat capability of armored units. This method can provide quantitative reference for improving the combat capability of armored units, and the adaptability and effectiveness of the method are verified by an example.
In view of the difficulty in quantifying the indicators and the difficulty of empowerment in the process of evaluating the equipment maintenance support capability of synthetic brigades after the adjustment of the scale structure reform of the Army, an evaluation index system is established and the data of the “equipment cloud” platform is combined,an analysis and calculation method for specific indicators are proposed. Then the entropy weight method and the analytic hierarchy process are used to calculate the subjective weight and objective weight of the index respectively, and the combined weights are calculated. Finally,the membership function of the index is analyzed in combination with the actual situation. The fuzzy comprehensive evaluation method is used to evaluate the maintenance support capability of the synthetic brigade. Through the example,the equipment maintenance support ability of a synthetic brigade is obtained scientifically and reasonably.
In the current field test, the full-trajectory live ammunition test has the problem of high cost and limited sample size, so most of the field test adopt the method of fixed point radar seeker to conduct the test and produce a large number of test data. How to effectively utilize these existing data and mine the useful information has become the focus of current research. Starting from the data mining application of passive jamming system for warship platform, this paper proposes to reconstruct the scene of field test with the data of field test set up at a fixed point, verify the simulation model, and extrapolate to the dynamic confrontation process of loaded trajectory, so as to form the dynamic inference system of passive jamming test of ship platform for the evaluation of combat effectiveness. The deduction results show that by extracting useful information from the field test data and combining it with operational simulation deduction method, the purpose of deep mining the test data for operational effectiveness evaluation can be achieved.
Aiming at the problem that commanders can't accurately grasp the power range of sounding radar under the conditions of terrain shelter and electromagnetic interference, a simulation system of sounding radar power is designed. The function of the system is analyzed, and the overall framework and composition of the system are introduced, through the specific design of the core modules such as power calculation, terrain shelter, interference calculation and simulation display, the visual display of the power range of the sounding radar under the influence of various factors is realized. It provides an efficient assistant decision-making means for commanders to use equipment in battle.
The simulation support platform which can effectively support the interoperation of the range test and training resources is a key technical issue to promote the rapid construction of the joint test and training system. Based on the analysis of the support requirements of joint test&training simulation and the current status of LVC simulation technology, the technical framework of JoSim, a joint test and training simulation support platform, is established. A standardized object modeling and automatic transformation method is proposed. A common middleware which is used in the fields of modeling simulation and test training is designed. The fast access of heterogeneous systems is realized by gateway and adapter. The practical application of JoSim in the logic range of surface warship air defense capability test and training shows that JoSim can support the rapid construction of logic range for different test and training purposes, and effectively promote the interoperability, reusability and composability of various simulation resources.
In order to find out the boundary conditions for the study of jamming performance of suspended projectile-carried communication jammer, the boundary conditions for studying the jamming performance of projectile-carried communication jammer are discussed and set up from eight aspects; the working frequency band of jammer, the signal propagation mode, transmitting power and communication distance, hopping speed and frequency interval, modulation and demodulation mode of jammed frequency hopping radio, jamming type, jamming efficiency and working environment of the jammer. Then, based on OPNET simulation platform, the model of projectile-carried communication jamming system under specific working environment has been constructed for verification. Relevant conclusions have certain reference value for studying the jamming performance of suspended projectile-carried communication jammer.
Aiming at the underwater requirements of a long time of high precision navigation and positioning, it designs the high precision underwater navigation positioning system framework based on beidou navigation, and system composition, positioning principle are analyzed. In view of the influence of the buoy in posture and acoustic gradient change on positional accuracy, the positioning error correction technology is studied. In addition, considering the requirement of the high stability of the beidou buoy, buoy stability design technology is studied. Through the simulation and water test, the system performance is fully verified and optimized. It shows that underwater beidou achieves underwater seamless extension of the high precision ability of beidou navigation. It can achieve track the underwater target, as well as navigate for the underwater target. It has great social significance and application prospect.
In order to analyze the new ideas and new practices of U.S. Armed Forces in the joint operation planning. By summarizing the literatures on the adaptive reform in recent years, the origin of the reform is analyzed, the main reform measures in six aspects including strategic guidance, in-process review, inter-organizational planning and coordination, planning products, operational assessment, and application of high-tech are sorted out. The current APEX enterprise of the U.S. Armed Forces' joint operational planning is introduced. Based on this, from the planning process, coordination mode and high-tech supported, three recommendations are made.
Based on the unmanned system roadmaps and relevant reports of autonomous underwater vehicle (AUV) released by U.S. Navy recent years, this paper analyzes and summarizes the current situation of U.S. military AUVs, including the classification methods and typical AUVs. The key technologies for improving AUV performance, such as artificial intelligence (AI) and underwater communication, are explained in the paper. At last, according to the changes of AUV task requirements, analyze and outlook its possible development trends. This paper can be used for reference by researchers in the related fields.
