Theory, Methodology, Tools and Applications for Modeling and Simulation of Complex Systems

In practical applications, noncircular turning tool feed system repeat the same control tasks over a finite time interval. But it does not have the ability to improve the tool position error from past repeated operations. This paper will use Partial Form Dynamic Linearization based Model-Free Adaptive Iterative Learning Control (PFDL-MFAILC) algorithm in noncircular turning tool feed. PFDL-MFAILC is a data-driven iterative learning control algorithms. The design of noncircular turning tool feed controller is just rely on input and output data. Simulation of PFDL-MFAILC algorithm show that the noncircular turning tool feed position error is improved as the number of repetitions increases. By contrast with PID control algorithm, position tracking accuracy of PFDL-MFAILC algorithm is significantly better than traditional PID control algorithm. After 60 iterations, the steady-state error of PFDL-MFAILC algorithm is much lower than the steady-state error of the PID algorithm. PFDL-MFAILC algorithm achieve the goal that improving position precision of noncircular turning tool feed and making noncircular turning tool feed have self-learning ability from past repeated operations.

The steady and rapid action of contact directly determines the overall performance of high voltage circuit breaker robot in closing and opening process. The current high-voltage circuit breaker robot reveals obvious contact vibration during use. In order to alleviate the ablation of contact due to the electric arc which is caused by the contact vibration, this essay establishes a forced vibration model of damped three degrees of freedom system through citing a new high-voltage circuit breaker, and analyzes the impacts of the mass and equivalent stiffness proportion among static contact, moving contact and support frame on the contact vibration frequency and amplitude by adopting Dunkerly and matrix iteration methods. Through adding vibration measuring instrument in the contact for experimental verification, the results show that appropriate and reasonable optimization of the mass proportion and equivalent stiffness proportion among static contact, moving contact and support frame can effectively reduce the amplitude of the first order of the main vibration mode, thus significantly alleviating the ablation of contact under the premise of effectively ensuring the opening speed to meet the conditions. The study results can remarkably reduce the vibration of high voltage circuit breaker contact, and then improve the overall performance of high voltage circuit breaker robot.

A inspection robot with four rocker-arms was designed adapting to the environment of fully-mechanized workface in thin coal seam. The walking mechanism and relevant parameters were determined according to the function requirements of the robot. Two sets of walking gaits were planned specifically for the four rocker-arms and were simulated virtually using ADAMS. By observing the behaviors of rocker-arms and analyzing the centroid displacement of the robot during the movements, both of the two sets of walking gaits were verified to be reasonable, one of them is recommended based on its better stability.

Self-balancing robot based on machine vision with two wheels sometimes can be more flexible and saving space than four wheels. We present design scheme and experiment implement of self-balancing robot based on machine vision in the paper. The robot can get road information by CCD camera, then process the image by algorithm to get the road path information and the robot can safely and reliably auto-drive along the road. Based on machine vision the self-balancing robot does not need person to control it, and it can reduce the human disturbance which can greatly improve the efficiency and safety of the system. The system needs to deal with lots of tasks under CPU resource limitation. We have considered the task scheduling under CPU resource limitation through experiment and theory prove. It is proved that the system has good intelligence and anti-interference ability by experiments and competition.

As the concept of more electric aircraft is mentioned, electrical power on aircraft can gradually replace hydraulic energy, pneumatic energy and other secondary energy, in order to achieve the purpose of saving energy. This paper is illustrated by the example of electromechanical actuator, analyzes the energy consumption of equipment and power loss of energy transmission. Appling in multi-domain modeling, it compares power loss with the traditional hydraulic actuator. As well as, it analyzes power loss of the secondary energy transformation device system including generator and hydraulic pump. The results of simulation show that aircraft actuator with electrical energy optimize energy utilization, especially for large passenger aircraft.

In this paper, firstly, it is discrete the engine output and constructed the Markov random process with the discrete state and continuous time. Secondly, we constructed a model of the multi-state Markov model and determined the transition intensity. Thirdly, we computed the expected of the engine capacity deficiency and the forced outage by the method.

Customized equipment manufacturing enterprises use cellular manufacturing system, material handling system transfer material between the units. It is difficult to obtain this type of manufacturing system accurate solution of optimization problem by traditional mathematics methods. This paper establish simulation model about process-assemble and material handling system based on computer simulation. Because the model has much variable and coupling influence, this paper uses sensitivity analysis method based on orthogonal experiments, determines the parameters optimization direction and the influence discipline between parameters, proposes a heuristic optimization method based on the optimization goal and practical constraints. Analysis shows that the proposed optimization method, which can be effective and reasonable configuration enterprise resources.

This paper proposes a new hybrid PSO optimization algorithm, which fuses GA and simulated annealing (SA) into the PSO algorithm. The crossover and mutation mechanism of GA algorithm make the new hybrid algorithm keep the diversity of population and retain the good factors in the population to jump out of local optimum. The sudden jump probability of SA also guarantees the diversity of the population, thus preventing local minimum of the hybrid PSO algorithm. This new hybrid algorithm is used to minimize the maximum completion time of the scheduling problems. The simulation results show that the performance of hybrid optimization algorithm outperforms another hybrid PSO algorithm. The hybrid PSO algorithm is not only in the structure of the algorithm, but also the search mechanism provides a powerful way to solve JSSP.

To solve the flexible job shop scheduling problem (FJSP), a chaotic differential evolution algorithm (CDEA) is proposed with makespan minimization criterion. In the CDEA, logistic mapping is used to generate chaotic numbers for the initialization, because it is helpful to diversify the CDEA population and improve its performance in preventing premature convergence to local minima. A compositive operation-machine-based (COMB) encoding method is employed to reduce computational burden. Meanwhile, a self-adaptive double mutation scheme and an elitist strategy in the selection operator are introduced to balance the population diversity and the convergence rate. The performance of schedules is evaluated in terms of makespan and relative error. The results are compared with different well-known algorithm from open literatures. The results indicate that the proposed CDEA is effective in reducing makespan because the small value of relative error and the faster convergence rate are observed.

Taking a super large twin-propeller twin-rudder container ship as research object, according to mechanism of MMG modeling, a mathematical model of ship motion for a super large ship with twin-propeller and twin-rudder is established. The validity of the ship model is verified by completing turning test. The maneuverability of the ship is discussed under several working conditions when left propeller and right propeller have the same speed and different speed. The active disturbance rejection control algorithm is adopted in the design of ship course controller and simulation calculation is accomplished. Fuzzy control algorithm is used to improve conventional active disturbance rejection controller (ADRC), which is applied to above ship course control. The simulation results show that the improved ADRC has better control performance, fast response and better control accuracy obtained.

Research on aircraft takeoff taxiing model has a great significance upon evaluation of aircraft performance, design of airport runway and warfare simulation. Because the aircraft takeoff taxiing is a complex nonlinear process, existing numerical calculation models are ordinarily based on the energy method or assumption of moment equilibrium, which leads to complex modeling process and low precision. In this paper, we establish a takeoff taxiing model by numerical integration method which is easy to implement, and obtain the parameters of running distance calculation formula based on Lagrange interpolation algorithm [1]. At the end of this paper, we use the characteristic parameters of a specific aircraft as model inputs to validate the correctness and accuracy of this model by comparing the model output and actual flight data.

Precision geometry alignment with rigorous deviation is one of the key assembly issues to be designed in product development. Propagation of tolerances causes large misalignment in assemblies leading to a cast-off and undesirable product functionality. There is always an imperativeness to calculate the assembly characteristic with tighter tolerances for functionally critical dimensions so that geometry imperfections can be compensated. Tolerance analysis and simulation techniques have been utilized by designer supporting such problem solving process. In this paper, a precise geometrical alignment of assembly design has been performed with the help of tolerance simulation. Both worst case and statistical tolerance analysis approaches are applied to provide design alternatives. In this research, a two-component thin walled alignment assembly structure has been chosen to present the design by tolerance simulation procedures. Tolerance ranges have been calculated and recommendations have been provided comparing the worst case and statistical analysis results.

This paper presents a method to predict the remaining useful life of bearings based on theories of Mixture of Gaussians Bayesian Belief Network (MoG-BBN) and Support Vector Data Description (SVDD). Our method extracts feature vectors from raw sensor data using wavelet packet decomposition (WPD). The features are then used to train the corresponding MoG-BBN and SVDD model. Genetic algorithm is employed to determine the initial value of training algorithm and enhance the stability of our model. The two models are combined to acquire a good generalization ability. The effectiveness of the proposed method is verified by actual bearing datasets from the NASA prognostic data repository.

To increase the credibility of decision-making model for close-in air-combat, a type-2 fuzzy logic system (FLS) based approach is introduced into its production rule base, which traditionally neglects the uncertainty of human cognition. By defining its interval type-2 Gaussian membership function for its inputs and outputs, and renovating the traditional production rules into type-2 fuzzy rules, the type-2 FLS model is quickly built. The model employs interval type-2 fuzzy sets to simplify the fuzzy operation and type-reduction while keeping the uncertainty of experiential knowledge. Simulation results show the feasibility and credibility of this approach.

The construction of joint operational system had become the important content of military modernization and preparations for military struggle. Comprehensive and accurate model of Operational System of Systems (OSoS) had become the key fundamental work. The most OSoS models which were constructed in single view had a weak integrated degree and could not accurately describe the system capacity. According to this problem, the framework and method of Multi-dimension and Multi-View Integrated OSoS Modeling was put forward. The OSoS was described in terms of lifecycle, generality, multi-view and multi-dimension. These models were integrated effectively. The emergence, dynamics and openness of OSoS were well reflected.

The performance of Infrared System is mainly determined by targets’ infrared radiation characteristic, ambient complex environment’s infrared radiation characteristic and the infrared radiation contrast between them. Meanwhile these factors are also influenced by geography location, weather condition, season and the time of a day. The IR imaging simulation system has the advantage of convenience, safety, lower cost etc. Using computer to simulate complex infrared scene has been more and more popular. In this paper, a new external rendering algorithm is proposed. This paper focuses on the designment and implementation of the external rendering of the complex scene for the IR imaging simulation. Finally, the effectiveness of the proposed algorithm is validated through an example.

In order to optimize the survivability of shipboard power system, a bi-level programming model was proposed. The master programming model had the topological structure of shipboard power network and electric nodes layouts as decision variables and optimal power system survivability as the objective function. The slave programming model was to minimize the total length of power cables in order to optimize the layouts of them. The standard genetic algorithm was improved to be the solving method of shipboard power network planning. Two genetic operators, which are repairing operator and equivalence operator respectively, are proposed and added to genetic algorithm. The repairing operator was proposed to deal with the infeasible solutions such as isolated nodes, isolated islands and loops, and the equivalence operator was proposed to optimize the scale of electric nodes. A computational example verified the validity and practicality of the improved genetic algorithm and the programming model.

This paper first analyzes the principle of four-point source decoying system to confront passive radar seeker (PRS) of anti-radiation missile (ARM). Then by establishing the decoying system model, the amplitude and phase of the compound electric field of various radiation sources in the PRS dynamic process are calculated and simulated, the simulation conclusion is analyzed. Finally the coordinates of the intersection point between the normal line of the synthetic wavefront with the ground are derived, and a quantitative simulation based on the system model is given. The simulation results show that the four-point source decoying system can not only protect the radar target, but also effectively avoid the bait attack.

From local war fares in recent times, miniature guided munition is playing an increasingly important role in modern wars. This paper proposed a miniature guided rocket projectile which is portable by a soldier. In order to study the influence on the aerodynamic characteristics of the projectile effected by its tails, this article carried on a numerical simulation on the projectile under different kinds of tails, knife edge rear tails and large area flat rear tails, and the lift coefficients, drag coefficients and pitching moment coefficients of the projectile aerodynamic shape are obtained as well as flow field characteristics and the surface pressure distribution when the angle of attack changes. The results showed that, compared with the knife edge rear tails, the flat rear tails have effective enough windward area and wing surface shape which has a good resistance effect. The projectile has a good static stability and good aerodynamic characteristics with enough high lift within a small scope of angle of attack. This research will perform as the theoretical basis for the choice of miniature guided rocket tails and the analysis of its flying stability.

Human body customization is an important task in computer graphics and computer simulation recently. The existing human body modeling methods may generate distorted model or generate a human body model that does not match the target measurements. These shortcomings are improved in this paper. This human body modeling module includes two parts: analyzing the linear relationship between human body shapes and measurements, and human shape modeling according to the target measurements. Instead of analyzing the linear relationship between measurements and the whole human body shapes, we segment each model into 16 rigid parts, and learn the mapping between 3D body shapes of each part and their measurements. By using the above method, we can obtain a new human body model with the control of target measurements. The reconstructed human models are used as infantry battle formation in the damage simulations.

With the development of small UAV, the ground station system is gradually improving. In order to realize the large field of view and wide angle detection of multi targets in the reconnaissance area, the image stitching module is integrated into the terminal of the ground station. Therefore, studying a real-time and efficient image matching algorithm is particularly important. This paper puts forward an improved algorithm based on SIFT (Scale Invariant Feature Transform) feature matching, which joins the UAV aerial parameters to estimate the overlapped region of the aerial images to reduce the sampling time, and adopts gauss quadratic D2OG feature detection operator instead of DOG operator to simplify the structure of the pyramid. Finally, we use the RANSAC algorithm to complete the precise image matching. Through the simulation results we can see that the proposed algorithm can reduce the complexity of data processing, simplify the area of feature matching, and meet the needs of the system on time.

Programming terminal high-low collaborative intercepting strategy scientifically and constructing assistant decision-making model with self-determination and intellectualization is one key problem to enhance operational efficiency. Assistant decision-making model has been constructed after analysis on collaborative intercepting principle; then Improved Clonal Selection Algorithm Optimizing Neural Network (ICLONALG-NN) is designed to solve the terminal anti-missile collaborative intercepting assistant decision-making model through introducing crossover operator to increase population diversity, introducing modified combination operator to make use of information before crossover and mutation, introducing population update operator into traditional CLONALG to optimize Neural Network parameters. Experimental simulation confirms the superiority and practicability of assistant decision-making model solved by ICLONALG-NN.

In order to isolate interference and, keep the line of sight of imaging load in inertial coordinate system, applying an inertially stabilized platform as a joint between aircraft and load is best choice. But it is difficult to design the controller of inertially stabilized platform for airborne remote sensing, because of its large mass and volume, which results low structural resonance interference. Based on the complex frequency domain analysis of various disturbances, a novel optimal controller design method is put forward. Firstly, the dual loop control structure is obtained by modeling main interference and ignoring the inessential disturbance. Secondly, the configuration of controller is determined considering the complexity and tracking performance. Moreover, constraints conditions are designed after analyzing the nyquist stability criterion and the influence of structural resonance. Finally, objective function about the evaluation index of anti-interference ability is proposed and, the controller optimal design is completed by using genetic algorithm. Simulation experiment is implemented based on a realistic prototype system and, the optimal controller design method is proved simple, effective and practical for engineering using.

Reconnaissance and Strike Integrated Unmanned Aerial Vehicles (briefed as R/S UAVs) have operational capabilities of performing many kinds of missions. Aiming at its cooperative task assignation problem, Binary Wolf Pack Algorithm (briefed as BWPA) is proposed to solve the problem. Firstly, model of task assignation is built on the degree of threaten and the degree of hard to attack that time-sensitive targets have. Secondly, basing on Wolf Pack Algorithm (briefed as WPA) and taking binary coding as entry point, BWPA is introduced, which abstracts a productive rule for leading wolf, a renewable mechanism for whole wolf pack, and three intelligent behaviors of scouting, summoning and beleaguering. Lastly, through flexible dealing with task list of time-sensitive targets and coding of R/S UAVs, BWPA is adopted to solve the problem of task assignation. Simulation results show that proposed model is verified, and BWPA has astringency and global optimization capability, is effective to solve the problem of cooperative task assignation of UAVs.

A point cloud data filtering method of laser radar imaging based on classification of curvature is investigated to resolve the deficiencies of the massive 3D point cloud data model filtering using single method. This method divides the region of point cloud data model by the average Gaussian curvature value in neighbor of the sampling point, and then the adaptive median filtering and adaptive bilateral filtering method are used for different region types. Static and dynamic targets are adopted respectively in simulation experiments, experiments show the method can effectively remove the noise of targets under the different motion states, it can also keep details of point cloud data models, and this method has better filtering performance compared with the single filtering method.

The database architecture is classified as database, database management system and users. The database management system is composed of several core modules named as internal mode, external mode, and mapping relations. Database design is discussed as conceptual model design and physical model design respectively. Important technologies such as database connection pool and ORM mapping framework database in database access layer are discussed specially. At last, design validations are presented.

Cooperative target-searching of multiple Unmanned Aerial Vehicles (UAVs) in uncertainty environment is an important research area in multi-UAVs cooperative control. The objective of multi-UAVs searching is to obtain the information of the searched area, decrease the uncertainty of this area and find the hidden targets as fast as possible. This paper introduces a new framework for UAV search operations and proposes a new approach to solve multi-UAVs cooperative searching problem. Aimed at the characteristics of the multi-UAVs cooperative searching problem, the modified probability map based cooperative searching strategy was discussed in detail. Based on the existing algorithms, the cooperative strategy was divided into three key parts, which are probability map initialization, probability map updating and the rules of UAV transfer. The search effectiveness of the Multi-UAVs system in the condition of Multi-target was analyzed based on the method of ABMS (Agent Based Modeling and Simulation). Simulation results demonstrated the effectiveness of the algorithm.

Aiming at the air-combat simulation training, refer to tactics that pilot use in actual air-combat, integrated air-combat knowledge and rules from BVR(Beyond-Visual-Range) to WVR(Within-Visual-Range) were established according to air situation, threat environment and airborne weapon performance. An expert system for target aircraft air-combat tactics decision was designed and developed by using Maneuver Sequence Automation (MSA) method based on basic flight maneuvers, an autonomous air-combat tactical simulation system was also established by using Visual C++ and Open Scene Graph (OSG). This study provided a high fidelity virtual opponent for simulation combat training, early-warning and detection, target recognition, information fusion etc.

A novel matching-area suitability assessment method for geomagnetic matching aided navigation is proposed, which is based on spectral moment characteristics. With the analysis of profile spectral moment and the surface spectral moment, parameters of both homogeneity and isotropy are employed to analyze the effect of direction on geomagnetic three-dimension surface topography. Finally, simulations are made with the computing result of spectral moment parameters. Compared the matching effect of the Inertial Navigation System (INS) indicated path and the real path in the geomagnetic field contour map in different directions, a conclusion is reached that there is a good agreement between flight path and directions. Simulation results prove the proposed assessment method is effective.

In the “man-aircraft” air combat simulation, the combat training of the intelligent Rival-Air-Plane may advance the pilots’ tactics. The intelligence is represented in the two aspects: the intelligent attack and the intelligent evade for the missile threats. This paper aims to use the shortest time when the intelligent aircraft arrives the nearest boundary of the weapon attach zone as the constraint condition, and further to plan out the optimal evasive route and enhance the evasive intelligence. As well, a set of sample data validates the simulation experiment, which is the scientific and feasible approach. Obviously, it not only achieves requirements of the strong real-time combat simulation, but also provides a reference of the actual combat missile to evade for the pilots.

With the rapid development of computing science and the new military revolution, the concept of the battlefield environment has experienced a developing process from the “battlefield” to the “operation space” and to the “operation environment”. Nowadays, the operation simulation environment is still mainly based on one or some certain aspect, and the construction haven’t satisfied the need of operation training enough. This paper analyzed the process of operation simulation environment construction covering the aspects of geography, electromagnetic, information and meteorological, and further discussed the construction process of the operation environment in systematically. What’s more, in view of the operation simulation environment elements, a comprehensive evaluation method was proposed to evaluate the operation simulation environment.

The characteristics of traditional methods for complex and large system development were analyzed, in view of the deficiencies, a kind of engineering development method based on simulation prototype was put forward, and the implementation content, steps and keys during the whole product development process were described. Some key techniques such as the construction of simulation framework, complex object modeling, virtual experiment and simulation evaluation were analyzed, and the solutions are put forward. Practice has proved that the method deepens and optimizes the design of product significantly, excavates potential demands effectively, reduces the development cycle and development cost.

Based on the wind profile radar data during stratiform precipitation processes in Beijing, a model of melting layer which describes the spectral parameter features has been presented. This model is applied to the simulation of WPR FFT data using Gaussian distribution. The simulated WPR FFT data proved to be effective for presenting the echo signal of precipitation particles compared with real data. The application in WPR terminal software test indicates the practicability of this melting layer model in software improvement.

The guided ammunition is a kind of ammunition that has the precise guidance system and high hitting probability. It has the advantages of low price and high precision. It is more and more widely applied in the combat. There are some differences of the working principle of the guidance and control system of the guided ammunition and the missile, so we need to model the guidance and control system of the guided ammunition individually. We built the guidance model and control model for the guided ammunition. Then we verified our model by running the digital simulation program. The simulation results prove the accuracy and effectiveness of our model. The model built in the paper can be used to model the guidance and control system for many other guided ammunitions. They can also be used in the computerized warfare simulation that the guided ammunition involved in.

To finally guarantee the stability and real-time data transmission of network-induced distributed systems, the study of the maximum allowable advancing step of a constructive flight simulation is carried out. Firstly, time performance indicators of the flight dynamics model and the closed control system model are obtained by analyzing their dynamic characteristics in time domain, respectively. Then, the advancing step is calculated based on theory analysis. Finally, several groups of comparative experiments on standalone and distributed systems are conducted to demonstrate the control effects under different advancing steps. To apply the conclusions of this paper to distributed simulation systems can help to solve the unstable problem of distributed models and enhance the distributed simulation performance.

Appropriate missile selection for hostile aerial targets is conceived as an important issue in military operations research. However, both roughness and fuzziness may exist in military data simultaneously due to the complexity of situation and subjectivity of human knowledge. Therefore, this paper presents a two-stage decision model based on rough set theory (RST) and fuzzy inference system (FIS) for missile selection. The LEM2 algorithm in RST is applied to derive decision rules. Next, a Mamdani fuzzy inference system is formed to identify the proper missile type depending on the Gaussian membership functions. Some experiments with respect to some practical parameters are performed to validate the proposed model. The computational results indicate that the proposed model is capable of producing high-quality solutions and is convenient to be incorporated in a military decision support software.

For airport runway, the specific battlefield target, the damage model is established. Based on damage data, this paper also proposes a method for airport runway damage assessment based on grid network scanning. It includes looking for the smallest aircraft landing window in the runway area, making up for the issue of identifying oblique landing window of the traditional algorithms, deciding the damage level of the airport runway, and assessing the function damage of airport runway. By computer simulation, the block situation of the airport caused by two different types of ammunition is analyzed. The paper makes airport runway damage assessment using different minimum landing window searching algorithm, and studies the significance of the oblique landing window on the assessment.

In the paper, a new learning method is presented based on rough set (RS) and extreme learning machine (ELM) for military simulation system application. ELM is a recently proposed algorithm, which can random choose the parameters of hidden neurons and analytically determines the output weights of single-hidden layer feedforward neural networks (SLFNs). Multivariate discretization method is implemented to convert continuous military simulation data into discrete data firstly. RS is then employed to generate simple rules and to remove irrelevant attributes. Finally, ELM is used to evaluate the performance of the reduced data set. The experimental results demonstrate that with the help of RS strategy, our method can produce good enough generalization performance.

In this paper, we present a function network modeling technique and two capability evaluation indicators for combat system-of-systems (CSoS) based on OODA loop. The function network mainly describes the weapon entities and the function interaction between them, such as can-be-observed, can-orient, can-be-decided and can-act. After discussing the CSoS function network model and its generation algorithm, two indicators are proposed to evaluate CSoS capability. A case study on air strike CSoS modeling and simulation analysis is presented to validate proposed method and the indicators. Results show that the number of OODA loops is positively relevant to the CSoS capability, and entropy $$ TE $$ reflects the distribution of OODA loops covering enemy targets.

Interoperation in warfare simulation actually represents that between the military field and the simulation field. The key problem in warfare simulation is that how military users can interact with simulation systems “naturally”, namely how to realize the semantic interoperation. An ontology is a formal, explicit specification of a shared conceptualization, which can be used to resolve heterogeneous problems in information systems. First, three ontology-based interoperation approaches was evaluated and compared. Then, a new approach-Ontology based Online Semantic Interoperation was proposed. At last, an example was given to illustrate the approach.

To meet the military requirement of reconfiguration for optimization of combat SoS, the main purpose of this research is to propose an efficiency evaluation model of combat SoS counterwork based on the directed and weighted network. Firstly, combat SoS is proceed by a network description, according to the nature, state and interaction mechanism of the combat units, to judge whether it can form a link between two nodes, and ascertain the link weight. Secondly, set up the network evaluation index of combat SoS, and establish an efficiency evaluation model of combat SoS counterwork. Lastly, design an experiment for the test of model. The results show that the model could efficiently discriminate and estimate the change of effectiveness in Combat SoS counterworks which caused by the microscopic state change of combat units.

In order to make an improved algorithm of good environmental adaptability and practicability, the simulation of three terrain matching algorithms were done on the condition of typical underwater navigation. Firstly, the differences of terrain aided navigation between on land and underwater were researched. Secondly, the underwater terrain aided navigation models were established as well as the underwater terrain aided navigation simulation of TERCOM, SITAN and ICCP. Finally, the influence of INS errors, measurement accuracy, and digital map resolution to algorithms performance were studied.

This paper proposes an integrated simulation system for air-to-ground guided munitions. Functionality and operating modes of the system are discussed. The structure design along with detailed analysis is given to illustrate its support for the whole process of simulation experiments. The simulation system comprised five subsystems, including experiment management system, data management system, mathematical simulation system, visual generating system and hardware-in-the-loop system. We propose a case study to illustrate the system’s feasibility and advantages in terms of full flow of operating process simulation in real battlefield environment.

In order to improve the localization precision of airborne electronic reconnaissance equipment, Missile-satellite cooperative location method is proposed. First of all, the principle of this method is analyzed. Then the structure and process of location system is explained. At last, the feasibility of the system is proved and simulation is implemented in STK.

With the development of military intelligence training system, behavior modeling becomes the core technology in Computer Generated Forces (CGF). The Context-based Reasoning (CxBR) Framework facilitates building behavior model of Computer Generated Forces because of its visualization, maintainability and non-predetermination. However, it is limited in its ability to support models that include affective reasoning. In order to imitate human behavior more realistically, human affect comes to be considered. By adding two abstract classes to the original framework which are analogous to the Mission and Context classes, representing Mood and Emotion, improved CxBR framework can remedy this limitation and consider the affect. This paper validates the feasibility of the improved CxBR framework by building air to ship fighter system structure using conversion logic of production representation.

Traditional behavior modeling methods rely on the knowledge representation derived from the induction and abstraction of subject matter experts, leading to the high barrier and long modeling period. To tackle this problem, we focus on a new behavior modeling approach, which extracts behavior knowledge from behavior data using recurrent neural network (RNN). A case study, take-off behavior modeling using long short-term memory (LSTM) network, was carried out in three phases: the data recording phase, the offline model training phase and the online model execution phase. A three-layer neural network was constructed to learn the pattern of take-off manipulations. The resulting take-off behavior model performed well to ‘pilot’ an airplane in the real-time test.

MMW/IR compound target simulation system can provide realistic millimeter and infrared signals in the occasion of performance verification for MMW/IR compound guidance missile. A reflector antenna arrangement with MMW/IR bispectral window is proposed for the MMW/IR compound target. Well, the edge diffraction of reflector which can cause angle measurement error seriously influences the uniformity of electric field. In order to enhance the accuracy, two reflector edge processing methods were given in this paper. Through the numerical calculation and comparison, the accuracy of angle measurement can be to 0.032° at 35 GHz. The edge processing can well oppress the edge diffraction and improve the uniformity of MMW field. And the edge processing can be used in the engineering of compound target simulation system design and development for HWIL simulation.

Optical Multi-mode Compound Guidance Simulation System is developed for optical compound guidance system HWIL simulation application in lab environment. In order to appraise the HWIL simulation reliability, this paper analyses some influence factors of evaluation objects, and founds an index system of the Optical Multi-mode Compound Guidance Simulation System for credibility evaluation.

An intelligent perceptual model is an essential component for a virtual soldier to simulate the attention mechanism in military games. Attention mechanism plays a critical role in the process of human evolution, enabling human beings to deal with information more efficiently by focusing on preferable objects and, at the same time, filtering the others. In this paper, we built a novel perceptual model for virtual soldiers based on attention theory. This model includes visual model and audial model. What’s more, we designed the attention controller to decide virtual soldier’s focus. The simulation tests show that by employing the proposed model, virtual soldiers gain more reasonable and efficient attention behaviour.

An effective model on airborne radar based on clutter spectrum is proposed in this paper. According to the simulation test, the calculating speed of the proposed model is close to that of the functional level radar model, and the simulation results of the proposed model are almost the same to that of the signal level radar model. The calculated quantity of the model proposed is much less than that of the signal level radar model. The model could be mainly used in simulation system with many airborne radars. In this paper, a simplified antenna model and a clutter spectrum model are proposed. The clutter spectrum is divided into four districts. The calculated quantity is further reduced since just one district’s clutter energy will be computed. The effective model could be applied to some huge simulating systems, especially in multi-radar and multi-target scenarios where calculating speed is a significant factor.

Air defense combat is one of the important parts in the research of surface ships. In the long-range missile intercepting maneuvering targets is a key link of warship air defense systems. Especially, in this paper, we mainly study for air defense intercept combat mission in CGF (Computer Generate Force) system, we study the designing and simulation of the ballistic of ship based-missiles adopt vertical launching mode. Aiming at the maneuver of the target, we do modeling and Simulation of interception trajectory model by using assumed target method. Finally, through the debugging operation with other CGF simulation against node, the feasibility of the model is verified.

The problem of decision-making in autonomous air combat is widely studied in domains like military training, UAV operation, computer game intelligence, etc. In this paper, the Grammar Evolution (GE) approach is applied to derive proper tactics strategy in air combat. GE approach finds solutions in the form of structured programs based on evolution principles, and thus being possible to bridge domain knowledge with generic evolution process to form a general approach for decision-making problems. In our work, the basic GE approach is first tested with a problem-related BNF grammar, which regulates the mapping between the genotype and the programs (phenotype). We firstly test some simple situations to prove that GE can search for a right solution. Next, however, when combat with the enemy with the Min-Max strategy, we find it is inefficient to get robust strategy in the dynamic environment of air combat because the derived strategy often cover only a small portion of the whole decision space. To this problem, we propose some improvement measures in the next work.

Aiming at the requirement of test and evaluation (T&E) on precision of missile target tacking under the guidance of radar, a virtual-real mixed simulation system is built up based on Xsim simulation platform. In the process, it involves some real equipment, such as radar and target. It solves some key technologies, including time-space consistency, data transmission and missile tracking trajectory modeling. In project, the test method of multiple virtual samples computing in a real time system running has been studied.

Constructing the intelligent situation assessment model of entities based on deep neural network (DNN) according to simulations of human cognition and reasoning mode, is significant for improving the ability of quick response and scientific decision-making of entities in the virtual battlefield. Stacked auto-encoder (SAE) is a commonly used deep learning (DL) model, and its feature extraction process is similar to the way of commander when they assess the battlefield situation. Therefore, we construct the intelligent situation assessment model of the simulation entity based on SAE algorithm. Both the initial feature collection and coding method for the model input, and the knowledge encapsulation and pattern parse method for the model output are proposed to construct the datasets. Then, the proposed situation assessment model based on the SAE is trained by the training datasets. Comparing with the models based on multi-layer perceptron (MLP) and logistic regression classifier (LRC), the convergence and accuracy of the proposed SAE is much better than MLP and LRC, indicating that the battlefield simulation assessment model based on SAE is an effective intelligent model.

In this paper, a temporal network model of system of systems (SoS) coordination is constructed and an algorithm of detecting corresponding temporal motif is designed according to the dynamics and evolution of SoS coordination. By the use of war gaming data, four methods including contrary analysis of temporal motif density in the whole process, the evolution analysis of temporal motif’s density, the analysis of correlation between motif density and operational task measure, generation analysis of SoS coordination pattern are used to analyze the characteristics of SoS coordination pattern, which will be a helpful reference for research in dynamics and evolutionary of SoS coordination.

Flying test and simulation test are the main means in weapon test range to estimate the performance of equipment. Flying test has a high credibility while the sample size is too small, which is contrary to simulation test. It is an effective method to use the two data to estimate collectively. In this paper, we put forward a new test data fusion method based on importance sampling mechanism, which treats the flying test distribution as the true distribution, and the simulation test distribution as the proposal distribution. So we give a new frame and strategy on how to fuse the two different kinds of test data. Numerical computations show the feasibility of the improved method.

Battlefield situation intelligent cognition is very important for the warfare, but this kind of technology is very hard for breakthrough. ‘Deep Green’ program which is launched by DARPA of USA to study military intelligent command and operation for modern warfare is unsuccessful. The key problem is that the technology of battlefield situation intelligent cognition is no breakthrough. Later on, USA pays more attention to the AI technology applied in the military filed, especially in the cognition intelligence aspect. The supported money from DARPA is increased apparently year by year. As some new technologies, such as Big Data and Bayesian Deduction, especially Deep Learning technology, are presented, and these new technologies are applied in the wider and wider range of domains, the development of the AI technology takes a big step forward. We also find some researches on the Deep Learning technology is applied in the situation cognition. Therefore, some significant inspirations for battlefield situation cognition in the military domain can be brought to us. In the paper, the main AI military programs launched by DARPA are combed and analyzed. Big Data, Bayesian Deduction and Deep Learning contained by the AI technology are introduced. Especially, the research status of Deep Learning and its applications in the situation cognition are generalized. On the basis of that, the researches on the military battlefield situation cognition based on the AI technology are prospected according to some relative knowledge.

This thesis introduces a case study of utilizing the decentralized partially observableMarkov decision process (DEC-POMDP) in modeling information delivery behavior under the network centric warfare settings. The deployed troops are modeled as bounded rational agents, and they have different communication success possibilities in the long range and the short range. Our DEC-POMDP model indicates how to deliver time-sensitive target information from the monitoring agent to the headquarter through multiple bridging agents. The bridging agents should intelligently decide which information to transmit through the long range or delegate the transfer to other agents via short range communications. Our models shed light on the improved information delivery behavior that are inferred from the DEC-POMDP model with simulation based experiments. We expect that this research would play a role in improving Tactics, Techniques, and Procedures (TTP) in the field operation.

The general structural design and operation process of the existing electromagnetic railgun (EMRG) simulation tools are defective. Aiming to solve the problems, including functional singleness, poor extensibility and inflexibility, the EMRG exterior ballistics simulation and data analysis tool is designed and developed in this paper. Firstly, in order to realize its functions, the requirement analysis of exterior ballistics simulation and data analysis is presented and the overall structural design and operation process of this tool are obtained. Secondly, this comprehensive tool is developed by Microsoft Visual Studio 2010, and based on the modularization theory, the tool realizes exterior ballistics simulation according to the six-DOF model, applications in various data analysis methods and report generation, etc. Finally, the application case shows that the tool can meet the requirement of the simulation and data analysis of the EMRG exterior ballistics and it is easy to operate. Therefore, the tool has significance to the study of the EMRG exterior ballistics.

Modeling combat behavior is an important, yet complicated task because the combat behavior emerges from the rationality as well as the irrationality. For instance, when a soldier confronts a dilemma on accomplishing his mission and saving his life, it is difficult to model his ongoing thoughts with a simple model. This paper presents (1) how to reconstruct a realistic combat environment with a virtual-constructive simulation, and (2) how to model such combat behavior with the inverse reinforcement learning. The virtual-constructive simulation is a well-known simulation application for soldier training. Previous works on this virtual-constructive simulation focuses on a small number of entities and mission phases, so it was difficult to observe the frequent behavior dilemma in the field. This work presents a large scale and a complete brigade-level operation to provide such synthetic environment to human player. Then, our second work is observing the com-bat behavior through the virtual-constructive simulations, and modeling the behavior with the inverse reinforcement learning. Surely, we can observe the descriptive statistics of the observed behavior, but the inverse reinforcement learning provides calibrated weights on the valuation on hypothetical rewards from conflicting goals. Our study is the first attempt on merging the large-scale virtual constructive simulation and the inverse reinforcement learning on such massive scale.

A simple and effective method for 3D skeleton based action recognition is proposed in this paper. Instead of taking the whole skeleton joints as the input, we select several active joints to represent the entire action which motion ranges are relatively large via evaluating their variance and give them different weights. Then by calculating the angles between these joints and the center joint in their three projections produces a feature set at each frame which is applied in a bag-of-words to form the 2D array. The final features are cascaded by these 2D arrays. During this process, the feature numbers can be reduced effectively. The random forest is utilized to classify different actions. Experiments on MSR-Action3D dataset demonstrate that our approach is able to achieve the state-of-the-art performance with high recognition rate and computational efficiency.

With the development of Internet+, developing an application of satellite simulation which executes on mobile terminals has an important significance for engineers to get visual knowledge of spaceflight. The application design and implementation method of satellite simulation software are discussed. A framework of satellite simulation system is presented, the modules and workflow of the satellite simulation system are described. Approaches to create the starry background of the satellite operation, environment lighting of the earth and the orbits of satellites in the software are presented in detail. The software uses Unity3D platform to create 3D simulation environment, uses SGP4/SDP4 orbit prediction models to calculate orbits of satellites, and uses databases to store TLE data.

In this paper, a novel viewpoint scoring method based on multi-attribute fusion is proposed. The perceptual model of viewpoint preference is explored from geometry and visual perception aspects. Modified mesh saliency entropy is presented as the crucial intrinsic geometric attribute. Several digital image factors which have influence on human visual perception form the viewpoint perception attributes. Evolution algorithm is utilized to select the canonical viewpoint automatically and intelligently. Experimental results demonstrate that the canonical viewpoint obtained by the proposed method contains more visible salient features and better conforms to human visual perception characteristic. Moreover, the method has high efficiency and requires no user interaction.

This paper implements the self-collision detection optimization during the clothes simulation, including the high-level and low-level tailoring optimization. During the high-level tailoring stage, this paper firstly implements the basic high-level tailoring in combination with the hierarchical bounding box algorithm and continuous normal vector cone information. On this basis, this paper implements the high-level tailoring optimization based on the radiation angle. Finally, this paper implements the high-level tailoring optimization based on the isolated set. During the low-level tailoring stage, this paper firstly implements the low-level tailoring optimization based on the characteristic distribution. In addition, this paper also implements the low-level tailoring optimization based on the non-coplanar filter. Experiment result shows that the two level tailoring optimization method in this paper can effectively cut off the redundant and non-collision primitive pair and further improve the efficiency of self-collision detection.

This research develops a theoretical method to evaluate the creep damage in a 91 steel with ultrasonic nonlinear technique. A three-dimensional finite element model is built to simulate the creep damage of steel P91 and the ultrasonic wave propagation through the degraded material. The nonlinear parameter is calculated from the amplitude of the second harmonic component of ultrasonic waves which is caused by the micro-structural changes of the material. Both the experimental and theoretical results show that the nonlinear parameter increases with the increase of temperature. The results show the feasibility of the model to evaluate the early stage of the degradation of materials by using ultrasonic nonlinearity.

Simulation scenario is the data base of simulation system. The current simulation scenario can often be applied to single simulation system. There are problems like the low reusability of old simulation scenario, long time when the large number of entities are created and so on. This paper studies the mapping relationship of simulation objects between different specifications, visually saving this mapping relationship to the data dictionary, and through the simulation object instantiation and data query to transform old entities to new entities that simulation system is able to identify. In MAXSIM Simulation System Development, this way greatly reduce the simulation scenario development time and ensures the accuracy of the data.

As a classic method of human motion alignment, original Dynamic Time Warping algorithm may cause poor effects while applied in the field of human motion blending, such as motion pause and motion distortion. To address this problem, we propose a motion segmentation based method of human motion alignment. We first make use of Isomap which is a classic algorithm in manifold learning to reduce dimensionality of all the motion frames in the original motion samples. Then we can obtain the segment of different motion samples by making use of the extreme point of all the corresponding low-dimensional coordinates of motion frames. Finally we apply Dynamic Time Warping method in the corresponding segments of different motion samples respectively to achieve refined alignment results. We conduct contrast experiment between our algorithm and traditional method, the result shows that our method can eliminate motion pause and motion distortion phenomenon caused by traditional method and make significant improvement in the application of human motion blending. The generated motion samples have a high degree of both fidelity and fluency.

The design environment of command compartment is developed based on the physical compartment and virtual scenes, and can be used for design and simulation training for command and control systems. The characteristics of the traditional virtual cabin are analyzed. A cost-effective hardware architecture is designed according to the construction target of the semi physical command cabin. A design method of the design environment of command compartment based on virtual reality is presented. And the key technologies of multi-channel virtual scene generation are analyzed and studied.

Kinect is an instrument that can interpret specific pictures with relatively high precision. In this paper, we utilize proper mathematical methods based on the point cloud data from Kinect 2.0 we tested before to analyze the deviation of Kinect scanning. By using Pearson Correlation Coefficient, regression model and Fuzzy Comprehensive Evaluation Method, we manage to analyze the deviation from both the position deviation and color deviation. Our work suggests the important parameters of Kinect—Elevation and Vertical Dimension. When searching for ways to reduce deviation of the color of the measured object, we should not only be concentrated on the optimized combination of the two parameters but also take the relation between vertical dimension and position accuracy into consideration.

We proposed methods for 3D visualization and inspection of different layers corresponding to different tissues. First, we give pipeline of extraction the MRI human brain datasets. Then we present the design of trapezoid opacity transfer function and present the framework of visualization for the CT and segmented MRI datasets, based on CUDA-based real-time volume ray-casting. Furthermore, we inspect its inner layers via virtual lenses by intuitive bimanual gesture interaction, achieving the focus+context visualization. Finally, the performance of our framework verified effective by our experiments.