This work proposes a recorded recurrent twin delayed deep deterministic(RRTD3)policy gradient algorithm to solve the challenge of constructing guidance laws for intercepting endoatmospheric maneuvering missiles with u...This work proposes a recorded recurrent twin delayed deep deterministic(RRTD3)policy gradient algorithm to solve the challenge of constructing guidance laws for intercepting endoatmospheric maneuvering missiles with uncertainties and observation noise.The attack-defense engagement scenario is modeled as a partially observable Markov decision process(POMDP).Given the benefits of recurrent neural networks(RNNs)in processing sequence information,an RNN layer is incorporated into the agent’s policy network to alleviate the bottleneck of traditional deep reinforcement learning methods while dealing with POMDPs.The measurements from the interceptor’s seeker during each guidance cycle are combined into one sequence as the input to the policy network since the detection frequency of an interceptor is usually higher than its guidance frequency.During training,the hidden states of the RNN layer in the policy network are recorded to overcome the partially observable problem that this RNN layer causes inside the agent.The training curves show that the proposed RRTD3 successfully enhances data efficiency,training speed,and training stability.The test results confirm the advantages of the RRTD3-based guidance laws over some conventional guidance laws.展开更多
In the multi-radar networking system,aiming at the problem of locating long-distance targets synergistically with difficulty and low accuracy,a dual-station joint positioning method based on the target measurement err...In the multi-radar networking system,aiming at the problem of locating long-distance targets synergistically with difficulty and low accuracy,a dual-station joint positioning method based on the target measurement error feature complementarity is proposed.For dual-station joint positioning,by constructing the target positioning error distribution model and using the complementarity of spatial measurement errors of the same long-distance target,the area with high probability of target existence can be obtained.Then,based on the target distance information,the midpoint of the intersection between the target positioning sphere and the positioning tangent plane can be solved to acquire the target's optimal positioning result.The simulation demonstrates that this method greatly improves the positioning accuracy of target in azimuth direction.Compared with the traditional the dynamic weighted fusion(DWF)algorithm and the filter-based dynamic weighted fusion(FBDWF)algorithm,it not only effectively eliminates the influence of systematic error in the azimuth direction,but also has low computational complexity.Furthermore,for the application scenarios of multi-radar collaborative positioning and multi-sensor data compression filtering in centralized information fusion,it is recommended that using radar with higher ranging accuracy and the lengths of baseline between radars are 20–100 km.展开更多
Based on the phenomenon of"split brick by Qigong",a mechanical model for short beam impact is proposed.Combined with the traditional energy method,a theoretical analysis of the impact of the short beam(Timos...Based on the phenomenon of"split brick by Qigong",a mechanical model for short beam impact is proposed.Combined with the traditional energy method,a theoretical analysis of the impact of the short beam(Timoshenko beam)closer to the real situation is made considering the quality and initial deformation.The optimal solution of short beam impact problem of how to choose the position where the short beam is most likely to break is obtained.The finite element numerical analysis and experimental test are used,and the results verify the applicability of the theoretical analysis of the proposed model.展开更多
This paper reports the use of machine learning to enhance the diagnosis of a dusty plasma.Dust in a plasma has a large impact on the properties of the plasma.According to a probe diagnostic experiment on a dust-free p...This paper reports the use of machine learning to enhance the diagnosis of a dusty plasma.Dust in a plasma has a large impact on the properties of the plasma.According to a probe diagnostic experiment on a dust-free plasma combined with machine learning,an experiment on a dusty plasma is designed and carried out.Using a specific experimental device,dusty plasma with a stable and controllable dust particle density is generated.A Langmuir probe is used to measure the electron density and electron temperature under different pressures,discharge currents,and dust particle densities.The diagnostic result is processed through a machine learning algorithm,and the error of the predicted results under different pressures and discharge currents is analyzed,from which the law of the machine learning results changing with the pressure and discharge current is obtained.Finally,the results are compared with theoretical simulations to further analyze the properties of the electron density and temperature of the dusty plasma.展开更多
The appropriate terminal attitude of missile is important to improve the damage effectiveness of fragmentation warhead against aircraft targets.In this paper,three missile terminal attitude selection methods are propo...The appropriate terminal attitude of missile is important to improve the damage effectiveness of fragmentation warhead against aircraft targets.In this paper,three missile terminal attitude selection methods are proposed to solve the problem of terminal attitude selection in different situations according to their respective evaluation indexes.The MVE-based method uses the damage probability of each detonation point around the aircraft at a given attitude of missile to calculate the Mean Volume of Effectiveness(MVE)at the corresponding attitude,and then uses the MVE for different attitudes to select the terminal strike attitude.The detonation position-based method addresses the case where the missile detonation position can be assessed in advance.Given the effects of missile guidance errors and fuze activation position errors,Monte Carlo simulations are used to calculate the damage probability of aircraft with different strike attitudes,from which the terminal strike attitude is selected.The BP-ANN model-based method uses the constructed Back Propagation Artificial Neural Network(BP-ANN)model instead of simulation to calculate the evaluation indexes in the corresponding cases of MVE-based method and Detonation position-based method,which can improve the efficiency of attitude selection.Simulations are conducted for different scenarios to verify the feasibility and effectiveness of the proposed method.展开更多
The path planning problem of complex wild environment with multiple elements still poses challenges.This paper designs an algorithm that integrates global and local planning to apply to the wild environmental path pla...The path planning problem of complex wild environment with multiple elements still poses challenges.This paper designs an algorithm that integrates global and local planning to apply to the wild environmental path planning.The modeling process of wild environment map is designed.Three optimization strategies are designed to improve the A-Star in overcoming the problems of touching the edge of obstacles,redundant nodes and twisting paths.A new weighted cost function is designed to achieve different planning modes.Furthermore,the improved dynamic window approach(DWA)is designed to avoid local optimality and improve time efficiency compared to traditional DWA.For the necessary path re-planning of wild environment,the improved A-Star is integrated with the improved DWA to solve re-planning problem of unknown and moving obstacles in wild environment with multiple elements.The improved fusion algorithm effectively solves problems and consumes less time,and the simulation results verify the effectiveness of improved algorithms above.展开更多
With the rapid development of deep learning-based detection algorithms,deep learning is widely used in the field of infrared small target detection.However,well-designed adversarial samples can fool human visual perce...With the rapid development of deep learning-based detection algorithms,deep learning is widely used in the field of infrared small target detection.However,well-designed adversarial samples can fool human visual perception,directly causing a serious decline in the detection quality of the recognition model.In this paper,an adversarial defense technology for small infrared targets is proposed to improve model robustness.The adversarial samples with strong migration can not only improve the generalization of defense technology,but also save the training cost.Therefore,this study adopts the concept of maximizing multidimensional feature distortion,applying noise to clean samples to serve as subsequent training samples.On this basis,this study proposes an inverse perturbation elimination method based on Generative Adversarial Networks(GAN)to realize the adversarial defense,and design the generator and discriminator for infrared small targets,aiming to make both of them compete with each other to continuously improve the performance of the model,find out the commonalities and differences between the adversarial samples and the original samples.Through experimental verification,our defense algorithm is not only able to cope with multiple attacks but also performs well on different recognition models compared to commonly used defense algorithms,making it a plug-and-play efficient adversarial defense technique.展开更多
The combustion of solid propellants can provide tremendous thrust to drive solid rocket motors.Solid propellant grains are subjected to mechanical loads during practical combustion applications.The causes of mechanica...The combustion of solid propellants can provide tremendous thrust to drive solid rocket motors.Solid propellant grains are subjected to mechanical loads during practical combustion applications.The causes of mechanical loads include,but are not limited to,impulsive loads during ignition and periodic loads(>5 Hz)due to mechanical oscillation[1,2].展开更多
On-orbit spacecraft face many threats,such as collisions with debris or other spacecraft.Therefore,perception of the surrounding space environment is vitally important for on-orbit spacecraft.Spacecraft require a dyna...On-orbit spacecraft face many threats,such as collisions with debris or other spacecraft.Therefore,perception of the surrounding space environment is vitally important for on-orbit spacecraft.Spacecraft require a dynamic attitude tracking ability with high precision for such missions.This paper aims to address the above problem using an improved backstepping controller.The tracking mission is divided into two phases:coarse alignment and fine alignment.In the first phase,a traditional saturation controller is utilized to limit the maximum attitude angular velocity according to the actuator’s ability.For the second phase,the proposed backstepping controller with different virtual control inputs is applied to track the moving target.To fulfill the high precision attitude tracking requirements,a hybrid attitude control actuator consisting of a Control Moment Gyro(CMG)and Reaction Wheel(RW)is constructed,which can simultaneously avoid the CMG singularity and RW saturation through the use of an angular momentum optimal management strategy,such as null motion.Finally,five simulation scenarios were carried out to demonstrate the effectiveness of the proposed control strategy and hybrid actuator.展开更多
Traditional coupled multi-disciplinary design optimization based on computational fluid dynamics/computational structure dynamics(CFD/CSD)aims to optimize the jig shape of aircraft,and general multi-disciplinary desig...Traditional coupled multi-disciplinary design optimization based on computational fluid dynamics/computational structure dynamics(CFD/CSD)aims to optimize the jig shape of aircraft,and general multi-disciplinary design optimization methodology is adopted.No special consideration is given to the aircraft itself during the optimization.The main drawback of these methodologies is the huge expanse and the low efficiency.To solve this problem,we put forward to optimize the cruise shape directly based on the fact that the cruise shape can be transformed into jig shape,and a methodology named reverse iteration of structural model(RISM)is proposed to get the aero-structural performance of cruise shape.The main advantage of RISM is that the efficiency can be improved by at least four times compared with loosely-coupled aeroelastic analysis and it maintains almost the same fidelity of loosely-coupled aeroelastic analysis.An optimization framework based on RISM is proposed.The aerodynamic and structural performances can be optimized simultaneously in this framework,so it may lead to the true optimal solution.The aerodynamic performance was predicted by N-S solver in this paper.Test shows that RISM predicts the aerodynamic and structural performances very well.A wing-body configuration was optimized by the proposed optimization framework.The drag and weight of the aircraft are decreased after optimization,which shows the effectiveness of the proposed framework.展开更多
In order to provide accurate launching pitching angular velocity(LPAV) for the exterior trajectory optimization design, multi-flexible body dynamics(MFBD) technology is presented to study the changing law of LPAV ...In order to provide accurate launching pitching angular velocity(LPAV) for the exterior trajectory optimization design, multi-flexible body dynamics(MFBD) technology is presented to study the changing law of LPAV of the rotating missile based on spiral guideway. An MFBD virtual prototype model of the rotating missile launching system is built using multi-body dynamics modeling technology based on the built flexible body models of key components and the special force model.The built model is verified with the frequency spectrum analysis. With the flexible body contact theory and nonlinear theory of MFBD technology, the research is conducted on the influence of a series of factors on LPAV, such as launching angle change, clearance between launching canister and missile,thrust change, thrust eccentricity and mass eccentricity, etc. Through this research, some useful values of the key design parameters which are difficult to be measured in physical tests are obtained. Finally,a simplified mathematical model of the changing law of LPAV is presented through fitting virtual test results using the linear regression method and verified by physical flight tests. The research results have important significance for the exterior trajectory optimization design.展开更多
In order to realize the agility,collaboration and visualization of alloy material devel-opment process,a product development platform based on simulation and modeling technologies is established in this study.In this ...In order to realize the agility,collaboration and visualization of alloy material devel-opment process,a product development platform based on simulation and modeling technologies is established in this study.In this platform,the whole-process simulation module builds multi-level simulation models based on metallurgical mechanisms from the production line level,the thermo-mechanical coupling field level and the microstructure evolution level.The design knowledge management module represents the multi-source heterogeneous material design knowledge through ontology model,including customers’requirement knowledge,material component knowledge,process design knowledge and quality inspection knowledge,and utilizes the case-based reasoning approach to reuse the knowledge.The data-driven modeling module applies machine learning algorithms to mine the relationships between product mechanical properties,material components,and process parameters from historical samples,and utilizes multi-objective optimiza-tion algorithms to find the optimal combination of process parameters.Application of the developed platform in actual steel mills shows that the proposed method helps to improve the efficiency of product design process.展开更多
基金supported by the National Natural Science Foundation of China(Grant No.12072090)。
文摘This work proposes a recorded recurrent twin delayed deep deterministic(RRTD3)policy gradient algorithm to solve the challenge of constructing guidance laws for intercepting endoatmospheric maneuvering missiles with uncertainties and observation noise.The attack-defense engagement scenario is modeled as a partially observable Markov decision process(POMDP).Given the benefits of recurrent neural networks(RNNs)in processing sequence information,an RNN layer is incorporated into the agent’s policy network to alleviate the bottleneck of traditional deep reinforcement learning methods while dealing with POMDPs.The measurements from the interceptor’s seeker during each guidance cycle are combined into one sequence as the input to the policy network since the detection frequency of an interceptor is usually higher than its guidance frequency.During training,the hidden states of the RNN layer in the policy network are recorded to overcome the partially observable problem that this RNN layer causes inside the agent.The training curves show that the proposed RRTD3 successfully enhances data efficiency,training speed,and training stability.The test results confirm the advantages of the RRTD3-based guidance laws over some conventional guidance laws.
文摘In the multi-radar networking system,aiming at the problem of locating long-distance targets synergistically with difficulty and low accuracy,a dual-station joint positioning method based on the target measurement error feature complementarity is proposed.For dual-station joint positioning,by constructing the target positioning error distribution model and using the complementarity of spatial measurement errors of the same long-distance target,the area with high probability of target existence can be obtained.Then,based on the target distance information,the midpoint of the intersection between the target positioning sphere and the positioning tangent plane can be solved to acquire the target's optimal positioning result.The simulation demonstrates that this method greatly improves the positioning accuracy of target in azimuth direction.Compared with the traditional the dynamic weighted fusion(DWF)algorithm and the filter-based dynamic weighted fusion(FBDWF)algorithm,it not only effectively eliminates the influence of systematic error in the azimuth direction,but also has low computational complexity.Furthermore,for the application scenarios of multi-radar collaborative positioning and multi-sensor data compression filtering in centralized information fusion,it is recommended that using radar with higher ranging accuracy and the lengths of baseline between radars are 20–100 km.
基金supported in part by the National College Students Innovation Training Program
文摘Based on the phenomenon of"split brick by Qigong",a mechanical model for short beam impact is proposed.Combined with the traditional energy method,a theoretical analysis of the impact of the short beam(Timoshenko beam)closer to the real situation is made considering the quality and initial deformation.The optimal solution of short beam impact problem of how to choose the position where the short beam is most likely to break is obtained.The finite element numerical analysis and experimental test are used,and the results verify the applicability of the theoretical analysis of the proposed model.
基金financially supported by National Natural Science Foundation of China(Nos.11775062,11805130 and 11905125)the Shanghai Sailing Program(Nos.19YF1420900 and 18YF1422200)。
文摘This paper reports the use of machine learning to enhance the diagnosis of a dusty plasma.Dust in a plasma has a large impact on the properties of the plasma.According to a probe diagnostic experiment on a dust-free plasma combined with machine learning,an experiment on a dusty plasma is designed and carried out.Using a specific experimental device,dusty plasma with a stable and controllable dust particle density is generated.A Langmuir probe is used to measure the electron density and electron temperature under different pressures,discharge currents,and dust particle densities.The diagnostic result is processed through a machine learning algorithm,and the error of the predicted results under different pressures and discharge currents is analyzed,from which the law of the machine learning results changing with the pressure and discharge current is obtained.Finally,the results are compared with theoretical simulations to further analyze the properties of the electron density and temperature of the dusty plasma.
基金supported by The Fundamental Research Funds for the Central Universities,China。
文摘The appropriate terminal attitude of missile is important to improve the damage effectiveness of fragmentation warhead against aircraft targets.In this paper,three missile terminal attitude selection methods are proposed to solve the problem of terminal attitude selection in different situations according to their respective evaluation indexes.The MVE-based method uses the damage probability of each detonation point around the aircraft at a given attitude of missile to calculate the Mean Volume of Effectiveness(MVE)at the corresponding attitude,and then uses the MVE for different attitudes to select the terminal strike attitude.The detonation position-based method addresses the case where the missile detonation position can be assessed in advance.Given the effects of missile guidance errors and fuze activation position errors,Monte Carlo simulations are used to calculate the damage probability of aircraft with different strike attitudes,from which the terminal strike attitude is selected.The BP-ANN model-based method uses the constructed Back Propagation Artificial Neural Network(BP-ANN)model instead of simulation to calculate the evaluation indexes in the corresponding cases of MVE-based method and Detonation position-based method,which can improve the efficiency of attitude selection.Simulations are conducted for different scenarios to verify the feasibility and effectiveness of the proposed method.
基金Industry-University-Research Cooperation Fund Project of the Eighth Research Institute of China Aerospace Science and Technology Corporation(No.USCAST2022-11)。
文摘The path planning problem of complex wild environment with multiple elements still poses challenges.This paper designs an algorithm that integrates global and local planning to apply to the wild environmental path planning.The modeling process of wild environment map is designed.Three optimization strategies are designed to improve the A-Star in overcoming the problems of touching the edge of obstacles,redundant nodes and twisting paths.A new weighted cost function is designed to achieve different planning modes.Furthermore,the improved dynamic window approach(DWA)is designed to avoid local optimality and improve time efficiency compared to traditional DWA.For the necessary path re-planning of wild environment,the improved A-Star is integrated with the improved DWA to solve re-planning problem of unknown and moving obstacles in wild environment with multiple elements.The improved fusion algorithm effectively solves problems and consumes less time,and the simulation results verify the effectiveness of improved algorithms above.
基金supported in part by the National Natural Science Foundation of China under Grant 62073164the Shanghai Aerospace Science and Technology Innovation Foundation under Grant SAST2022-013.
文摘With the rapid development of deep learning-based detection algorithms,deep learning is widely used in the field of infrared small target detection.However,well-designed adversarial samples can fool human visual perception,directly causing a serious decline in the detection quality of the recognition model.In this paper,an adversarial defense technology for small infrared targets is proposed to improve model robustness.The adversarial samples with strong migration can not only improve the generalization of defense technology,but also save the training cost.Therefore,this study adopts the concept of maximizing multidimensional feature distortion,applying noise to clean samples to serve as subsequent training samples.On this basis,this study proposes an inverse perturbation elimination method based on Generative Adversarial Networks(GAN)to realize the adversarial defense,and design the generator and discriminator for infrared small targets,aiming to make both of them compete with each other to continuously improve the performance of the model,find out the commonalities and differences between the adversarial samples and the original samples.Through experimental verification,our defense algorithm is not only able to cope with multiple attacks but also performs well on different recognition models compared to commonly used defense algorithms,making it a plug-and-play efficient adversarial defense technique.
基金supported by the National Key R&D Program of China(Grant No.2019YFA0405600)。
文摘The combustion of solid propellants can provide tremendous thrust to drive solid rocket motors.Solid propellant grains are subjected to mechanical loads during practical combustion applications.The causes of mechanical loads include,but are not limited to,impulsive loads during ignition and periodic loads(>5 Hz)due to mechanical oscillation[1,2].
基金the support provided by the National Natural Science Foundation of China(No.61973153)the National Key Research and Development Plan of China(No.2016YFB0500901)the Open Fund of the National Defense Key Discipline Laboratory of Micro-Spacecraft Technology of China(No.HIT.KLOF.MST.201705)
文摘On-orbit spacecraft face many threats,such as collisions with debris or other spacecraft.Therefore,perception of the surrounding space environment is vitally important for on-orbit spacecraft.Spacecraft require a dynamic attitude tracking ability with high precision for such missions.This paper aims to address the above problem using an improved backstepping controller.The tracking mission is divided into two phases:coarse alignment and fine alignment.In the first phase,a traditional saturation controller is utilized to limit the maximum attitude angular velocity according to the actuator’s ability.For the second phase,the proposed backstepping controller with different virtual control inputs is applied to track the moving target.To fulfill the high precision attitude tracking requirements,a hybrid attitude control actuator consisting of a Control Moment Gyro(CMG)and Reaction Wheel(RW)is constructed,which can simultaneously avoid the CMG singularity and RW saturation through the use of an angular momentum optimal management strategy,such as null motion.Finally,five simulation scenarios were carried out to demonstrate the effectiveness of the proposed control strategy and hybrid actuator.
基金supported by the National Natural Science Foundation of China(Grant Nos.11272005,10902082 and 91016008)the Funds for the Central Universities(Grant No.xjj2014135)partially supported by the open project of State Key Laboratory for Strength and Vibration of Mechanical Structures of Xi’an Jiaotong University(SV2014-KF-10)
文摘Traditional coupled multi-disciplinary design optimization based on computational fluid dynamics/computational structure dynamics(CFD/CSD)aims to optimize the jig shape of aircraft,and general multi-disciplinary design optimization methodology is adopted.No special consideration is given to the aircraft itself during the optimization.The main drawback of these methodologies is the huge expanse and the low efficiency.To solve this problem,we put forward to optimize the cruise shape directly based on the fact that the cruise shape can be transformed into jig shape,and a methodology named reverse iteration of structural model(RISM)is proposed to get the aero-structural performance of cruise shape.The main advantage of RISM is that the efficiency can be improved by at least four times compared with loosely-coupled aeroelastic analysis and it maintains almost the same fidelity of loosely-coupled aeroelastic analysis.An optimization framework based on RISM is proposed.The aerodynamic and structural performances can be optimized simultaneously in this framework,so it may lead to the true optimal solution.The aerodynamic performance was predicted by N-S solver in this paper.Test shows that RISM predicts the aerodynamic and structural performances very well.A wing-body configuration was optimized by the proposed optimization framework.The drag and weight of the aircraft are decreased after optimization,which shows the effectiveness of the proposed framework.
基金supported by the Key Special Funds for National Defense Basic Scientific Research of China (No. C0320110005)
文摘In order to provide accurate launching pitching angular velocity(LPAV) for the exterior trajectory optimization design, multi-flexible body dynamics(MFBD) technology is presented to study the changing law of LPAV of the rotating missile based on spiral guideway. An MFBD virtual prototype model of the rotating missile launching system is built using multi-body dynamics modeling technology based on the built flexible body models of key components and the special force model.The built model is verified with the frequency spectrum analysis. With the flexible body contact theory and nonlinear theory of MFBD technology, the research is conducted on the influence of a series of factors on LPAV, such as launching angle change, clearance between launching canister and missile,thrust change, thrust eccentricity and mass eccentricity, etc. Through this research, some useful values of the key design parameters which are difficult to be measured in physical tests are obtained. Finally,a simplified mathematical model of the changing law of LPAV is presented through fitting virtual test results using the linear regression method and verified by physical flight tests. The research results have important significance for the exterior trajectory optimization design.
基金This research is supported by the National Key R&D Program of China under the Grant No.2018YFB1701602the National Natural Science Foundation of China under the Grant No.61903031the Fundamental Research Funds for the Cen-tral Universities under the Grant No.FRF-TP-20-050A2.
文摘In order to realize the agility,collaboration and visualization of alloy material devel-opment process,a product development platform based on simulation and modeling technologies is established in this study.In this platform,the whole-process simulation module builds multi-level simulation models based on metallurgical mechanisms from the production line level,the thermo-mechanical coupling field level and the microstructure evolution level.The design knowledge management module represents the multi-source heterogeneous material design knowledge through ontology model,including customers’requirement knowledge,material component knowledge,process design knowledge and quality inspection knowledge,and utilizes the case-based reasoning approach to reuse the knowledge.The data-driven modeling module applies machine learning algorithms to mine the relationships between product mechanical properties,material components,and process parameters from historical samples,and utilizes multi-objective optimiza-tion algorithms to find the optimal combination of process parameters.Application of the developed platform in actual steel mills shows that the proposed method helps to improve the efficiency of product design process.