This paper considers the problem of generating a flight trajectory for a single fixed-wing unmanned combat aerial vehicle （UCAV） performing an air-to-surface multi-target attack （A/SMTA） mission using satellite-gu...This paper considers the problem of generating a flight trajectory for a single fixed-wing unmanned combat aerial vehicle （UCAV） performing an air-to-surface multi-target attack （A/SMTA） mission using satellite-guided bombs. First, this problem is formulated as a variant of the traveling salesman problem （TSP）, called the dynamic-constrained TSP with neighborhoods （DCT- SPN）. Then, a hierarchical hybrid approach, which partitions the planning algorithm into a roadmap planning layer and an optimal control layer, is proposed to solve the DCTSPN. In the roadmap planning layer, a novel algorithm based on an updatable proba- bilistic roadmap （PRM） is presented, which operates by randomly sampling a finite set of vehicle states from continuous state space in order to reduce the complicated trajectory planning problem to planning on a finite directed graph. In the optimal control layer, a collision-free state-to-state trajectory planner based on the Gauss pseudospectral method is developed, which can generate both dynamically feasible and optimal flight trajectories. The entire process of solving a DCTSPN consists of two phases. First, in the offline preprocessing phase, the algorithm constructs a PRM, and then converts the original problem into a standard asymmet- ric TSP （ATSP）. Second, in the online querying phase, the costs of directed edges in PRM are updated first, and a fast heuristic searching algorithm is then used to solve the ATSP. Numerical experiments indicate that the algorithm proposed in this paper can generate both feasible and near-optimal solutions quickly for online purposes.展开更多
In order to achieve the optimal attack outcome in the air combat under the beyond visual range(BVR)condition,the decision-making(DM)problem which is to set a proper assignment for the friendly fighters on the hostile ...In order to achieve the optimal attack outcome in the air combat under the beyond visual range(BVR)condition,the decision-making(DM)problem which is to set a proper assignment for the friendly fighters on the hostile fighters is the most crucial task for cooperative multiple target attack(CMTA).In this paper,a heuristic quantum genetic algorithm(HQGA)is proposed to solve the DM problem.The originality of our work can be supported in the following aspects:(1)the HQGA assigns all hostile fighters to every missile rather than fighters so that the HQGA can encode chromosomes with quantum bits(Q-bits);(2)the relative successful sequence probability(RSSP)is defined,based on which the priority attack vector is constructed;(3)the HQGA can heuristically modify quantum chromosomes according to modification technique proposed in this paper;(4)last but not the least,in some special conditions,the HQGA gets rid of the constraint described by other algorithms that to obtain a better result.In the end of this paper,two examples are illustrated to show that the HQGA has its own advantage over other algorithms when dealing with the DM problem in the context of CMTA.展开更多
Based on the wide application of cloud computing and wireless sensor networks in various fields,the Sensor-Cloud System(SCS)plays an indispensable role between the physical world and the network world.However,due to t...Based on the wide application of cloud computing and wireless sensor networks in various fields,the Sensor-Cloud System(SCS)plays an indispensable role between the physical world and the network world.However,due to the close connection and interdependence between the physical resource network and computing resource network,there are security problems such as cascading failures between systems in the SCS.In this paper,we propose a model with two interdependent networks to represent a sensor-cloud system.Besides,based on the percolation theory,we have carried out a formulaic theoretical analysis of the whole process of cascading failure.When the system’s subnetwork presents a steady state where there is no further collapse,we can obtain the largest remaining connected subgroup components and the penetration threshold.Theoretically,this result is the critical maximum that the coupled SCS can withstand.To verify the correctness of the theoretical results,we further carried out actual simulation experiments.The results show that a scale-free network priority attack’s percolation threshold is always less than that of ER network which is priority attacked.Similarly,when the scale-free network is attacked first,adding the power law exponentλcan be more intuitive and more effective to improve the network’s reliability.展开更多
Object detection finds wide application in various sectors,including autonomous driving,industry,and healthcare.Recent studies have highlighted the vulnerability of object detection models built using deep neural netw...Object detection finds wide application in various sectors,including autonomous driving,industry,and healthcare.Recent studies have highlighted the vulnerability of object detection models built using deep neural networks when confronted with carefully crafted adversarial examples.This not only reveals their shortcomings in defending against malicious attacks but also raises widespread concerns about the security of existing systems.Most existing adversarial attack strategies focus primarily on image classification problems,failing to fully exploit the unique characteristics of object detectionmodels,thus resulting in widespread deficiencies in their transferability.Furthermore,previous research has predominantly concentrated on the transferability issues of non-targeted attacks,whereas enhancing the transferability of targeted adversarial examples presents even greater challenges.Traditional attack techniques typically employ cross-entropy as a loss measure,iteratively adjusting adversarial examples to match target categories.However,their inherent limitations restrict their broad applicability and transferability across different models.To address the aforementioned challenges,this study proposes a novel targeted adversarial attack method aimed at enhancing the transferability of adversarial samples across object detection models.Within the framework of iterative attacks,we devise a new objective function designed to mitigate consistency issues arising from cumulative noise and to enhance the separation between target and non-target categories(logit margin).Secondly,a data augmentation framework incorporating random erasing and color transformations is introduced into targeted adversarial attacks.This enhances the diversity of gradients,preventing overfitting to white-box models.Lastly,perturbations are applied only within the specified object’s bounding box to reduce the perturbation range,enhancing attack stealthiness.Experiments were conducted on the Microsoft Common Objects in Context(MS COCO)dataset using You Only Look Once version 3(YOLOv3),You Only Look Once version 8(YOLOv8),Faster Region-based Convolutional Neural Networks(Faster R-CNN),and RetinaNet.The results demonstrate a significant advantage of the proposed method in black-box settings.Among these,the success rate of RetinaNet transfer attacks reached a maximum of 82.59%.展开更多
深度神经网络在很多应用领域取得了显著成功。然而,近年来的研究表明,它们容易受到对抗攻击的威胁。尤其是有目标的对抗攻击,能够精确控制未知模型的输出,对数据隐私和系统安全构成严重挑战。生成式攻击方法因其高效生成对抗样本的能力...深度神经网络在很多应用领域取得了显著成功。然而,近年来的研究表明,它们容易受到对抗攻击的威胁。尤其是有目标的对抗攻击,能够精确控制未知模型的输出,对数据隐私和系统安全构成严重挑战。生成式攻击方法因其高效生成对抗样本的能力,近年来逐渐应用于有目标攻击的研究中。然而,现有的生成式攻击方法通常针对单一目标类别生成对抗样本,在多目标任务中表现出计算效率低下、灵活性不足和扩展性有限等问题。针对这些不足,提出了一种基于双重信息的多目标生成式攻击(multitarget generative attack based on dual-information,MTGA-DI)方法。该方法通过设计一个条件生成模型,充分融合目标类别的语义和视觉信息,不仅具备多目标攻击能力,还显著提升了对抗样本的迁移性和鲁棒性。实验结果表明,与现有多目标攻击方法相比,MTGA-DI在标准训练模型和鲁棒模型上的性能更优,在应对基于输入预处理的防御模型时也展现出更强的适应能力。展开更多
Combining the heuristic algorithm （HA） developed based on the specific knowledge of the cooperative multiple target attack （CMTA） tactics and the particle swarm optimization （PSO）, a heuristic particle swarm opt...Combining the heuristic algorithm （HA） developed based on the specific knowledge of the cooperative multiple target attack （CMTA） tactics and the particle swarm optimization （PSO）, a heuristic particle swarm optimization （HPSO） algorithm is proposed to solve the decision-making （DM） problem. HA facilitates to search the local optimum in the neighborhood of a solution, while the PSO algorithm tends to explore the search space for possible solutions. Combining the advantages of HA and PSO, HPSO algorithms can find out the global optimum quickly and efficiently. It obtains the DM solution by seeking for the optimal assignment of missiles of friendly fighter aircrafts （FAs） to hostile FAs. Simulation results show that the proposed algorithm is superior to the general PSO algorithm and two GA based algorithms in searching for the best solution to the DM problem.展开更多
文摘This paper considers the problem of generating a flight trajectory for a single fixed-wing unmanned combat aerial vehicle （UCAV） performing an air-to-surface multi-target attack （A/SMTA） mission using satellite-guided bombs. First, this problem is formulated as a variant of the traveling salesman problem （TSP）, called the dynamic-constrained TSP with neighborhoods （DCT- SPN）. Then, a hierarchical hybrid approach, which partitions the planning algorithm into a roadmap planning layer and an optimal control layer, is proposed to solve the DCTSPN. In the roadmap planning layer, a novel algorithm based on an updatable proba- bilistic roadmap （PRM） is presented, which operates by randomly sampling a finite set of vehicle states from continuous state space in order to reduce the complicated trajectory planning problem to planning on a finite directed graph. In the optimal control layer, a collision-free state-to-state trajectory planner based on the Gauss pseudospectral method is developed, which can generate both dynamically feasible and optimal flight trajectories. The entire process of solving a DCTSPN consists of two phases. First, in the offline preprocessing phase, the algorithm constructs a PRM, and then converts the original problem into a standard asymmet- ric TSP （ATSP）. Second, in the online querying phase, the costs of directed edges in PRM are updated first, and a fast heuristic searching algorithm is then used to solve the ATSP. Numerical experiments indicate that the algorithm proposed in this paper can generate both feasible and near-optimal solutions quickly for online purposes.
基金supported by National Nature Science Foundation of China,and the supporting project is“Study on parallel intelligent optimization simulation with combination of qualitative and quantitative method”(61004089)supported by the Graduate Student Innovation Practice Foundation of Beihang University in China(YCSJ-01-201205),which is“Research of an efficient and intelligent optimization method and application in aircraft shape design”.
文摘In order to achieve the optimal attack outcome in the air combat under the beyond visual range(BVR)condition,the decision-making(DM)problem which is to set a proper assignment for the friendly fighters on the hostile fighters is the most crucial task for cooperative multiple target attack(CMTA).In this paper,a heuristic quantum genetic algorithm(HQGA)is proposed to solve the DM problem.The originality of our work can be supported in the following aspects:(1)the HQGA assigns all hostile fighters to every missile rather than fighters so that the HQGA can encode chromosomes with quantum bits(Q-bits);(2)the relative successful sequence probability(RSSP)is defined,based on which the priority attack vector is constructed;(3)the HQGA can heuristically modify quantum chromosomes according to modification technique proposed in this paper;(4)last but not the least,in some special conditions,the HQGA gets rid of the constraint described by other algorithms that to obtain a better result.In the end of this paper,two examples are illustrated to show that the HQGA has its own advantage over other algorithms when dealing with the DM problem in the context of CMTA.
基金supported by National Natural Science Foundation of China under Grant No.62072412,61902359,U1736115in part by the Opening Project of Shanghai Key Laboratory of Integrated Administration Technologies for Information Security under Grant No.AGK2018001.
文摘Based on the wide application of cloud computing and wireless sensor networks in various fields,the Sensor-Cloud System(SCS)plays an indispensable role between the physical world and the network world.However,due to the close connection and interdependence between the physical resource network and computing resource network,there are security problems such as cascading failures between systems in the SCS.In this paper,we propose a model with two interdependent networks to represent a sensor-cloud system.Besides,based on the percolation theory,we have carried out a formulaic theoretical analysis of the whole process of cascading failure.When the system’s subnetwork presents a steady state where there is no further collapse,we can obtain the largest remaining connected subgroup components and the penetration threshold.Theoretically,this result is the critical maximum that the coupled SCS can withstand.To verify the correctness of the theoretical results,we further carried out actual simulation experiments.The results show that a scale-free network priority attack’s percolation threshold is always less than that of ER network which is priority attacked.Similarly,when the scale-free network is attacked first,adding the power law exponentλcan be more intuitive and more effective to improve the network’s reliability.
文摘Object detection finds wide application in various sectors,including autonomous driving,industry,and healthcare.Recent studies have highlighted the vulnerability of object detection models built using deep neural networks when confronted with carefully crafted adversarial examples.This not only reveals their shortcomings in defending against malicious attacks but also raises widespread concerns about the security of existing systems.Most existing adversarial attack strategies focus primarily on image classification problems,failing to fully exploit the unique characteristics of object detectionmodels,thus resulting in widespread deficiencies in their transferability.Furthermore,previous research has predominantly concentrated on the transferability issues of non-targeted attacks,whereas enhancing the transferability of targeted adversarial examples presents even greater challenges.Traditional attack techniques typically employ cross-entropy as a loss measure,iteratively adjusting adversarial examples to match target categories.However,their inherent limitations restrict their broad applicability and transferability across different models.To address the aforementioned challenges,this study proposes a novel targeted adversarial attack method aimed at enhancing the transferability of adversarial samples across object detection models.Within the framework of iterative attacks,we devise a new objective function designed to mitigate consistency issues arising from cumulative noise and to enhance the separation between target and non-target categories(logit margin).Secondly,a data augmentation framework incorporating random erasing and color transformations is introduced into targeted adversarial attacks.This enhances the diversity of gradients,preventing overfitting to white-box models.Lastly,perturbations are applied only within the specified object’s bounding box to reduce the perturbation range,enhancing attack stealthiness.Experiments were conducted on the Microsoft Common Objects in Context(MS COCO)dataset using You Only Look Once version 3(YOLOv3),You Only Look Once version 8(YOLOv8),Faster Region-based Convolutional Neural Networks(Faster R-CNN),and RetinaNet.The results demonstrate a significant advantage of the proposed method in black-box settings.Among these,the success rate of RetinaNet transfer attacks reached a maximum of 82.59%.
文摘深度神经网络在很多应用领域取得了显著成功。然而,近年来的研究表明,它们容易受到对抗攻击的威胁。尤其是有目标的对抗攻击,能够精确控制未知模型的输出,对数据隐私和系统安全构成严重挑战。生成式攻击方法因其高效生成对抗样本的能力,近年来逐渐应用于有目标攻击的研究中。然而,现有的生成式攻击方法通常针对单一目标类别生成对抗样本,在多目标任务中表现出计算效率低下、灵活性不足和扩展性有限等问题。针对这些不足,提出了一种基于双重信息的多目标生成式攻击(multitarget generative attack based on dual-information,MTGA-DI)方法。该方法通过设计一个条件生成模型,充分融合目标类别的语义和视觉信息,不仅具备多目标攻击能力,还显著提升了对抗样本的迁移性和鲁棒性。实验结果表明,与现有多目标攻击方法相比,MTGA-DI在标准训练模型和鲁棒模型上的性能更优,在应对基于输入预处理的防御模型时也展现出更强的适应能力。
文摘Combining the heuristic algorithm （HA） developed based on the specific knowledge of the cooperative multiple target attack （CMTA） tactics and the particle swarm optimization （PSO）, a heuristic particle swarm optimization （HPSO） algorithm is proposed to solve the decision-making （DM） problem. HA facilitates to search the local optimum in the neighborhood of a solution, while the PSO algorithm tends to explore the search space for possible solutions. Combining the advantages of HA and PSO, HPSO algorithms can find out the global optimum quickly and efficiently. It obtains the DM solution by seeking for the optimal assignment of missiles of friendly fighter aircrafts （FAs） to hostile FAs. Simulation results show that the proposed algorithm is superior to the general PSO algorithm and two GA based algorithms in searching for the best solution to the DM problem.
