Drone swarm systems,equipped with photoelectric imaging and intelligent target perception,are essential for reconnaissance and strike missions in complex and high-risk environments.They excel in information sharing,an...Drone swarm systems,equipped with photoelectric imaging and intelligent target perception,are essential for reconnaissance and strike missions in complex and high-risk environments.They excel in information sharing,anti-jamming capabilities,and combat performance,making them critical for future warfare.However,varied perspectives in collaborative combat scenarios pose challenges to object detection,hindering traditional detection algorithms and reducing accuracy.Limited angle-prior data and sparse samples further complicate detection.This paper presents the Multi-View Collaborative Detection System,which tackles the challenges of multi-view object detection in collaborative combat scenarios.The system is designed to enhance multi-view image generation and detection algorithms,thereby improving the accuracy and efficiency of object detection across varying perspectives.First,an observation model for three-dimensional targets through line-of-sight angle transformation is constructed,and a multi-view image generation algorithm based on the Pix2Pix network is designed.For object detection,YOLOX is utilized,and a deep feature extraction network,BA-RepCSPDarknet,is developed to address challenges related to small target scale and feature extraction challenges.Additionally,a feature fusion network NS-PAFPN is developed to mitigate the issue of deep feature map information loss in UAV images.A visual attention module(BAM)is employed to manage appearance differences under varying angles,while a feature mapping module(DFM)prevents fine-grained feature loss.These advancements lead to the development of BA-YOLOX,a multi-view object detection network model suitable for drone platforms,enhancing accuracy and effectively targeting small objects.展开更多
The growing field of urban monitoring has increasingly recognized the potential of utilizing autonomous technologies,particularly in drone swarms.The deployment of intelligent drone swarms offers promising solutions f...The growing field of urban monitoring has increasingly recognized the potential of utilizing autonomous technologies,particularly in drone swarms.The deployment of intelligent drone swarms offers promising solutions for enhancing the efficiency and scope of urban condition assessments.In this context,this paper introduces an innovative algorithm designed to navigate a swarm of drones through urban landscapes for monitoring tasks.The primary challenge addressed by the algorithm is coordinating drone movements from one location to another while circumventing obstacles,such as buildings.The algorithm incorporates three key components to optimize the obstacle detection,navigation,and energy efficiency within a drone swarm.First,the algorithm utilizes a method to calculate the position of a virtual leader,acting as a navigational beacon to influence the overall direction of the swarm.Second,the algorithm identifies observers within the swarm based on the current orientation.To further refine obstacle avoidance,the third component involves the calculation of angular velocity using fuzzy logic.This approach considers the proximity of detected obstacles through operational rangefinders and the target’s location,allowing for a nuanced and adaptable computation of angular velocity.The integration of fuzzy logic enables the drone swarm to adapt to diverse urban conditions dynamically,ensuring practical obstacle avoidance.The proposed algorithm demonstrates enhanced performance in the obstacle detection and navigation accuracy through comprehensive simulations.The results suggest that the intelligent obstacle avoidance algorithm holds promise for the safe and efficient deployment of autonomous mobile drones in urban monitoring applications.展开更多
With the advancement of unmanned aerial vehicle(UAV)technology,the market for drones and the cooperation of many drones are expanding.Drone swarms move together in multiple regions to perform their tasks.A Ground Cont...With the advancement of unmanned aerial vehicle(UAV)technology,the market for drones and the cooperation of many drones are expanding.Drone swarms move together in multiple regions to perform their tasks.A Ground Control Server(GCS)located in each region identifies drone swarmmembers to prevent unauthorized drones from trespassing.Studies on drone identification have been actively conducted,but existing studies did not consider multiple drone identification environments.Thus,developing a secure and effective identification mechanism for drone swarms is necessary.We suggested a novel approach for the remote identification of drone swarms.For an efficient identification process between the drone swarm and the GCS,each Reader drone in the region collects the identification information of the drone swarmand submits it to the GCS for verification.The proposed identification protocol reduces the verification time for a drone swarm by utilizing batch verification to verify numerous drones in a drone swarmsimultaneously.To prove the security and correctness of the proposed protocol,we conducted a formal security verification using ProVerif,an automatic cryptographic protocol verifier.We also implemented a non-flying drone swarmprototype usingmultiple Raspberry Pis to evaluate the proposed protocol’s computational overhead and effectiveness.We showed simulation results regarding various drone simulation scenarios.展开更多
Drones have become indispensable tools in various domains, from surveillance and environmental monitoring to disaster response and communication relay. However, their growing use in critical missions necessitates robu...Drones have become indispensable tools in various domains, from surveillance and environmental monitoring to disaster response and communication relay. However, their growing use in critical missions necessitates robust security measures to protect against potential threats and ensure the integrity of operations. This research presents a novel secure architecture for a swarm of drones deployed on surveillance missions. Leveraging a reliable foundation established through Delaunay triangulation for communication among drones, this work introduces advanced security protocols to enhance the protection and integrity of the network. The architecture employs a mesh network topology connecting six drones, each configured for specific surveillance tasks, including perimeter monitoring, area scanning, thermal imaging, traffic observation, communication relay, and incident response. The mesh network design ensures extended coverage, redundancy, load balancing, and self-configuration, significantly improving reliability and resilience. Security validation was conducted using GNS3 and Ettercap, simulating various vulnerability scenarios. Comparative performance analysis between a classic drone network and the proposed secure mesh network demonstrates superior traffic management and robustness against potential attacks. The results underscore the architecture’s suitability for secure and reliable operations in critical surveillance environments.展开更多
In the third decade of the 21st century,the aerospace field is evolving at an unprecedented pace towards intelligence and autonomy.As competition for space resource development intensifies,breakthroughs in near-space ...In the third decade of the 21st century,the aerospace field is evolving at an unprecedented pace towards intelligence and autonomy.As competition for space resource development intensifies,breakthroughs in near-space vehicle technology emerge,and the concept of drone swarm warfare matures,traditional rule-based and experience-driven battlefield situation awareness models are struggling to meet the demands of complex adversarial environments.This special issue brings together the latest research findings from fields such as computer science and technology,electronic engineering,and cognitive science,systematically exploring the cognitive revolution driven by the deep integration of next-generation artificial intelligence and aerospace engineering,all centered around the core theme of"Intelligent Situation Awareness"(ISA).展开更多
The application field for Unmanned Aerial Vehicle (UAV) technology and its adoption rate have been increasingsteadily in the past years. Decreasing cost of commercial drones has enabled their use at a scale broader th...The application field for Unmanned Aerial Vehicle (UAV) technology and its adoption rate have been increasingsteadily in the past years. Decreasing cost of commercial drones has enabled their use at a scale broader thanever before. However, increasing the complexity of UAVs and decreasing the cost, both contribute to a lack ofimplemented securitymeasures and raise new security and safety concerns. For instance, the issue of implausible ortampered UAV sensor measurements is barely addressed in the current research literature and thus, requires moreattention from the research community. The goal of this survey is to extensively review state-of-the-art literatureregarding common sensor- and communication-based vulnerabilities, existing threats, and active or passive cyberattacksagainst UAVs, as well as shed light on the research gaps in the literature. In this work, we describe theUnmanned Aerial System (UAS) architecture to point out the origination sources for security and safety issues.Weevaluate the coverage and completeness of each related research work in a comprehensive comparison table as wellas classify the threats, vulnerabilities and cyber-attacks into sensor-based and communication-based categories.Additionally, for each individual cyber-attack, we describe existing countermeasures or detectionmechanisms andprovide a list of requirements to ensureUAV’s security and safety.We also address the problem of implausible sensormeasurements and introduce the idea of a plausibility check for sensor data. By doing so, we discover additionalmeasures to improve security and safety and report on a research niche that is not well represented in the currentresearch literature.展开更多
1 Introduction On-device deep learning(DL)on mobile and embedded IoT devices drives various applications[1]like robotics image recognition[2]and drone swarm classification[3].Efficient local data processing preserves ...1 Introduction On-device deep learning(DL)on mobile and embedded IoT devices drives various applications[1]like robotics image recognition[2]and drone swarm classification[3].Efficient local data processing preserves privacy,enhances responsiveness,and saves bandwidth.However,current ondevice DL relies on predefined patterns,leading to accuracy and efficiency bottlenecks.It is difficult to provide feedback on data processing performance during the data acquisition stage,as processing typically occurs after data acquisition.展开更多
Due to their advantages in flexibility,scalability,survivability,and cost-effectiveness,drone swarms have been increasingly used for reconnaissance tasks and have posed great challenges to their opponents on modern ba...Due to their advantages in flexibility,scalability,survivability,and cost-effectiveness,drone swarms have been increasingly used for reconnaissance tasks and have posed great challenges to their opponents on modern battlefields.This paper studies an optimization problem for deploying air defense systems against reconnaissance drone swarms.Given a set of available air defense systems,the problem determines the location of each air defense system in a predetermined region,such that the cost for enemy drones to pass through the region would be maximized.The cost is calculated based on a counterpart drone path planning problem.To solve this adversarial problem,we first propose an exact iterative search algorithm for small-size problem instances,and then propose an evolutionary framework that uses a specific encoding-decoding scheme for large-size problem instances.We implement the evolutionary framework with six popular evolutionary algorithms.Computational experiments on a set of different test instances validate the effectiveness of our approach for defending against reconnaissance drone swarms.展开更多
基金supported by the Natural Science Foundation of China,Grant No.62103052.
文摘Drone swarm systems,equipped with photoelectric imaging and intelligent target perception,are essential for reconnaissance and strike missions in complex and high-risk environments.They excel in information sharing,anti-jamming capabilities,and combat performance,making them critical for future warfare.However,varied perspectives in collaborative combat scenarios pose challenges to object detection,hindering traditional detection algorithms and reducing accuracy.Limited angle-prior data and sparse samples further complicate detection.This paper presents the Multi-View Collaborative Detection System,which tackles the challenges of multi-view object detection in collaborative combat scenarios.The system is designed to enhance multi-view image generation and detection algorithms,thereby improving the accuracy and efficiency of object detection across varying perspectives.First,an observation model for three-dimensional targets through line-of-sight angle transformation is constructed,and a multi-view image generation algorithm based on the Pix2Pix network is designed.For object detection,YOLOX is utilized,and a deep feature extraction network,BA-RepCSPDarknet,is developed to address challenges related to small target scale and feature extraction challenges.Additionally,a feature fusion network NS-PAFPN is developed to mitigate the issue of deep feature map information loss in UAV images.A visual attention module(BAM)is employed to manage appearance differences under varying angles,while a feature mapping module(DFM)prevents fine-grained feature loss.These advancements lead to the development of BA-YOLOX,a multi-view object detection network model suitable for drone platforms,enhancing accuracy and effectively targeting small objects.
文摘The growing field of urban monitoring has increasingly recognized the potential of utilizing autonomous technologies,particularly in drone swarms.The deployment of intelligent drone swarms offers promising solutions for enhancing the efficiency and scope of urban condition assessments.In this context,this paper introduces an innovative algorithm designed to navigate a swarm of drones through urban landscapes for monitoring tasks.The primary challenge addressed by the algorithm is coordinating drone movements from one location to another while circumventing obstacles,such as buildings.The algorithm incorporates three key components to optimize the obstacle detection,navigation,and energy efficiency within a drone swarm.First,the algorithm utilizes a method to calculate the position of a virtual leader,acting as a navigational beacon to influence the overall direction of the swarm.Second,the algorithm identifies observers within the swarm based on the current orientation.To further refine obstacle avoidance,the third component involves the calculation of angular velocity using fuzzy logic.This approach considers the proximity of detected obstacles through operational rangefinders and the target’s location,allowing for a nuanced and adaptable computation of angular velocity.The integration of fuzzy logic enables the drone swarm to adapt to diverse urban conditions dynamically,ensuring practical obstacle avoidance.The proposed algorithm demonstrates enhanced performance in the obstacle detection and navigation accuracy through comprehensive simulations.The results suggest that the intelligent obstacle avoidance algorithm holds promise for the safe and efficient deployment of autonomous mobile drones in urban monitoring applications.
基金supported by Institute of Information&Communications Technology Planning&Evaluation(IITP)grant funded by the Korea government(MSIT)(No.RS-2023-00225201,Development of Control Rights Protection Technology to Prevent Reverse Use of Military Unmanned Vehicles,50)by MSIT under the ITRC(Information Technology Research Center)Supported Program(IITP-2023-2018-0-01417,Industrial 5G Bigdata Based Deep Learning Models Development and Human Resource Cultivation,50)supervised by the IITP.
文摘With the advancement of unmanned aerial vehicle(UAV)technology,the market for drones and the cooperation of many drones are expanding.Drone swarms move together in multiple regions to perform their tasks.A Ground Control Server(GCS)located in each region identifies drone swarmmembers to prevent unauthorized drones from trespassing.Studies on drone identification have been actively conducted,but existing studies did not consider multiple drone identification environments.Thus,developing a secure and effective identification mechanism for drone swarms is necessary.We suggested a novel approach for the remote identification of drone swarms.For an efficient identification process between the drone swarm and the GCS,each Reader drone in the region collects the identification information of the drone swarmand submits it to the GCS for verification.The proposed identification protocol reduces the verification time for a drone swarm by utilizing batch verification to verify numerous drones in a drone swarmsimultaneously.To prove the security and correctness of the proposed protocol,we conducted a formal security verification using ProVerif,an automatic cryptographic protocol verifier.We also implemented a non-flying drone swarmprototype usingmultiple Raspberry Pis to evaluate the proposed protocol’s computational overhead and effectiveness.We showed simulation results regarding various drone simulation scenarios.
文摘Drones have become indispensable tools in various domains, from surveillance and environmental monitoring to disaster response and communication relay. However, their growing use in critical missions necessitates robust security measures to protect against potential threats and ensure the integrity of operations. This research presents a novel secure architecture for a swarm of drones deployed on surveillance missions. Leveraging a reliable foundation established through Delaunay triangulation for communication among drones, this work introduces advanced security protocols to enhance the protection and integrity of the network. The architecture employs a mesh network topology connecting six drones, each configured for specific surveillance tasks, including perimeter monitoring, area scanning, thermal imaging, traffic observation, communication relay, and incident response. The mesh network design ensures extended coverage, redundancy, load balancing, and self-configuration, significantly improving reliability and resilience. Security validation was conducted using GNS3 and Ettercap, simulating various vulnerability scenarios. Comparative performance analysis between a classic drone network and the proposed secure mesh network demonstrates superior traffic management and robustness against potential attacks. The results underscore the architecture’s suitability for secure and reliable operations in critical surveillance environments.
文摘In the third decade of the 21st century,the aerospace field is evolving at an unprecedented pace towards intelligence and autonomy.As competition for space resource development intensifies,breakthroughs in near-space vehicle technology emerge,and the concept of drone swarm warfare matures,traditional rule-based and experience-driven battlefield situation awareness models are struggling to meet the demands of complex adversarial environments.This special issue brings together the latest research findings from fields such as computer science and technology,electronic engineering,and cognitive science,systematically exploring the cognitive revolution driven by the deep integration of next-generation artificial intelligence and aerospace engineering,all centered around the core theme of"Intelligent Situation Awareness"(ISA).
基金the FederalMinistry of Education and Research of Germany under Grant Numbers 16ES1131 and 16ES1128K.
文摘The application field for Unmanned Aerial Vehicle (UAV) technology and its adoption rate have been increasingsteadily in the past years. Decreasing cost of commercial drones has enabled their use at a scale broader thanever before. However, increasing the complexity of UAVs and decreasing the cost, both contribute to a lack ofimplemented securitymeasures and raise new security and safety concerns. For instance, the issue of implausible ortampered UAV sensor measurements is barely addressed in the current research literature and thus, requires moreattention from the research community. The goal of this survey is to extensively review state-of-the-art literatureregarding common sensor- and communication-based vulnerabilities, existing threats, and active or passive cyberattacksagainst UAVs, as well as shed light on the research gaps in the literature. In this work, we describe theUnmanned Aerial System (UAS) architecture to point out the origination sources for security and safety issues.Weevaluate the coverage and completeness of each related research work in a comprehensive comparison table as wellas classify the threats, vulnerabilities and cyber-attacks into sensor-based and communication-based categories.Additionally, for each individual cyber-attack, we describe existing countermeasures or detectionmechanisms andprovide a list of requirements to ensureUAV’s security and safety.We also address the problem of implausible sensormeasurements and introduce the idea of a plausibility check for sensor data. By doing so, we discover additionalmeasures to improve security and safety and report on a research niche that is not well represented in the currentresearch literature.
基金supported by the National Science Fund for Distinguished Young Scholars(62025205)the National Natural Science Foundation of China(Grant Nos.62032020,62102317)CityU APRC Grant(9610633).
文摘1 Introduction On-device deep learning(DL)on mobile and embedded IoT devices drives various applications[1]like robotics image recognition[2]and drone swarm classification[3].Efficient local data processing preserves privacy,enhances responsiveness,and saves bandwidth.However,current ondevice DL relies on predefined patterns,leading to accuracy and efficiency bottlenecks.It is difficult to provide feedback on data processing performance during the data acquisition stage,as processing typically occurs after data acquisition.
基金supported by the National Natural Science Foundation of China(No.61872123)the Natural Science Foundation of Zhejiang Province(No.LR20F030002).
文摘Due to their advantages in flexibility,scalability,survivability,and cost-effectiveness,drone swarms have been increasingly used for reconnaissance tasks and have posed great challenges to their opponents on modern battlefields.This paper studies an optimization problem for deploying air defense systems against reconnaissance drone swarms.Given a set of available air defense systems,the problem determines the location of each air defense system in a predetermined region,such that the cost for enemy drones to pass through the region would be maximized.The cost is calculated based on a counterpart drone path planning problem.To solve this adversarial problem,we first propose an exact iterative search algorithm for small-size problem instances,and then propose an evolutionary framework that uses a specific encoding-decoding scheme for large-size problem instances.We implement the evolutionary framework with six popular evolutionary algorithms.Computational experiments on a set of different test instances validate the effectiveness of our approach for defending against reconnaissance drone swarms.
