A method and system for secure authentication of a drone swarm

Abstract

The application relates to the technical field of unmanned aerial vehicle cluster security control, and particularly discloses an unmanned aerial vehicle cluster security authentication method and system. The method comprises the following steps: verifying the identity of each unmanned aerial vehicle and assigning roles through a security key and a digital certificate; performing security authentication on a generated flight path through a preset security authentication model, and performing secondary security authentication on the generated flight path through a preset secondary reinforcement security authentication model to identify potential security risks; performing global monitoring by a ground control center; and when a security event and a system failure are detected, immediately starting an emergency response program, executing a safety path re-planning, isolating affected unmanned aerial vehicles, and starting an automatic return program instruction. The application further discloses a corresponding security authentication system. The method and system can improve the security and stability of the unmanned aerial vehicle cluster through twice security authentication by the security authentication model.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle cluster, more particularly, the present application relates to a kind of unmanned aerial vehicle cluster security authentication method and system. BACKGROUND

[0002] With the rapid development of wireless communication technology and mobile devices towards intelligent, unmanned aerial vehicle cluster is widely used in dangerous environment monitoring, search and rescue tasks, etc., based on the interaction of a variety of sensors and actuators, unmanned aerial vehicle cluster urgently needs a variety of intelligent algorithms to support the interaction of sensors and actuators, to ensure the efficiency and success rate of unmanned aerial vehicle cluster to execute tasks, and when unmanned aerial vehicle distributes autonomous planning flight path, in order to avoid collision, it needs to consume a lot of computing resources, and the generated trajectory will exist the phenomenon of not smooth, so the use cost of unmanned aerial vehicle cluster is higher.The existing leader-follower algorithm, traditional dynamics equation method, reinforcement learning, artificial potential field and biological inspired obstacle avoidance algorithm of swarm intelligence has provided effective and efficient autonomous mechanism for the cooperation of unmanned aerial vehicle cluster, but the security of swarm intelligence algorithm in security authentication exists hidden danger, wind speed has a great influence on unmanned aerial vehicle flight, and the safety and sensitivity parameters of unmanned aerial vehicle cluster obstacle avoidance cannot be guaranteed, which affects the normal execution of unmanned aerial vehicle cluster task, and is easy to cause huge loss.

[0003] Therefore, it is urgent to provide an unmanned aerial vehicle cluster security authentication method and system to authenticate and reinforce the safety of unmanned aerial vehicle cluster obstacle avoidance, and improve the safety and stability of unmanned aerial vehicle cluster. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the present application provides an unmanned aerial vehicle cluster security authentication method and system, which authenticates and reinforces the safety of unmanned aerial vehicle cluster obstacle avoidance by presetting security authentication formula and secondary reinforcement security authentication formula, and improves the safety and stability of unmanned aerial vehicle cluster, to solve the problems proposed in the above background.

[0005] To achieve the above purpose, according to one aspect of the present application, an unmanned aerial vehicle cluster security authentication method is provided, comprising the following steps:

[0006] S100 Before starting the task, the identity of each unmanned aerial vehicle is verified and the role is assigned by security key and digital certificate;

[0007] S200 In the process of flight, real-time collection of environmental data and unmanned aerial vehicle state parameters is carried out, environmental monitoring and obstacle detection are carried out, the generated flight path is authenticated by presetting security authentication model, and the generated flight path is authenticated by presetting secondary reinforcement security authentication model, to identify potential security risks;

[0008] S300 The UAV cluster communicates with the ground control center in real time through the wireless communication network, transmits flight data and environmental monitoring results, and the ground control center uses these data for global monitoring;

[0009] S400 When a security event and system failure are detected, an emergency response program is immediately started, and instructions for re-planning a safe path, isolating affected UAVs, and starting an automatic return program are executed;

[0010] Further, the relative speed between the UAV and the obstacle determines the speed at which the UAV approaches the obstacle, reflecting the vector difference between the two objects. The relative speed determines the speed at which the UAV and the obstacle approach each other. The reaction time of the UAV cluster is the time required from the UAV's perception of the obstacle to the start of the obstacle avoidance action, including the total time for sensor detection, data processing, decision making, and control command issuance. The distance covered by the UAV within the reaction time is V rel ·T react The safety distance standard value is the minimum distance that needs to be maintained from the obstacle for safe flight. This distance depends on the size, speed, and maneuvering sensitivity of the UAV, and also takes into account the distance required for the UAV to stop or change direction in the worst case. The distance to the nearest obstacle is the straight-line distance between the UAV and the obstacle in front. This distance determines how much time and space the UAV has to take evasive action,

[0011] The preset safety authentication model is:

[0012]

[0013] In the formula: SEI is the safety evaluation index, D obs is the distance to the nearest obstacle, V rel is the speed of the UAV cluster relative to the obstacle, T react is the reaction time of the UAV cluster, S safe is the safety distance standard value, related to the size, speed, and maneuvering sensitivity of the UAV,

[0014] Based on the safety authentication formula, a secondary reinforcement safety authentication formula is further developed, which takes into account the impact of wind speed on the safety of the UAV's flight path. It is used to reflect the adaptability of the UAV to obstacle avoidance and path planning under different wind speed conditions. The preset secondary reinforcement safety authentication model is:

[0015]

[0016] In the formula: SEI wind is the safety evaluation index considering wind speed, V windLet f(V) be the wind speed of the wind field where the drone swarm is located. wind ) is a wind speed-related function used to adjust for relative velocity changes caused by wind speed, [V rel +f(V wind )]·T react This indicates the distance the drone will travel within its reaction time after taking wind speed into account. This distance is a key parameter for assessing whether the drone can safely stop or avoid obstacles when it encounters them. If this distance plus the safety distance is greater than the distance to the obstacle, it means that the drone cannot safely avoid the obstacle under the current speed and wind conditions and needs to slow down or adjust its flight path to improve safety.

[0017] As a preferred embodiment of the present invention, the parameters in the security authentication model are obtained from D. obs Data is acquired through radar and optical sensors, utilizing radar to detect distances of over 45 meters from the drone swarm. obs Using optical sensors to detect D within 45 meters of the drone swarm obs V rel Calculations were performed using an inertial measurement unit (IMU) and GPS. The IMU provides the instantaneous velocity and acceleration of the UAV swarm, while GPS provides the position data. react The preset constants for the drone swarm are determined through numerical analysis based on the processing speed and execution response time of the drone swarm.

[0018] As a preferred embodiment of the present invention, the parameters of the secondary reinforcement safety certification formula are obtained from f(V) wind Using machine learning models to learn V from flight data of drone swarms win d vs V rel The system automatically identifies the optimal function form based on the impact of changes, and the wind speed V in the wind field where the drone swarm is located. wind The results were obtained through numerical analysis using wind field simulation software.

[0019] Among them, f(V) is obtained using a machine learning model. wind This includes data acquisition and preprocessing, model selection and training, model evaluation and optimization. The acquired data includes wind speed, the actual speed of the UAV, flight altitude, and temperature. The machine learning model selected is a gradient boosting tree model. The model's performance is evaluated using mean squared error, and the model structure and parameters are adjusted based on the evaluation results. Wind field simulation software is used to obtain V... wind The process includes selecting wind field simulation software, acquiring wind field data, and predicting wind field data. The wind field simulation software includes, but is not limited to, Meteodyn WT. Based on the UAV's predicted flight time and low point, numerical analysis is performed using the wind field simulation software to obtain the wind speed V at the corresponding time and location.wind Predicted data, and data V wind as one of the inputs of the machine learning model.

[0020] As a preferred technical solution of the present application, when the safety evaluation index SEI obtained by the safety authentication formula is greater than or equal to 1, the current path is safe, and the current flight trajectory is executed; when the safety evaluation index SEI is less than 1, the flight path needs to be adjusted.

[0021] The safety evaluation index SEI considering the influence of wind speed obtained by the secondary reinforcement safety authentication formula wind When greater than or equal to 1, the current wind speed has no negative impact on the safety of the flight path, and the UAV cluster can improve the flight speed and optimize the flight path; the safety evaluation index SEI considering the influence of wind speed wind When less than 1, the current wind speed has a negative impact on the safety of the flight path of the UAV cluster, and the safety distance between the UAV cluster and the obstacle needs to be increased and the flight speed needs to be slowed down.

[0022] As a preferred technical solution of the present application, the safety evaluation index and the safety evaluation index considering the influence of wind speed are input into a preset improved risk assessment formula for safety event detection, and the improved risk assessment formula is:

[0023]

[0024] In the formula, R improved is an improved risk assessment index, ω and λ are weights of the safety evaluation index and the safety evaluation index considering the influence of wind speed respectively, and are adjusted according to actual task requirements, and ε is 0.01, which is used to avoid zero denominator.

[0025] In this scheme, the improved risk assessment formula is designed based on the mathematical principles of reciprocal and weighted average, aiming to comprehensively consider the influence of the basic safety evaluation index (SEI) and the safety evaluation index considering the influence of wind speed (SEI wind ) on the safety of UAV flight, and this formula makes the safety evaluation more flexible and accurate by introducing weights and a smoothing term ε, and the reciprocal is used in the formula to emphasize the importance of lower SEI or SEI wind values (i.e. higher risk levels) in risk assessment, because when SEI or SEI wind is smaller, its reciprocal will be larger, thereby giving a higher risk score in the calculation of R improved , reflecting a higher risk, the weights ω and λ are used to adjust the relative importance of SEI and SEI wind in risk assessment, allowing flexible adjustment of the contribution of the two to the risk score according to specific task requirements and flight environment, and the smoothing term ε is 0.01 to avoid the case of zero denominator and ensure that the formula is valid when SEI or SEIwind The stability and calculability at near zero, this small positive number also helps to handle extreme cases, avoiding the distortion of the evaluation results caused by extreme values. The reciprocal form makes the formula very sensitive to small changes in SEI and SEI wind , especially when these values are very small, helping to quickly identify and respond to potential risks, ensuring the safe flight of the UAV. By adjusting the weights ω and λ, the operator can adjust the focus of the safety assessment according to the actual situation. In environments where wind speed changes greatly, the value of λ can be increased, making the impact of wind speed weigh more heavily in the risk assessment.

[0026] As a preferred technical solution of the present application, according to R improved , the strategy for adjusting the flight parameters of the UAV cluster is: according to the task requirements, the low risk threshold and the high risk threshold of R improved are set in advance, the low risk threshold and the high risk threshold are R low and R high , respectively, R improved <R low , the flight task risk level is low risk, the UAV cluster maintains the current flight path and speed, and increases the flight height; R high ≥R improved ≥R low , the flight task risk level is medium risk, the flight speed is slowed down, the flight path is adjusted, the high-risk area is avoided, and the distance between the UAVs is increased; R improved >R high , the flight task risk level is high risk, the flight path is immediately adjusted, a safe area is found for temporary refuge, the UAV cluster slows down or temporarily hovers, and in extreme cases, an emergency landing instruction is started or the take-off point is returned.

[0027] As a preferred technical solution of the present application, when the flight task is low risk, the increase value of the flight height is wherein is the height adjustment experience coefficient, h re is the current flight height; when the flight task is medium risk, the slow-down value of the flight speed is wherein is the flight speed adjustment experience coefficient, v re is the current flight speed, and the increase value of the distance between the UAVs is wherein is the distance adjustment experience coefficient, d re is the current distance between the UAVs; when the flight task is high risk, the slow-down value of the UAV cluster speed is wherein is the high-risk speed sharp reduction experience coefficient, and

[0028] The flight safety of UAVs depends not only on avoiding collisions with fixed obstacles, but also on maintaining a safe distance from other UAVs and responding to sudden changes in wind speed. Adjusting flight height, speed, and spacing are effective means to achieve these goals. improved As a risk score that considers both basic flight safety and wind speed effects, it can reflect the overall risk level of the current flight mission. By linking this score with flight parameter adjustments, dynamic and intelligent adjustments of flight parameters can be achieved.

[0029] Experience coefficient The introduction of experience coefficients is based on practical experience in UAV flight safety management. These coefficients allow for flexible adjustment of response strategies for different risk levels based on historical data and flight safety research. Low-risk adjustment (increase in flight height): When the risk is low, increasing the flight height can further reduce the likelihood of collision with ground obstacles, and also provide more buffer space to avoid potential airborne obstacles (such as birds, etc.). Medium-risk adjustment (slow down flight speed and increase spacing): In the case of medium risk, slowing down the flight speed and increasing the spacing between UAVs can provide more reaction time and space for UAVs to respond to unexpected events or complex flight environments. High-risk adjustment (significant reduction in flight speed): When the risk level is high, the speed needs to be reduced quickly, where Reflects the more aggressive speed reduction strategy taken in high-risk situations to maximize obstacle avoidance success and reduce collision risk.

[0030] As a preferred technical solution of the present application, the specific process of global monitoring includes: selecting appropriate wireless communication technology according to the operating environment and task requirements of UAVs, building a communication network including ground control stations, relay UAVs, and target UAVs, collecting flight speed, position, and height in real time using sensors equipped on each UAV, collecting environmental monitoring data, transmitting the collected data back to the ground control center in real time, receiving and integrating the data of the UAV cluster by the ground control center, and constructing real-time flight status and environmental monitoring pictures.

[0031] According to another aspect of the present application, a UAV cluster security authentication system is also disclosed for the above-mentioned UAV cluster security authentication method.

[0032] According to still another aspect of the present application, a UAV cluster security authentication system is provided, comprising an initialization and identity verification module, a dynamic obstacle avoidance and path planning module, a real-time monitoring and safety control module, and a data analysis and risk assessment module, wherein,

[0033] The initialization and identity verification module is used to verify the identity of each UAV in the UAV cluster and assign roles through security keys and digital certificates before the task begins.

[0034] The dynamic obstacle avoidance and path planning module includes a flight control system and a data processing unit, the flight control system is used for preliminary obstacle avoidance and path planning, and the data processing unit performs safety authentication on the flight path through a preset safety authentication model and a secondary reinforced safety authentication model.

[0035] The real-time monitoring and safety control module includes a ground control center and a real-time data processing and decision support system, wherein the ground control center performs global monitoring on the UAV cluster according to the transmitted data, and the real-time data processing and decision support system is used to execute the flight parameter adjustment strategy.

[0036] The data analysis and risk assessment module is used to collect and analyze flight data and environmental monitoring data, learn and predict risks from the data by using a machine learning model, perform safety event detection, and develop a flight parameter adjustment strategy based on the safety event detection result.

[0037] Further, the initialization and identity verification module includes a key management system, an identity verification server and a security chip built-in the UAV, which is used to verify the identity of each UAV in the UAV cluster and assign roles through a security key and a digital certificate before the task starts.

[0038] The dynamic obstacle avoidance and path planning module includes optical sensors, radars and infrared sensors carried by the UAV, a flight control system and a data processing unit, wherein the flight control system is used to perform preliminary obstacle avoidance and path planning through a bio-inspired swarm intelligence algorithm and an artificial potential field method, and the data processing unit performs safety authentication on the flight path through a safety authentication formula and a secondary reinforced safety authentication formula.

[0039] The real-time monitoring and safety control module includes a wireless communication device, a ground control station, a relay UAV, a real-time data processing and decision support system, the ground control station performs global monitoring on the UAV cluster according to the transmitted data, and the real-time data processing and decision support system executes the flight parameter adjustment strategy and adjusts the flight path of the UAV or executes an emergency obstacle avoidance instruction when necessary.

[0040] The data analysis and risk assessment module collects and analyzes flight data and environmental monitoring data, learns and predicts risks from the data by using a machine learning model, and executes an improved risk assessment function formula to perform safety event detection and develop a flight parameter adjustment strategy.

[0041] Preferably, it further includes a communication network construction and management module for building and managing a communication network including the ground control station, the relay UAV and the target UAV, to ensure the real-time and safety of data.

[0042] Preferably, a security event response and recovery module is further included, comprising an emergency response system, an unmanned aerial vehicle autonomous navigation system, a fault diagnosis and processing unit

[0043] Overall, compared with the prior art, the above technical solutions conceived by the present application have the following beneficial effects:

[0044] The safety authentication method of the present application more accurately evaluates the safety of the flight path of the unmanned aerial vehicle through the safety evaluation index and the safety evaluation index considering the wind speed, especially in complex environmental conditions, not only considering the position of the obstacle and the speed of the unmanned aerial vehicle, but also especially considering the influence of wind speed on flight, thereby improving the safety and stability of the unmanned aerial vehicle flight. In addition, the present application scheme automatically identifies the best wind speed adjustment function through the machine learning model, which can flexibly adapt to the flight requirements under different wind speed conditions, and the wind speed data obtained through the wind field simulation software enables the unmanned aerial vehicle to more accurately predict and adapt to the flight environment it will face, further improving the safety and efficiency of flight. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 A flow chart of a safety authentication method of an unmanned aerial vehicle cluster according to an embodiment of the present application;

[0046] Figure 2 A structural block diagram of an unmanned aerial vehicle cluster system according to an embodiment of the present application. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and embodiments. In addition, the technical features involved in each embodiment of the present application described below can be combined with each other as long as there is no conflict between them.

[0048] As shown in Figure 1 A safety authentication method of an unmanned aerial vehicle cluster according to an embodiment of the present application includes the following steps:

[0049] Step S100, initialization and identity verification: before the task starts, the unmanned aerial vehicle cluster verifies the identity of each unmanned aerial vehicle and assigns roles through a security key and a digital certificate;

[0050] Step S200, Dynamic Obstacle Avoidance and Path Planning Safety Certification: During flight, environmental data and UAV status parameters are collected in real time through optical sensors, radar, and infrared sensors to perform environmental monitoring and obstacle detection. Initial obstacle avoidance and path planning are performed using bio-inspired swarm intelligence algorithms and artificial potential field methods. Reinforcement learning is combined to optimize the path and obstacle avoidance strategies. The generated flight path is certified for safety using a preset safety certification formula. Taking into account the influence of wind speed, the generated flight path is further reinforced for safety certification using a preset secondary reinforcement safety certification formula to identify potential safety hazards.

[0051] Step S300, Real-time monitoring and safety control: The UAV swarm communicates with the ground control center in real time through the wireless communication network, transmitting flight data and environmental monitoring results. The ground control center uses this data for global monitoring.

[0052] Step S400, Security Incident Response and Recovery: When a security incident is detected, execute the drone swarm flight parameter adjustment strategy.

[0053] When a system malfunction is detected, the flight parameter adjustment strategy for the drone swarm can also be implemented.

[0054] For example, adjusting strategies could involve adjusting flight paths, altitudes, speeds, and / or spacing, or initiating emergency response procedures to replan safe routes, isolate affected drones, and / or activate automatic return-to-home commands.

[0055] In step S200, the preset security authentication model or formula is:

[0056]

[0057] Where: SEI is the security assessment index, D obs V is the distance to the nearest obstacle. rel T represents the speed of the drone swarm relative to the obstacle. react S represents the reaction time of the drone swarm. safe This is a standard safe distance value, which is related to the size, speed, and control sensitivity of the drone.

[0058] The preset secondary reinforcement safety certification model or formula is as follows:

[0059]

[0060] In the formula: SEI wind To take into account the safety assessment index of wind speed, V wind Let f(V) be the wind speed of the wind field where the drone swarm is located. wind ) is a wind speed-related function used to adjust for relative speed changes caused by wind speed.

[0061] By using the safety evaluation index SEI and the safety evaluation index SEI considering the wind speed wind The safety risk faced by the UAV during flight can be comprehensively evaluated. Among them, the SEI considers the distance between the UAV and the obstacle, the relative speed of the UAV, the reaction time and the safety distance, while the SEI wind Further considering the wind speed, the accuracy and practicality of safety evaluation are improved, so that the UAV can dynamically avoid obstacles and plan paths according to real-time environmental conditions and its own state, significantly improving the safety of UAV flight. f(V wind ) in the formula is the influence function of wind speed on the relative speed of the UAV, which is learned from a large amount of flight data through machine learning, which enables the obstacle avoidance strategy to be dynamically adjusted according to the current wind speed, considering the dynamic obstacle avoidance and path planning method affected by the environmental factor wind speed, which enables the UAV to maintain efficient and safe flight in complex wind field conditions, reducing flight accidents caused by changes in wind speed; by transmitting flight data and environmental monitoring results to the ground control center in real time, and combining the safety authentication formula for real-time safety evaluation, the ground control center can fully master the flight state and environmental conditions of the UAV cluster. This real-time data monitoring and global control mechanism not only improves the adaptability of the UAV to environmental changes, but also enhances the response speed and efficiency to sudden situations, ensuring the smooth completion of the flight mission; when detecting safety events or system failures, the risk level is quickly identified according to the calculation results of SEI and SEI win , and the corresponding emergency response program is started. This data-driven emergency response mechanism enables the UAV cluster to take measures such as re-planning the flight path, isolating affected UAVs or starting automatic return when encountering risks, effectively reducing the likelihood of accidents and potential losses. The UAV cluster safety authentication method proposed in the present application realizes significant improvement in UAV flight safety, optimization of dynamic obstacle avoidance and path planning, and effective improvement of the emergency response mechanism through precise safety authentication formulas and secondary reinforcement safety authentication formulas, which is of great significance for improving the efficiency and success rate of UAV clusters in executing complex tasks.

[0062] It should be noted that the parameters in the safety authentication formula of step S200, wherein,

[0063] D obs is obtained by radar and optical sensors, D obs is obtained by radar and optical sensors, D obs ;

[0064] V relThe inertial measurement unit and GPS are used to calculate the instantaneous speed and acceleration of the UAV cluster, and the GPS provides position data.

[0065] T react The preset constant of the UAV cluster is determined by numerical analysis based on the processing speed and response time of the UAV cluster.

[0066] The scheme of the embodiments of the present application can accurately measure the distance between the UAV and the obstacle in different distance ranges by combining the use of radar and optical sensors. Radar is suitable for long-distance detection, while optical sensors provide high-precision image recognition at close range, enabling the UAV to more accurately judge and respond to the surrounding environment. The relative speed calculated by the inertial measurement unit and GPS not only provides the current speed and acceleration information of the UAV, but also ensures that the UAV adjusts the flight strategy according to the real-time position data, such as obstacle avoidance and path planning, enhancing the dynamic adaptability of the flight process. The reaction time determined by numerical analysis reflects the time required for the UAV to perceive the obstacle and complete the obstacle avoidance action. This preset considers the processing speed and response time of the UAV, and can be optimized based on the actual performance of the UAV to ensure that the safety certification formula can be applied to different types of UAV clusters. By accurately obtaining D obs and V rel , and combining appropriate T react , the UAV can more effectively plan an obstacle avoidance path, avoiding unnecessary sudden changes or emergency operations, thereby making the flight path smoother and more efficient. This method of using multiple sensors and technical means to obtain parameters not only improves the reliability of safety evaluation, but also makes the safety certification method applicable to UAV clusters of different performance and configuration, increasing the applicability and flexibility of the system.

[0067] It should be noted that in the parameters of the secondary reinforcement safety certification model or formula of step S200, f(V wind ) learns the influence of V wind on V rel change from the flight data of the UAV cluster using a machine learning model, automatically identifies the best function form, and the wind speed V wind of the wind field where the current UAV cluster is located is obtained by numerical analysis using wind field simulation software.

[0068] The machine learning model is used to automatically learn the influence of wind speed on the flight speed of the UAV, which can make the UAV more accurately consider environmental factors when planning the flight path, especially in complex or variable wind field conditions. The improvement of adaptability and accuracy helps the UAV to better avoid obstacles and improve flight efficiency. Through the analysis and application of real-time wind field data, the UAV can dynamically adjust its flight strategy to adapt to the current wind speed and direction, which not only reduces the flight deviation and potential danger caused by wind changes, but also optimizes energy consumption and improves flight endurance; wind speed is an important environmental factor that affects the safety of UAV cluster cooperation. By accurately calculating the influence of wind speed on flight speed, the UAV cluster can more effectively coordinate the obstacle avoidance and formation flight within the cluster, reduce the risk of collision, and ensure the overall safety of cluster flight.

[0069] It should be noted that in step S200, when the safety evaluation index SEI obtained by the safety authentication formula is greater than or equal to 1, the current path is safe, and the current flight trajectory is executed. When the safety evaluation index SEI is less than 1, the flight path needs to be adjusted.

[0070] The safety evaluation index SEI considering the influence of wind speed obtained by the secondary reinforcement safety authentication formula wind When greater than or equal to 1, the current wind speed has no negative impact on the safety of the flight path, and the UAV cluster can improve the flight speed and optimize the flight path. The safety evaluation index SEI considering the influence of wind speed wind When less than 1, the current wind speed has a negative impact on the safety of the flight path of the UAV cluster, and the safety distance between the UAV and the obstacle needs to be increased and the flight speed needs to be reduced.

[0071] The embodiments of the present application set a clear threshold value, so that the UAV control system can directly judge whether the current flight state is safe, simplifying the process of flight safety decision-making, helping to quickly make adjustment decisions, improving the reaction speed of the UAV cluster, and the safety evaluation index considering the influence of wind speed enables the UAV to dynamically adjust the flight strategy according to real-time wind speed data, such as reducing the flight speed or increasing the safety distance when the wind speed is relatively large and poses a threat to flight safety. Such adaptability enhances the safety of UAV flight in complex environments, and when SEI or SEI wind When greater than or equal to 1, it means that the current flight state of the UAV is relatively safe, and no excessive adjustment is needed, so the flight speed can be maintained or improved, and the flight path can be optimized, which not only improves the task execution efficiency, but also reduces energy consumption; by monitoring SEI and SEI wind in real time and adjusting the flight strategy when the value is less than 1, potential flight safety accidents can be effectively avoided, and the impact on the UAV cluster and task execution can be reduced; according to SEI and SEI windThe specific values of the risk response strategies are adopted at different levels, such as adjusting the flight path, changing the flight speed or height, and increasing the spacing between unmanned aerial vehicles, and such a hierarchical response mechanism enables the unmanned aerial vehicles to take more reasonable measures to respond to different levels of risk; by timely adjusting the flight strategy to respond to potential risks, the unmanned aerial vehicle cluster can more reliably complete various tasks, especially when performing key tasks such as search and rescue and environmental monitoring, the reliability of the system is particularly important.

[0072] It should be noted that in step S400, the safety evaluation index obtained in step S200 and the safety evaluation index considering the influence of wind speed are input into the preset improved risk assessment formula for safety event detection, and the improved risk assessment formula is:

[0073]

[0074] In the formula, R improved is the improved risk assessment index, ω and λ are the weights of the safety evaluation index and the safety evaluation index considering the influence of wind speed, respectively, which are adjusted according to actual task requirements, and ε is 0.01, which is used to avoid a zero denominator.

[0075] The formula improves the sensitivity to low SEI and SEI wind by using the reciprocal form, making the system more sensitive to potential risks in the flight path, even slight changes in risk can be detected and amplified, so that necessary risk avoidance measures can be taken in time; the introduction of weights ω and λ allows the influence of the basic safety evaluation index and the safety evaluation index considering the influence of wind speed to be adjusted according to actual flight tasks and environmental conditions, thereby refining risk assessment to ensure that risk assessment is more in line with actual flight conditions and improve the accuracy of risk control; different flight tasks and environments require different safety evaluation strategies, and by adjusting ω and λ, various flight scenarios can be flexibly adapted to, whether in open areas with large wind speed changes or in narrow spaces requiring precise obstacle avoidance; the introduction of a small positive number ε prevents the denominator from being zero when SEI or SEI wind approaches zero, ensuring that the risk assessment formula can function normally in all cases, avoiding calculation errors and ensuring the stability and reliability of the evaluation results; through the calculation of R improved , the safety risk level in flight can be more accurately identified, providing a scientific basis for developing and implementing safety strategies such as flight path adjustment, speed control, etc., thereby promoting the implementation of flight safety management; the improved risk assessment formula supports dynamic risk management, which can dynamically adjust the risk assessment results according to real-time flight data and environmental changes, providing continuous safety protection for unmanned aerial vehicle flight.

[0076] Optionally, in step S400, according to R improvedThe strategy for adjusting the flight parameters of a drone swarm is as follows:

[0077] Based on task requirements, pre-set R improved The low-risk threshold and high-risk threshold are R and R, respectively. low and R high R improved <R low At that time, the flight mission risk level was low, and the drone swarm maintained its current flight path and speed while increasing its flight altitude; R high ≥R improved ≥R low At that time, the flight mission risk level was medium risk. The flight speed was reduced, the flight path adjusted, high-risk areas avoided, and the spacing between drones increased. improved >R high If the flight mission is classified as high-risk, the flight path should be adjusted immediately, and a safe area should be sought for temporary refuge. The drone swarm should slow down or hover temporarily. In extreme cases, an emergency landing command should be initiated or the drone should return to the takeoff point.

[0078] The method in this application allows unmanned aerial vehicle (UAV) systems to dynamically adjust their flight behavior based on real-time risk assessments, achieving dynamic and real-time risk management and ensuring that flight safety strategies can respond promptly to environmental changes and potential threats; by using R improved Compared with preset risk thresholds, the system can clearly identify the current risk level faced by the drone and take corresponding safety measures, such as increasing flight altitude, adjusting flight path, decelerating, or hovering, significantly improving flight safety. Under low-risk conditions, the drone swarm can maintain its current flight path and speed, optimizing resource utilization and improving mission efficiency. Only when medium or high risk is detected will measures such as deceleration, obstacle avoidance, or emergency landing be taken, thus balancing safety and efficiency. This strategy allows the drone system to flexibly set R based on different mission requirements and environmental conditions. low and R high The value provides high adaptability and flexibility, making it suitable for various flight missions and environments. By defining clear response strategies for different levels of risk, unnecessary flight adjustments due to false alarms or minor risks can be avoided, thereby reducing overreaction and maintaining the continuity and stability of flight missions. The immediate response strategy for high-risk situations, such as emergency landing or return to the takeoff point, enhances the emergency response capability of the UAV system in the face of emergencies and reduces potential risks and losses.

[0079] Optionally, in step S400, when the flight mission is of low risk, the increase in flight altitude is... in To highly adjust the empirical coefficients, h reis the current flight height; the flight speed reduction value is wherein is the flight speed adjustment experience coefficient, v re is the current flight speed; the spacing between unmanned aerial vehicles increases by wherein is the spacing adjustment experience coefficient, d re is the current spacing between unmanned aerial vehicles; the unmanned aerial vehicle cluster slows down by wherein is the high-risk speed sharp reduction experience coefficient, and

[0080] By introducing the experience coefficient to adjust the flight parameters, the response of the unmanned aerial vehicle is allowed to be finely managed, and each adjustment measure is based on the real-time assessment of the risk level, ensuring that the response measure matches the actual risk and avoiding a one-size-fits-all approach; the use of the coefficient reflects the need to dynamically adjust the flight parameters to adapt to different risk levels, and this dynamic adaptation mechanism enables the unmanned aerial vehicle to maximize flight risk while maintaining task execution efficiency; in low-risk situations, unnecessary flight adjustments are avoided by increasing the flight height, while in medium to high-risk situations, safety is improved by slowing down the speed and adjusting the spacing, balancing flight safety and efficiency; special responses to high-risk situations, such as significant speed reduction, enhance the emergency handling capability of the unmanned aerial vehicle when encountering unexpected events or high-risk environments, helping to protect the unmanned aerial vehicle cluster from damage at critical moments and ensuring that the flight mission can be completed safely; the introduction of the experience coefficient provides flexibility in adjusting flight parameters, allowing more accurate and reliable decisions based on past experience and real-time data; by adjusting the spacing between unmanned aerial vehicles, the safety of collaborative work within the cluster is enhanced, especially in complex or emergency flight environments, reducing the risk of collision and ensuring safe flight of the entire cluster.

[0081] Optionally, the specific implementation process of step S300 includes: selecting an appropriate wireless communication technology according to the unmanned aerial vehicle operating environment and task requirements, building a communication network including a ground control station, a relay unmanned aerial vehicle and a target unmanned aerial vehicle, using sensors equipped on each unmanned aerial vehicle to collect flight speed, position and height in real time, while collecting environmental monitoring data, transmitting the collected data back to the ground control center in real time, the ground control center receives and integrates the data of the unmanned aerial vehicle cluster, and constructs a real-time flight state and environmental monitoring picture.

[0082] The scheme of the embodiment can realize comprehensive monitoring and management of the flight state of the unmanned aerial vehicle by collecting and transmitting flight data and environmental monitoring data to the ground control center in real time. The ground control center can immediately understand the position, speed, height and surrounding environment of each unmanned aerial vehicle, thereby providing strong data support for flight safety and task execution. Real-time data monitoring enables the ground control center to timely discover potential safety risks and take measures to intervene or adjust the flight strategy, thereby significantly improving the safety of unmanned aerial vehicle flight. When the unmanned aerial vehicle is about to enter a dangerous area, its flight path is immediately adjusted. According to the real-time collected flight data and environmental information, the ground control center can dynamically adjust the flight plan of the unmanned aerial vehicle according to the current situation, optimize the flight path, and improve the efficiency and success rate of task execution. The demand for real-time data is particularly important when the unmanned aerial vehicle cluster performs complex tasks, searches and rescues in a large range, and monitors the environment. By building an efficient communication network and real-time data transmission, the unmanned aerial vehicle cluster can better work together, share information, and improve the coordination and efficiency of task execution. Real-time monitoring data can also help the ground control center optimize resource allocation, adjust the use frequency and flight area of the unmanned aerial vehicle, ensure the concentrated use of resources in key areas, and reduce resource waste. In the event of an emergency, such as unmanned aerial vehicle failure or sudden environmental changes, real-time data transmission and monitoring capabilities enable the ground control center to quickly respond and guide the unmanned aerial vehicle to safely land or return, or dispatch other unmanned aerial vehicles for rescue, thereby minimizing losses. By integrating real-time flight data and environmental information, the ground control center can make more accurate and reasonable decisions. The data-driven decision-making process can improve the overall efficiency and safety of unmanned aerial vehicle flight operations.

[0083] As shown in Figure 2 The application also proposes a safety authentication system for unmanned aerial vehicle clusters based on the above safety authentication method.

[0084] Specifically, the safety authentication system includes an initialization and identity verification module, a dynamic obstacle avoidance and path planning module, a real-time monitoring and safety control module, a safety event response and recovery module, and a data analysis and risk assessment module

[0085] It can also preferably include a safety event response and recovery module, or more preferably include a communication network construction and management module.

[0086] The initialization and identity verification module is connected to the dynamic obstacle avoidance and path planning module. The dynamic obstacle avoidance and path planning module is connected to the real-time monitoring and safety control module. The real-time monitoring and safety control module is connected to the data analysis and risk assessment module. The data analysis and risk assessment module is connected to the safety event response and recovery module.

[0087] The security event response and recovery module is connected to the real-time monitoring and security control module, and the communication network construction and management module is connected to all other modules.

[0088] Before the task begins, the initialization and authentication module authenticates and assigns roles to each drone, ensuring that all drones participating in the task are authorized, thereby preventing unauthorized access and potential security threats. After authentication, the module assigns specific roles and tasks to each drone, such as leader or follower, as well as their specific position and responsibilities in the swarm.

[0089] After authentication, the drone enables its dynamic obstacle avoidance and path planning module, which uses optical sensors, radar, and infrared sensors on board the drone to collect real-time environmental data and drone state parameters for environmental monitoring and obstacle detection. Based on the collected data, combined with swarm intelligence algorithms, artificial potential field method, and reinforcement learning, the module generates and optimizes flight paths for the drone, achieving efficient dynamic obstacle avoidance.

[0090] After dynamic obstacle avoidance and path planning, the real-time monitoring and security control module collects flight data (speed, position, height) and environmental monitoring data from the drones through the communication network construction and management module, and transmits them back to the ground control center in real time. The ground control center uses the received data to monitor the entire swarm of drones, analyzes the flight state and environmental conditions, and if necessary, the module will guide the drones to adjust the flight path or execute emergency obstacle avoidance instructions to ensure flight safety.

[0091] Real-time monitoring data is transmitted to the data analysis and risk assessment module, which uses pre-set safety authentication formulas and secondary reinforced safety authentication formulas considering wind speed to comprehensively consider the impact of wind speed and other factors on flight safety, and assesses the safety risks in real time.

[0092] If the data analysis and risk assessment module detects that the safety risk exceeds the acceptable range, it can adjust the flight strategy through the security event response and recovery module. This module is responsible for starting emergency response measures, such as guiding drones to re-plan safe paths, isolating affected drones, or executing automatic return procedures, to deal with detected security incidents or system failures.

[0093] The communication network construction and management module provides continuous communication support throughout the entire flight task, ensuring real-time and accurate transmission of information between all modules and between drones and the ground control center, ensuring the coordinated operation of the system and real-time updating of information.

[0094] Specifically, the initialization and authentication module includes a key management system, an authentication server, and a security chip built into the drone.

[0095] The dynamic obstacle avoidance and path planning module includes optical sensors, radars, and infrared sensors carried by the UAV, a flight control system, and a data processing unit. The flight control system is used to perform preliminary obstacle avoidance and path planning through bio-inspired swarm intelligence algorithms and artificial potential field methods. The data processing unit performs safety authentication of the flight path through a safety authentication formula and a secondary reinforced safety authentication formula.

[0096] The real-time monitoring and safety control module includes wireless communication devices, a ground control station, relay UAVs, a real-time data processing and decision support system. The ground control station performs global monitoring of the UAV swarm based on transmitted data. The real-time data processing and decision support system adjusts the flight path of the UAV or executes emergency obstacle avoidance instructions when necessary.

[0097] The data analysis and risk assessment module collects and analyzes flight data and environmental monitoring data, uses machine learning models to learn from the data and predict risks, and executes an improved risk assessment function formula to detect safety events and develop flight parameter adjustment strategies.

[0098] The safety event response and recovery module includes an emergency response system, a UAV autonomous navigation system, a fault diagnosis and processing unit. The communication network construction and management module is used to build and manage the communication network including the ground control station, relay UAVs, and target UAVs, to ensure the real-time and safety of data.

[0099] Those skilled in the art will readily understand that the above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for security authentication of a UAV swarm, characterized in that, The method comprises the following steps: S100 Before starting the task, the unmanned aerial vehicle cluster verifies the identity of each unmanned aerial vehicle and assigns roles through a security key and a digital certificate; S200 Real-time collection of environmental data and unmanned aerial vehicle state parameters during flight, environmental monitoring and obstacle detection, security authentication of the generated flight path through a preset security authentication model, and secondary reinforcement security authentication of the generated flight path through a preset secondary reinforcement security authentication model to identify potential safety hazards; S300 Real-time communication between the unmanned aerial vehicle cluster and the ground control center through a wireless communication network, transmission of flight data and environmental monitoring results, and global monitoring by the ground control center using the flight data and the above monitoring result data; S400 Execution of unmanned aerial vehicle cluster flight parameter adjustment strategies when a safety event is detected; The preset security authentication model is: In the formula, is a safety evaluation index, is a distance from the nearest obstacle, is a speed of the UAV cluster relative to the obstacle, is a reaction time of the UAV cluster, is a safety distance standard value; The preset secondary reinforcement security authentication model is: wherein, is a safety evaluation index taking into account the wind speed, is the wind speed of the current wind field where the UAV cluster is located, is a function related to the wind speed, used to adjust the relative speed change caused by the wind speed; The safety evaluation index and the safety evaluation index considering the influence of the wind speed are input into a preset improved risk evaluation formula to detect a safety event, wherein the improved risk evaluation formula is: , wherein, is an improved risk evaluation index, and are a weight of the safety evaluation index and a weight of the safety evaluation index considering the influence of the wind speed, respectively, is a preset value. According to the improved risk assessment index The strategy for adjusting the flight parameters of the UAV cluster is to pre-set the low risk threshold and the high risk threshold of the improved risk assessment index ; if the risk assessment index is less than the low risk threshold , the risk level of the flight task is low, the UAV cluster maintains the current flight path and speed, and the flight height is increased; if the risk assessment index is greater than the low risk threshold and less than the high risk threshold , the risk level of the flight task is medium, the flight speed is slowed down, the flight path is adjusted, the high-risk area is avoided, and the distance between the UAVs is increased; if the risk assessment index is greater than the high risk threshold , the risk level of the flight task is high, the flight path is immediately adjusted, a safe area is searched for temporary refuge, the UAV cluster is slowed down or temporarily hovered, and in an extreme case, an emergency landing instruction is started or the take-off point is returned to. 2.The method of claim 1, wherein, The distance from the nearest obstacle in the security authentication model and / or the secondary reinforcement security authentication model Acquired by radar and optical sensors, wherein the distance from the UAV cluster is outside a preset threshold range Acquired by radar detection, wherein the distance from the UAV cluster is within a preset threshold range Acquired by optical sensor detection Speed of the drone swarm relative to an obstacle Calculated using an inertial measurement unit and GPS; Reaction time of the drone swarm Is a preset constant. 3.The method of claim 1, wherein, a function related to wind speed in the secondary reinforcement security authentication formula learning the wind speed of the wind field where the current UAV cluster is located from the flight data of the UAV cluster by using a machine learning model the speed of the UAV cluster relative to the obstacle the wind speed of the wind field where the current UAV cluster is located obtained by numerical analysis using wind field simulation software. 4.The method of claim 1, wherein, The security assessment index acquired by the security authentication model When greater than or equal to 1, execute the current flight path, and the security assessment index When less than 1, the flight path needs to be adjusted; and / or The secondary reinforcement security authentication model obtains a safety evaluation index considering the influence of wind speed When greater than or equal to 1, the current wind speed has no negative impact on the safety of the flight path; when less than 1, the current wind speed has a negative impact on the safety of the flight path of the unmanned aerial vehicle cluster.

5. The method of claim 1, wherein, The increase value of the flight height is The flight speed slow-down value is The spacing increase value between the unmanned aerial vehicles is The unmanned aerial vehicle cluster slow-down slow-down value when the flight task is high-risk is ; wherein is a height adjustment experience coefficient, is a current flight height, is a speed adjustment experience coefficient, is a current flight speed, is a distance adjustment experience coefficient, is a current distance between unmanned aerial vehicles, is a high risk speed sharp reduction experience coefficient, and .

6. The method of claim 1, wherein, The global monitoring process is: A communication network including a ground control center, a relay unmanned aerial vehicle, and target unmanned aerial vehicles is built; Flight speed, position, and height of each target unmanned aerial vehicle in the unmanned aerial vehicle cluster are collected in real time, and environmental monitoring data are collected simultaneously; The collected data are transmitted back to the ground control center in real time through the relay unmanned aerial vehicle; The ground control center receives and integrates the collected data of the unmanned aerial vehicle cluster for real-time monitoring.

Images