CN-121477939-B - Unmanned aerial vehicle system formation control method and system for resisting DoS attack

CN121477939BCN 121477939 BCN121477939 BCN 121477939BCN-121477939-B

Abstract

The application provides a method and a system for controlling formation of an unmanned aerial vehicle system for resisting DoS attack, and relates to the technical field of unmanned aerial vehicle formation control, wherein the method comprises the steps of constructing a dynamic model comprising a guiding unmanned aerial vehicle and a plurality of following unmanned aerial vehicles; the method comprises the steps of constructing a communication topology model according to communication links among unmanned aerial vehicles, introducing time-varying characteristics of DoS attacks based on the communication topology model to construct a dynamic Laplace matrix, fusing the dynamic model and the dynamic Laplace matrix to obtain a distributed state estimator, estimating flight states of the unmanned aerial vehicles to calculate estimated state deviation and formation tracking deviation of the unmanned aerial vehicles, generating control instructions by using a preset distributed state controller based on the estimated state deviation and the formation tracking deviation to drive the unmanned aerial vehicles to gradually converge to an expected formation state in a DoS attack environment, and accordingly achieving control targets of controlling unmanned aerial vehicle systems to achieve expected formation.

Inventors

WANG MINGKUAN
ZHANG JUAN

Assignees

东北大学
东北大学佛山研究生创新学院

Dates

Publication Date: 20260508
Application Date: 20260109

Claims (5)

1. The unmanned aerial vehicle system formation control method for resisting DoS attack is characterized by comprising the following steps of: according to the three-degree-of-freedom rotary motion of the unmanned aerial vehicle, constructing a dynamic model comprising a guiding unmanned aerial vehicle and a plurality of following unmanned aerial vehicles; Constructing a communication topology model according to the communication links between the guiding unmanned aerial vehicle and each following unmanned aerial vehicle and between each following unmanned aerial vehicle; Based on the communication topology model, introducing time-varying characteristics of the DoS attack to construct a dynamic laplace matrix, wherein elements in the dynamic laplace matrix are dynamically switched according to whether a corresponding communication link is subjected to the DoS attack or not, and the method comprises the steps that in a DoS attack period, elements corresponding to the communication link represent the communication link to be interrupted; The dynamic model and the dynamic Laplace matrix are fused, a distributed state estimator is designed, and the distributed state estimator is used for estimating the flight state of each unmanned aerial vehicle and calculating estimated state deviation and formation tracking deviation of each unmanned aerial vehicle based on the estimated flight state and expected formation state; based on the estimated state deviation and the formation tracking deviation, generating a control instruction by using a preset distributed state controller to drive each unmanned aerial vehicle to gradually converge to the expected formation state in a DoS attack environment; the dynamics model comprises a sub-dynamics model of the guiding unmanned aerial vehicle and a sub-dynamics model of the following unmanned aerial vehicle; The sub-dynamics model of the guiding unmanned aerial vehicle is as follows: ; the follow unmanned aerial vehicle's sub-dynamics model is: ; Wherein, the Respectively represent the first The frame follows the unmanned aerial vehicle and guides the flight state of the unmanned aerial vehicle, Representation of The vector of the real numbers is maintained, Respectively represent the first The frame follows the unmanned aerial vehicle and guides the flight observations of the unmanned aerial vehicle, Respectively represent the first Frame following unmanned aerial vehicle and guiding unmanned aerial vehicle The rate of change of the time of flight state; respectively represent the first The rack follows the input variables of the drone and the lead drone, Representation of A dimension real number vector; respectively represent the first The rack follows non-linear factors present in the drone and the lead drone, 、、 Respectively representing a system matrix, an input variable matrix and an observation matrix of the unmanned aerial vehicle system; the distributed state estimator is: ; Wherein, the Is a feedback gain matrix of the distributed state estimator, Representation of A dimension real number vector; Respectively represent Is a function of the estimated value of (2); Setting a node event triggering mechanism for the guided unmanned aerial vehicle, and setting an edge event triggering mechanism for each communication link between the guided unmanned aerial vehicle and each following unmanned aerial vehicle respectively; When the trigger condition of the node event trigger mechanism or the edge event trigger mechanism is reached, controlling the distributed state estimator to update the estimated value of the flight state of each unmanned aerial vehicle; Setting a node event triggering mechanism for the guided unmanned aerial vehicle, and setting an edge event triggering mechanism for each communication link between the guided unmanned aerial vehicle and each following unmanned aerial vehicle respectively; When the trigger condition of the node event trigger mechanism or the edge event trigger mechanism is reached, controlling the distributed state estimator to update the estimated value of the flight state of each unmanned aerial vehicle; the node event triggering mechanism is as follows: ; Wherein, the Representation of guided drone At the moment of the secondary triggering, Representing the maximum lower bound of the set, Representing node event trigger functions and , Indicating a predicted state deviation of the lead unmanned aerial vehicle, Representation of Guiding the predicted value of the flight state of the unmanned aerial vehicle at any time, Representing taking the norm of the vector or matrix, All represent trigger parameters; A threshold representing a node event trigger mechanism, and/or, ; Wherein, the Representation of a lead drone and a first Frame follows the first of the communication links between the drones At the moment of the secondary triggering, Representing the maximum lower bound of the set, Representing edge event trigger functions and , Representation of a lead drone and a first The rack follows the predicted state deviations between the drones, Representation of Guided drone and first time prediction The rack follows the relative state between the drones, A feedback gain matrix representing the edge event trigger function, Representing the predicted tracking deviation(s), Representing taking the norm of the vector or matrix, The trigger parameter is indicated as such, Representing adaptive coupling weights; A threshold value representing the edge event trigger mechanism, and/or, ; Wherein, the Represent the first Frame following unmanned aerial vehicle and first Frame follows the first of the communication links between the drones At the moment of the secondary triggering, Representing the maximum lower bound of the set, Representing edge event trigger functions and , Represent the first Frame following unmanned aerial vehicle and first The rack follows the predicted state deviations between the drones, Representation of Time of day (time) Frame following unmanned aerial vehicle and first The rack follows the relative state between the drones, A feedback gain matrix representing the edge event trigger function, Representing the predicted tracking deviation(s), Representing taking the norm of the vector or matrix, The trigger parameter is indicated as such, Representing adaptive coupling weights; A threshold value representing the edge event trigger mechanism; the distributed state controller is as follows: ; Wherein, the The input variable is represented by a representation of the input variable, All of which represent the adaptive coupling weights, adaptive coupling weights The derivative of (2) is expressed as And (2) and , Representation of Time of day (time) Frame following unmanned aerial vehicle and first The rack follows the real-time weight values of the communication links between the drones, Representing the predicted tracking deviation(s), 、 Are all indicated Time of day (time) Frame following unmanned aerial vehicle and first The rack follows the relative state between the drones, 、 Respectively represent the first Frame following unmanned aerial vehicle and first The frame follows the predicted value of the flight state of the unmanned aerial vehicle, The right upper corner mark T represents transposition and is the feedback gain matrix of the self-adaptive coupling weight The derivative of (2) is expressed as And (2) and , Representation of Time of day (time) The rack follows the real-time weight values of the communication link between the drone and the lead drone, Representing the predicted tracking deviation(s), Representation of Time-of-day guided drone and item The rack follows the relative state between the drones, Representation of Guiding the flight state estimated value of the unmanned aerial vehicle at any time; Representing the compensation signal.
2. The DoS attack resistant unmanned aerial vehicle system formation control method according to claim 1, wherein constructing a communication topology model from the communication links between the lead unmanned aerial vehicle and each following unmanned aerial vehicle and between each following unmanned aerial vehicle comprises: Adopting an undirected communication graph to represent communication links among all following unmanned aerial vehicles, wherein the undirected communication graph is as follows: ; Wherein, node set Indicating the presence of The frame follows the unmanned aerial vehicle, The representation is composed of The shelves follow a local communication topology of communication links between the drones, Represent the first Frame following unmanned aerial vehicle and first The frame follows the weight value of the communication link between the unmanned aerial vehicles; adding the guiding unmanned aerial vehicle into the undirected communication graph to obtain a tree-shaped communication topology model; wherein when the guiding unmanned aerial vehicle and the first unmanned aerial vehicle are When the frame is communicated with the communication link between the unmanned aerial vehicle, the guiding unmanned aerial vehicle and the first unmanned aerial vehicle Frame follows weight values of communication links between unmanned aerial vehicles 。
3. The unmanned aerial vehicle system formation control method of claim 2, wherein introducing the time-varying characteristic of DoS attacks based on the communication topology model to construct a dynamic laplace matrix comprises: define the time period sets of the DoS attack and the DoS attack disappearing as respectively 、 ; The real-time weight value of each communication link in the communication topology model is as follows: ; According to the real-time weight value of each communication link Build degree matrix ; According to the real-time weight value of each communication link in the local communication topology Obtaining an initial Laplace matrix , ; According to the degree matrix And the initial Laplace matrix Constructing a dynamic Laplace matrix 。
4. A method of controlling formation of a DoS attack resistant unmanned aerial vehicle system according to any of claims 1 to 3, wherein the method further comprises: establishing a mathematical model of the DoS attack, and taking the DoS attack and the period of disappearance of the DoS attack as an attack interval and a safety interval respectively through the mathematical model; Constructing a Lyapunov function; Based on different characteristics of the dynamic Laplace matrix in the attack interval and the safety interval, the change trend of the Lyapunov function is analyzed, and the change trend of the Lyapunov function proves that the real-time flight state of each unmanned aerial vehicle can be converged to the expected formation state.
5. An unmanned aerial vehicle system formation control system resistant to DoS attacks, comprising: The first construction module is used for constructing a dynamic model comprising a guiding unmanned aerial vehicle and a plurality of following unmanned aerial vehicles according to the three-degree-of-freedom rotary motion of the unmanned aerial vehicle; The second construction module is used for constructing a communication topology model according to the communication links between the guiding unmanned aerial vehicle and each following unmanned aerial vehicle and between each following unmanned aerial vehicle; The third construction module is used for introducing time-varying characteristics of the DoS attack based on the communication topology model to construct a dynamic laplace matrix, wherein elements in the dynamic laplace matrix are dynamically switched according to whether a corresponding communication link is subjected to the DoS attack or not, and the third construction module comprises the steps that in the DoS attack period, elements corresponding to the communication link represent the communication link to be interrupted; The fourth construction module is used for fusing the dynamic model and the dynamic Laplace matrix, designing a distributed state estimator, and calculating estimated state deviation and formation tracking deviation of each unmanned aerial vehicle based on the estimated flight state and expected formation state; The data generation module is used for generating control instructions by using a preset distributed state controller based on the estimated state deviation and the formation tracking deviation so as to drive each unmanned aerial vehicle to gradually converge the real-time flight state to the expected formation state in a DoS attack environment; the dynamics model comprises a sub-dynamics model of the guiding unmanned aerial vehicle and a sub-dynamics model of the following unmanned aerial vehicle; The sub-dynamics model of the guiding unmanned aerial vehicle is as follows: ; the follow unmanned aerial vehicle's sub-dynamics model is: ; Wherein, the Respectively represent the first The frame follows the unmanned aerial vehicle and guides the flight state of the unmanned aerial vehicle, Representation of The vector of the real numbers is maintained, Respectively represent the first The frame follows the unmanned aerial vehicle and guides the flight observations of the unmanned aerial vehicle, Respectively represent the first Frame following unmanned aerial vehicle and guiding unmanned aerial vehicle The rate of change of the time of flight state; respectively represent the first The rack follows the input variables of the drone and the lead drone, Representation of A dimension real number vector; respectively represent the first The rack follows non-linear factors present in the drone and the lead drone, 、、 Respectively representing a system matrix, an input variable matrix and an observation matrix of the unmanned aerial vehicle system; the distributed state estimator is: ; Wherein, the Is a feedback gain matrix of the distributed state estimator, Representation of A dimension real number vector; Respectively represent Is a function of the estimated value of (2); Setting a node event triggering mechanism for the guided unmanned aerial vehicle, and setting an edge event triggering mechanism for each communication link between the guided unmanned aerial vehicle and each following unmanned aerial vehicle respectively; When the trigger condition of the node event trigger mechanism or the edge event trigger mechanism is reached, controlling the distributed state estimator to update the estimated value of the flight state of each unmanned aerial vehicle; Setting a node event triggering mechanism for the guided unmanned aerial vehicle, and setting an edge event triggering mechanism for each communication link between the guided unmanned aerial vehicle and each following unmanned aerial vehicle respectively; When the trigger condition of the node event trigger mechanism or the edge event trigger mechanism is reached, controlling the distributed state estimator to update the estimated value of the flight state of each unmanned aerial vehicle; the node event triggering mechanism is as follows: ; Wherein, the Representation of guided drone At the moment of the secondary triggering, Representing the maximum lower bound of the set, Representing node event trigger functions and , Indicating a predicted state deviation of the lead unmanned aerial vehicle, Representation of Guiding the predicted value of the flight state of the unmanned aerial vehicle at any time, Representing taking the norm of the vector or matrix, All represent trigger parameters; A threshold representing a node event trigger mechanism, and/or, ; Wherein, the Representation of a lead drone and a first Frame follows the first of the communication links between the drones At the moment of the secondary triggering, Representing the maximum lower bound of the set, Representing edge event trigger functions and , Representation of a lead drone and a first The rack follows the predicted state deviations between the drones, Representation of Guided drone and first time prediction The rack follows the relative state between the drones, A feedback gain matrix representing the edge event trigger function, Representing the predicted tracking deviation(s), Representing taking the norm of the vector or matrix, The trigger parameter is indicated as such, Representing adaptive coupling weights; A threshold value representing the edge event trigger mechanism, and/or, ; Wherein, the Represent the first Frame following unmanned aerial vehicle and first Frame follows the first of the communication links between the drones At the moment of the secondary triggering, Representing the maximum lower bound of the set, Representing edge event trigger functions and , Represent the first Frame following unmanned aerial vehicle and first The rack follows the predicted state deviations between the drones, Representation of Time of day (time) Frame following unmanned aerial vehicle and first The rack follows the relative state between the drones, A feedback gain matrix representing the edge event trigger function, Representing the predicted tracking deviation(s), Representing taking the norm of the vector or matrix, The trigger parameter is indicated as such, Representing adaptive coupling weights; A threshold value representing the edge event trigger mechanism; the distributed state controller is as follows: ; Wherein, the The input variable is represented by a representation of the input variable, All of which represent the adaptive coupling weights, adaptive coupling weights The derivative of (2) is expressed as And (2) and , Representation of Time of day (time) Frame following unmanned aerial vehicle and first The rack follows the real-time weight values of the communication links between the drones, Representing the predicted tracking deviation(s), 、 Are all indicated Time of day (time) Frame following unmanned aerial vehicle and first The rack follows the relative state between the drones, 、 Respectively represent the first Frame following unmanned aerial vehicle and first The frame follows the predicted value of the flight state of the unmanned aerial vehicle, The right upper corner mark T represents transposition and is the feedback gain matrix of the self-adaptive coupling weight The derivative of (2) is expressed as And (2) and , Representation of Time of day (time) The rack follows the real-time weight values of the communication link between the drone and the lead drone, Representing the predicted tracking deviation(s), Representation of Time-of-day guided drone and item The rack follows the relative state between the drones, Representation of Guiding the flight state estimated value of the unmanned aerial vehicle at any time; Representing the compensation signal.

Description

Unmanned aerial vehicle system formation control method and system for resisting DoS attack Technical Field The application relates to the technical field of unmanned aerial vehicle formation control, in particular to a method and a system for unmanned aerial vehicle system formation control for resisting DoS attack. Background Along with the increase of electricity demand and the expansion of power grid scale, the number of power transmission lines formed by towers, wires and the like is increased, and unmanned aerial vehicles are required to carry out safety inspection on the power transmission lines. In the inspection process, multiple unmanned aerial vehicles are required to be mutually matched, for example, one unmanned aerial vehicle is used as a guiding unmanned aerial vehicle, the other multiple unmanned aerial vehicles are used as following unmanned aerial vehicles, the guiding unmanned aerial vehicle and the multiple following unmanned aerial vehicles are inspected according to a preset path and a formation, comprehensive inspection of a power transmission line is achieved, and technical support is provided for power transmission line detection of a smart grid. The process of controlling the lead drone and the plurality of follow drones to patrol according to the preset path and formation is also referred to as the formation control of the drone, which requires the plurality of drones to communicate their status information via the communication link. However, the communication link is vulnerable to denial of service (Denial of Service, doS) attacks and is cut off, resulting in a failure to share the flight status in time between multiple unmanned aerial vehicles, thereby breaking formation consensus and failing the inspection task. In order to overcome the interference of DoS attack on a communication link, the existing chinese patent CN119292057a proposes to construct a "virtual neighbor" by using the latest neighbor information received before the attack, and provide a stable reference target for an isolated unmanned aerial vehicle by using the state of the virtual neighbor, so that a passive communication topology switching mechanism is designed based on the state of the virtual neighbor, so that the unmanned aerial vehicle can automatically switch to a mode depending on the virtual neighbor after the communication link is interrupted, that is, in the case that the communication link is attacked by DoS, formation control can still be realized. However, such virtual neighbor mechanisms based on old information are prone to error accumulation in a DoS attack or high dynamic environment for a long time, resulting in serious hysteresis of state estimation, which may cause formation runaway. Meanwhile, the formation control method idealizes the flight models of the unmanned aerial vehicles into linear models, ignores nonlinear factors such as air resistance, attitude coupling and the like, and is difficult to adapt to complex interference in actual flight. For this reason, development of a formation control method with high fault tolerance and interference immunity is needed to cope with the challenges of communication link interruption. Disclosure of Invention The application provides a formation control method and a formation control system for an unmanned aerial vehicle system for resisting DoS attack, and aims to control the unmanned aerial vehicle system to achieve a control target of expected formation under the condition that a communication link is attacked by DoS. The technical scheme is as follows: In a first aspect, a method for controlling formation of an unmanned aerial vehicle system for resisting DoS attack is provided, where the method includes: according to the three-degree-of-freedom rotary motion of the unmanned aerial vehicle, constructing a dynamic model comprising a guiding unmanned aerial vehicle and a plurality of following unmanned aerial vehicles; Constructing a communication topology model according to the communication links between the guiding unmanned aerial vehicle and each following unmanned aerial vehicle and between each following unmanned aerial vehicle; Based on the communication topology model, introducing time-varying characteristics of the DoS attack to construct a dynamic laplace matrix, wherein elements in the dynamic laplace matrix are dynamically switched according to whether a corresponding communication link is subjected to the DoS attack or not, and the method comprises the steps that in a DoS attack period, elements corresponding to the communication link represent the communication link to be interrupted; The dynamic model and the dynamic Laplace matrix are fused, a distributed state estimator is designed, and the distributed state estimator is used for estimating the flight state of each unmanned aerial vehicle and calculating estimated state deviation and formation tracking deviation of each unmanned aerial vehicle based on the estimated flight stat