Distributed cooperative control method for vehicle platoon under sensor and actuator dual attack

Abstract

The application relates to a vehicle fleet distributed cooperative control method under sensor and actuator dual attacks, and relates to the technical field of vehicle fleet safe operation control. The application solves the problems of poor environmental adaptability, insufficient attack compensation capability and insufficient robustness of traditional vehicle fleet cooperative safety control methods. The application adopts a hierarchical control idea, and the designed hierarchical control includes a basic cooperative controller of each vehicle, a first cooperative feedback controller of each vehicle, an advanced cooperative controller of each vehicle and a second cooperative feedback controller; the first cooperative feedback controller involved guarantees the stability of the system under sensor attacks, and the advanced cooperative controller is designed to offset the influence of attacks on the control loop; the actuator attack forms a negative compensation through the second cooperative feedback controller and is fed back to a feedforward compensation module of the vehicle, so that the system disturbance caused by the actuator attack can be resolved in real time, and the robustness of control is guaranteed. The application is mainly applied to vehicle fleet cooperative control.

Description

Technical Field

[0001] This invention relates to the field of fleet safety operation control technology. Background Technology

[0002] With the development of intelligent transportation systems, vehicle queuing control has significantly improved road capacity and energy efficiency by shortening vehicle spacing and optimizing traffic flow. However, the openness and interconnectivity of the Internet of Vehicles (IoV) expose it to the dual threats of actuator attacks (such as throttle / brake signal tampering) and sensor attacks (such as position / speed data forgery), which may lead to queue instability or even traffic accidents.

[0003] Actuator attacks disrupt vehicle dynamics by directly altering control commands. For example, an attacker could inject a false throttle signal, causing the vehicle to accelerate uncontrollably.

[0004] Sensor attacks can mislead control decisions by falsifying or interfering with sensor data (such as vehicle speed and spacing). For example, tampering with GPS signals can lead to incorrect calculations of queue spacing.

[0005] When a coordinated attack causes multiple vehicles to be attacked simultaneously, the attack effect may spread through inter-vehicle communication, triggering cascading failures.

[0006] Currently, a major challenge facing fleet systems is accurately modeling and detecting actuator and sensor attacks. Traditional attack detection methods, such as model-based, rule-based, or data-based methods, still face numerous challenges in their application to fleet systems. Due to the complexity and high dynamism of fleet systems, how to identify and defend against these attacks in real time and accurately remains an important research topic.

[0007] Furthermore, fleet systems require strong fault tolerance and robustness, especially in ensuring system stability and performance during attacks. Existing fault-tolerant control methods primarily rely on redundant system design or optimization of control algorithms to compensate for failed sensors or actuators. However, under attack scenarios, existing robust control methods still struggle to provide efficient protection, particularly in high-dimensional complex systems.

[0008] Meanwhile, the security issues of distributed cooperative control are particularly complex. Each vehicle in a fleet system collaborates through local information. Ensuring that the local control of each vehicle remains unaffected while maintaining global coordination in the face of sensor and actuator attacks is a crucial challenge. Traditional distributed control methods, lacking security safeguards, are vulnerable to attacks, leading to inconsistent or insecure system behavior.

[0009] Furthermore, attack defense mechanisms and recovery strategies are crucial for solving fleet system security issues. Existing research mainly focuses on attack detection, response mechanisms, and recovery strategies, such as using encrypted communication, firewalls, and fault detection algorithms to prevent attacks from occurring and spreading. However, how to quickly restore the system to normal operation, especially when actuators and sensors are attacked, remains a pressing problem to be solved.

[0010] In summary, existing protection technologies mainly focus on detection and isolation, with most research concentrating on attack detection (such as Kalman filtering combined with χ² detectors) and isolation (such as sensor redundancy), but lacking sufficient attack compensation and system recovery capabilities. Furthermore, existing technologies have poor adaptability to nonlinear and dynamic environments. In addition, existing technologies lack robustness in cooperative control, and traditional methods do not fully consider the attack propagation effects in multi-vehicle distributed cooperation.

[0011] Therefore, it is necessary to invent a multi-vehicle distributed cooperative control method that can adapt to complex environments and has strong robustness and attack compensation capabilities under dual attacks. Summary of the Invention

[0012] The purpose of this invention is to address the problems of poor environmental adaptability, insufficient attack compensation capabilities, and insufficient robustness in traditional fleet cooperative safety control methods. This invention provides a distributed cooperative control method for fleets under dual attacks from sensors and actuators.

[0013] A distributed cooperative control method for a fleet under dual attacks of sensors and actuators is proposed. This cooperative control method is used to control each vehicle in the fleet that has gained communication and whose sensors and actuators in two adjacent vehicles are attacked, thereby realizing the cooperative control of all two adjacent vehicles in the fleet that have gained communication under topology switching signals.

[0014] The control of each vehicle is achieved based on the vehicle's collaborative control system, which includes a basic collaborative controller, an advanced collaborative controller, a first collaborative feedback controller, a second collaborative feedback controller, and a feedforward compensation module.

[0015] The specific process of controlling each vehicle in a convoy that has established communication and whose sensors and actuators in adjacent vehicles have been attacked includes:

[0016] Set up communication with adjacent vehicles. and a car Inter-expected safe distance matrix ;

[0017] Construct the linear matrix inequality of observation gain and solve for the observation gain matrix. First positive definite matrix Second positive definite matrix ;according to and Obtain the feedback gain matrix ;

[0018] vehicle The basic collaborative controller, based on , ,vehicle estimated state ,vehicle estimated state Received a car Neighborhood error equation And in combination with the set controller parameters To generate a vehicle Basic control signals ; For time;

[0019] vehicle The first collaborative feedback controller, according to Sensor interference estimation first gain matrix Sensor interference estimation second gain matrix Sensor attack estimation matrix ,vehicle estimated state ,vehicle estimated state And from vehicles The sensor's output measurement value Received a car First feedback control signal ;

[0020] vehicle Advanced collaborative controller, based on , , and Constructed Lyapunov parameters ,as well as and Identify the vehicle Cooperative control function ; and then according to and Output vehicles Advanced control signals ;

[0021] Feedforward compensation module, according to , and Generate vehicles compensation signal ;

[0022] vehicle The second collaborative feedback controller, according to , , and the set actuator interference estimation gain matrix. and a car The actual output instructions of the actuator Received a car The second feedback control signal ;

[0023] Feedforward compensation module, according to and Output the final control signal For vehicles The actuators are controlled to achieve vehicle operation. The cooperative control system controls the vehicle's sensors and actuators under attack.

[0024] Preferably, in step S1, the expression for the linear matrix inequality of the observation gain is:

[0025] ;

[0026] in, The convergence parameter is greater than zero. For a single vehicle's system matrix, For the input matrix, Indicates the output matrix. Let be the gain estimation constant. It is a 1x1 identity matrix. It is a 2x2 identity matrix. It is a 3x3 identity matrix. This is a transpose.

[0027] Preferably, , ;

[0028] in, For the input matrix, For transpose, Topology switching signal The car that got off With a vehicle Topology parameters between, , For a vehicle With a vehicle The expected safe distance between them.

[0029] Preferably, according to , , and Constructed Lyapunov parameters The expression is:

[0030] .

[0031] Preferably, the vehicle is determined. Cooperative control function The implementation method is as follows:

[0032] , This represents the L2 norm.

[0033] Preferably, generating a vehicle Basic control signals The implementation methods include: .

[0034] Preferably, the vehicle is obtained. First feedback control signal The implementation methods include:

[0035] ;

[0036] ;

[0037] ;

[0038] ;

[0039] in, For vehicles The feedback state error of the first collaborative feedback controller, for The derivative, Topology switching signal The car that got off With a vehicle Topology parameters between, It is a vehicle The sensor measurements after compensation by the first collaborative feedback controller, For vehicles The estimated value of sensor attacks, For the output matrix, , For a vehicle With a vehicle The expected safe distance between them.

[0040] Preferably, the vehicle is obtained. Advanced control signals The implementation methods include:

[0041] .

[0042] Preferably, the vehicle is obtained. compensation signal The implementation methods include:

[0043] .

[0044] Preferably, the vehicle is obtained. The second feedback control signal The implementation methods include:

[0045] ;

[0046] in, Topology switching signal The car that got off With a vehicle Topology parameters between, Estimate the gain matrix for actuator disturbances.

[0047] Advantages of this invention:

[0048] This invention provides a distributed cooperative control method for vehicle platooning under dual attacks from sensors and actuators. This enables cooperative and secure control of the vehicle platooning system even when subjected to both actuator and sensor attacks. The invention fully considers the state of each vehicle within the system and the influence of the external environment, exhibiting strong adaptability to complex environments. It employs a hierarchical control approach, with the designed hierarchical control system including a basic cooperative controller for each vehicle, a first cooperative feedback controller for each vehicle, an advanced cooperative controller for each vehicle, and a second cooperative feedback controller. The underlying logic first designs the basic cooperative controllers for each vehicle to achieve cooperative and secure control of the vehicle system in the absence of attacks. Then, the first cooperative feedback controllers for each vehicle handle sensor attacks. Next, the advanced cooperative controllers for each vehicle dynamically correct residual errors caused by external attacks and the system itself. Finally, the second cooperative feedback controller handles actuator attacks. This method possesses strong attack compensation capabilities and can suppress attack propagation effects in multi-vehicle distributed cooperation.

[0049] A hierarchical distributed collaborative control method for the fleet system was adopted, improving the fleet's flexibility, adaptability, and reliability. Furthermore, it not only solves the problem of coordinated control of vehicle speeds within the fleet but also considers vehicle displacement path tracking within the fleet system, enhancing security. Moreover, it enables coordinated and secure control of the vehicle platoon system when it is attacked by actuators and sensors. The method effectively defends against attacks such as spoofed data injection and actuator hijacking without relying on hardware redundancy. It addresses the problems of weak environmental adaptability, insufficient robustness, inadequate attack compensation capabilities, and slow system recovery speed found in existing methods.

[0050] In this processing mode, sensor attacks (e.g., spoofing) and actuator attacks (e.g., command tampering) are treated as time-varying external disturbances. A first cooperative feedback controller ensures system stability under sensor attacks, and an advanced cooperative controller is designed to counteract the impact of attacks on the control loop. Actuator attacks are compensated for by a second cooperative feedback controller, which feeds back to the vehicle's feedforward compensation module. This real-time mitigation of system disturbances caused by actuator attacks ensures control robustness. It effectively defends against attacks such as spoofing and actuator hijacking without relying on hardware redundancy, and features high real-time performance and strong compatibility.

[0051] The designed linear matrix inequality for constructing observation gain can more accurately determine the required gain, avoiding the waste or insufficiency of gain caused by excessive or insufficient gain. Attached Figure Description

[0052] Figure 1 This is a schematic diagram illustrating the principle of the distributed cooperative control method for fleets under dual attacks from sensors and actuators as described in this invention.

[0053] Figure 2 This is a topology switching mode diagram for intelligent connected vehicles; among them, Figure 2 a to Figure 2 d represents the topology diagram of the first to fourth communication modes of the intelligent connected vehicle fleet; and number 0 is the leader vehicle, and numbers 1 to 4 are the first to fourth vehicles;

[0054] Figure 3 This is a switching signal diagram for the intelligent connected vehicle of this invention;

[0055] Figure 4 yes Figure 3 A schematic diagram showing the vehicle's position under the cyclical alternation of four communication modes;

[0056] Figure 5 for Figure 3 A schematic diagram illustrating the speed changes of a vehicle under the cyclical alternation of four communication modes;

[0057] Figure 6 for Figure 3 A schematic diagram illustrating the changes in vehicle acceleration under the cyclical alternation of four communication modes;

[0058] Figure 7 A graph showing the relationship between the actual position and the estimated position of each vehicle in the convoy;

[0059] Figure 8 A graph showing the relationship between the actual speed and the estimated speed of each vehicle in the convoy;

[0060] Figure 9 A graph showing the relationship between the actual acceleration and the estimated acceleration of each vehicle in the convoy; Detailed Implementation

[0061] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0063] This invention proposes a distributed cooperative control method for vehicle fleets under dual attacks from sensors and actuators. Based on the estimated states of observed vehicles, it utilizes a designed basic cooperative controller for each vehicle to achieve cooperative control of the vehicle system in the absence of attacks. Sensor attacks are then mitigated by a first cooperative feedback controller for each vehicle. Further, advanced cooperative controllers dynamically correct residual errors caused by external attacks and the system itself. Finally, actuator attacks are mitigated by a second cooperative feedback controller for each vehicle. This method effectively defends against attacks such as spoofed data injection and actuator hijacking without relying on hardware redundancy, and features high real-time performance and strong compatibility. It addresses the problems of poor environmental adaptability, insufficient robustness, inadequate attack compensation capabilities, and slow system recovery speed in existing methods. The specific implementation is as follows:

[0064] Detailed Implementation Method 1, see [link / reference] Figure 1 This embodiment describes a distributed cooperative control method for a fleet under dual attacks on sensors and actuators. This cooperative control method is used to control each vehicle in the fleet that has obtained communication and whose sensors and actuators are attacked in two adjacent vehicles, thereby achieving cooperative control of all two adjacent vehicles in the fleet that have obtained communication in the communication state.

[0065] The control of each vehicle is achieved based on the vehicle's collaborative control system, which includes a basic collaborative controller, an advanced collaborative controller, a first collaborative feedback controller, a second collaborative feedback controller, and a feedforward compensation module.

[0066] The specific process of controlling each vehicle in a convoy that has established communication and whose sensors and actuators in adjacent vehicles have been attacked includes:

[0067] Set up communication with adjacent vehicles. and a car Inter-expected safe distance matrix ;

[0068] Construct the linear matrix inequality of observation gain and solve for the observation gain matrix. First positive definite matrix Second positive definite matrix ;according to and Obtain the feedback gain matrix Specifically, ,in, For the input matrix, For transpose;

[0069] vehicle The basic collaborative controller, based on , ,vehicle estimated state ,vehicle estimated state Received a car Neighborhood error equation And in combination with the set controller parameters To generate a vehicle Basic control signals ; For time;

[0070] vehicle The first collaborative feedback controller, according to Sensor interference estimation first gain matrix Sensor interference estimation second gain matrix Sensor attack estimation matrix ,vehicle estimated state ,vehicle estimated state And from vehicles The sensor's output measurement value Received a car First feedback control signal ;

[0071] vehicle Advanced collaborative controller, based on , , and Constructed Lyapunov parameters ,as well as and Identify the vehicle Cooperative control function ; and then according to and Output vehicles Advanced control signals ;

[0072] Feedforward compensation module, according to , and Generate vehicles compensation signal ;

[0073] vehicle The second collaborative feedback controller, according to , , and the set actuator interference estimation gain matrix. and a car The actual output instructions of the actuator Received a car The second feedback control signal ;

[0074] Feedforward compensation module, according to and Output the final control signal For vehicles The actuators are controlled to achieve vehicle operation. The cooperative control system controls the vehicle's sensors and actuators under attack.

[0075] In this embodiment, the communication status of each vehicle in the convoy is changed during convoy operation via a topology switching signal. This clearly identifies adjacent vehicles that have established communication. and a car Determining the optimal safe distance not only improves fleet coordination efficiency and optimizes traffic flow, but also enhances fleet safety, automation, and real-time vehicle control. Furthermore, each topology switching signal corresponds to a topology switching mode. In this mode, only the status information of communicating and adjacent vehicles is needed, avoiding a large amount of unnecessary global computation and information transmission. Moreover, because it relies only on information from adjacent vehicles, the impact of faulty or unstable vehicles will not spread to the entire fleet, ensuring fleet reliability in complex environments.

[0076] The observation gain matrix is ​​solved by constructing the linear matrix inequality of observation gain. and This ensures that the convoy can achieve coordinated safety control under dual attacks, and allows for more accurate calculation of the required gain, avoiding waste or insufficient gain due to excessive or insufficient gain.

[0077] The estimated states of each vehicle can be achieved using existing technologies, specifically by analyzing the vehicle positions in their respective dynamic models. ,speed and acceleration The state observer, constructed using the linearized relationship between these relationships, enables the estimated state observation of the vehicle. This ensures the safety and stability of the vehicle even when all state information is unavailable.

[0078] The distributed cooperative control method for vehicle fleets under dual attacks from sensors and actuators provided in this embodiment adopts the concept of hierarchical control. The designed hierarchical control includes a basic cooperative controller for each vehicle, a first cooperative feedback controller for each vehicle, an advanced cooperative controller for each vehicle, and a second cooperative feedback controller. The underlying logic is to first design the basic cooperative controller for each vehicle to achieve cooperative safety control of the vehicle system in the absence of attacks, then use the first cooperative feedback controller for each vehicle to handle sensor attacks, then use the advanced cooperative controller for each vehicle to dynamically correct the residual errors caused by external attacks and the system itself, and finally use the second cooperative feedback controller to handle actuator attacks. It has strong attack compensation capabilities and can suppress the attack propagation effect in multi-vehicle distributed cooperation.

[0079] In this processing mode, sensor attacks (e.g., spoofing) and actuator attacks (e.g., command tampering) are treated as time-varying external disturbances. A first cooperative feedback controller ensures system stability under sensor attacks, and an advanced cooperative controller is designed to counteract the impact of attacks on the control loop. Actuator attacks are compensated for by a second cooperative feedback controller, which feeds back to the vehicle's feedforward compensation module. This real-time mitigation of system disturbances caused by actuator attacks ensures control robustness. It effectively defends against attacks such as spoofing and actuator hijacking without relying on hardware redundancy, and features high real-time performance and strong compatibility.

[0080] Furthermore, the observation gain matrix is ​​solved by constructing the linear matrix inequality of observation gain. First positive definite matrix Second positive definite matrix Specifically, in step S1, the expression for the linear matrix inequality of the observation gain is:

[0081] ;

[0082] in, The convergence parameter is greater than zero. For a single vehicle's system matrix, For the input matrix, Indicates the output matrix. Let be the gain estimation constant. It is a 1x1 identity matrix. It is a 2x2 identity matrix. It is a 3x3 identity matrix. This is a transpose.

[0083] In this preferred embodiment, the solution obtained from the given observation gain linear matrix inequality is... as well as More accurate, which can better guarantee the feedback gain matrix The gain effect and convergence parameters and gain estimation constant The introduction of this allows us to reasonably constrain [the system] according to the actual needs of the situation. and Size.

[0084] To be further specific, according to , , and Constructed Lyapunov parameters The expression is: ;

[0085] It also gives, .

[0086] Furthermore, we obtain the vehicle. Advanced control signals The implementation methods include:

[0087] ;

[0088] ;in, This represents the L2 norm.

[0089] In this preferred embodiment, the advanced cooperative controller of each vehicle is responsible for dynamically correcting errors to further realize the cooperative safety control of the fleet system in the event of a sensor attack when there is communication between the vehicle and other vehicles. The generated advanced control signal is transmitted to the feedforward compensation module, whose input signal is the estimated state from the vehicle itself. Estimated status from neighboring vehicles The expected real state and Lyapunov parameters The output signal is This allows for dynamic error correction. Furthermore, in this embodiment, sensor attacks (such as spoofing) are treated as external disturbances. A first cooperative feedback controller ensures system stability, and an advanced cooperative controller is designed to counteract the impact of the attack on the control loop.

[0090] Furthermore, the generated vehicle is given. Basic control signals The implementation methods include:

[0091] ;

[0092] ;

[0093] in, Topology switching signal The car that got off With a vehicle Topology parameters between, , For a vehicle With a vehicle The expected safe distance between them.

[0094] The basic control signal given in this preferred embodiment The specific implementation method has the advantage that, through the given controller parameters and feedback gain matrix, basic cooperative control of the vehicle can be well achieved under the condition of no external attack; separating the basic cooperative control from other controllers can avoid excessive coupling of the system, thereby improving working efficiency and providing a basic guarantee for cooperative control. Its input signal is the estimated state from the vehicle itself. Estimated status from neighboring vehicles and the expected real state The output signal is (Passed to the feedforward compensation module).

[0095] Specifically, it also provides information on obtaining a car. First feedback control signal The implementation methods include:

[0096] ;

[0097] ;

[0098] ;

[0099] ;

[0100] in, For vehicles The feedback state error of the first collaborative feedback controller, for The derivative, Topology switching signal The car that got off With a vehicle Topology parameters between, It is a vehicle The sensor measurements after compensation by the first collaborative feedback controller, For vehicles The estimated value of sensor attacks, For the output matrix, , For a vehicle With a vehicle The expected safe distance between them.

[0101] In this preferred embodiment, the input signals of the first cooperative feedback controller are signals from the vehicle itself. Signals from sensors Expected distance matrix and state estimation information from neighboring and communicating vehicles. In utilizing and An estimate of a sensor attack is generated within the first feedback controller. Afterwards, with Together, the compensated sensor measurements are generated. ,pass , as well as After processing within the first feedback controller, the final output feedback control signal is generated. .

[0102] Specifically, it also provides information on obtaining the vehicle. compensation signal The implementation methods include:

[0103] .

[0104] Specifically, to get a car The second feedback control signal The implementation methods include:

[0105] ;

[0106] in, Topology switching signal The car that got off With a vehicle Topology parameters between, Estimate the gain matrix for actuator disturbances.

[0107] In this preferred embodiment, the signal input of the second cooperative feedback controller is the control signal output by the basic cooperative controller. The control signal output by the first collaborative feedback controller Control signals of advanced co-controllers and the actual output instructions of the actuator The output signal is a negative compensation signal. (That is: the second feedback control signal).

[0108] Verification experiment:

[0109] Figure 2 The topology diagrams represent four different communication modes, which are connected via, for example... Figure 3 The switching method shown converts the communication mode between vehicles by applying a certain parameter to each vehicle. Figure 1 The control scheme shown is used to verify the feasibility of the present invention under the cyclical alternation of four communication modes.

[0110] like Figure 4 The diagram shown illustrates the vehicle's position under four alternating communication modes. Figure 4It can be seen that, even when the actuators and sensors are attacked, the positions of the vehicles remain within a safe and controllable range after the application of this cooperative control scheme, despite the continuous changes in the communication mode, demonstrating the effectiveness of this method.

[0111] like Figure 5 The diagram illustrates the speed changes of vehicles under four alternating communication modes. It shows that when the communication mode changes, the speed of each vehicle fluctuates due to actuator and sensor attacks. However, under the control scheme of this invention, the speed of each vehicle quickly returns to a stable normal speed, demonstrating the strong robustness of this method. Furthermore, the fact that the speed of each vehicle quickly returns to normal each time the communication mode changes indicates that this method effectively suppresses the propagation effect of attacks.

[0112] like Figure 6 The diagram illustrates the acceleration changes of a vehicle under four alternating communication modes. It shows that due to attacks on the actuators and sensors, the vehicle's acceleration fluctuates to some extent when the communication mode changes. However, under the control scheme of this invention, the acceleration trajectories of each vehicle quickly stabilize to zero. Furthermore, with each change in communication mode, the acceleration of each vehicle stabilizes to zero within a very short time, demonstrating that the method of this invention enables rapid coordinated control between vehicles, fast adjustment and recovery speed under attack, and strong attack compensation capabilities.

[0113] like Figures 7 to 9 These are graphs showing the actual position, velocity, and acceleration of each vehicle, and their corresponding estimated positions, velocities, and accelerations. Figures 7 to 9 As can be seen, the actual position, speed, and acceleration curves of each vehicle completely coincide with the estimated position, speed, and acceleration curves of each vehicle, indicating that the method of the present invention can better adapt to complex external environments and can effectively handle the impact of actuator attacks and sensor attacks on the system loop.

[0114] Furthermore, from Figures 7 to 9 As can be seen, by applying the method and inequalities of this invention, the estimation error can be converged to a compact set, and combined with... Figures 4 to 6 It can be seen that the tracking error between vehicles remains within a safe and controllable range. Therefore, this invention can achieve the cooperative control objective of position and speed coordination between adjacent and communicating vehicles in a vehicle platooning system. This demonstrates the effectiveness of the method.

[0115] In the cooperative control method of this invention, each vehicle has its own basic cooperative controller, first cooperative feedback controller, second cooperative feedback controller, and advanced cooperative controller. The advantage of this distributed approach is that each vehicle can independently and dynamically adjust its position and speed based on its own state and measurement output information, saving energy and improving efficiency. Furthermore, the hierarchical control design has strong attack compensation capabilities and can suppress the attack propagation effect in multi-vehicle distributed cooperation, exhibiting high real-time performance and strong compatibility.

[0116] This invention employs a distributed collaborative control method for fleet systems under dual attacks from sensors and actuators. Based on hierarchical control, it improves the fleet's flexibility, adaptability, and reliability; simultaneously, as... Figure 5 As shown, this not only solves the problem of coordinated speed control of vehicles within a convoy, but also simultaneously integrates vehicle displacement path tracking (such as...) into the convoy system. Figure 4 Taking this into account, it offers enhanced security. Furthermore, it enables collaborative security control of the vehicle queuing system when it is attacked by actuators and sensors. The method employed effectively defends against attacks such as spoofed data injection and actuator hijacking without relying on hardware redundancy. It addresses the shortcomings of existing methods, including weak environmental adaptability, insufficient robustness, inadequate attack compensation capabilities, and slow system recovery speed.

[0117] The distributed cooperative control method for convoys under dual attacks of sensors and actuators used in this invention is effective for controlling multiple vehicles and can ensure the speed coordination of the convoy (e.g., Figure 5 ) and the control of safe distance (such as Figure 4 ), which only requires each vehicle to communicate with and for information to be exchanged between adjacent vehicles (such as Figure 2 This eliminates the need for communication between each vehicle, significantly reducing the consumption of communication resources.

[0118] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. A distributed cooperative control method for vehicle fleets under dual attacks from sensors and actuators, characterized in that, This cooperative control method is used to control each vehicle in a convoy that has gained communication and whose sensors and actuators in two adjacent vehicles are under attack, thereby achieving cooperative control of all adjacent vehicles in the convoy that have gained communication under a topology switching signal. The control of each vehicle is achieved based on the vehicle's cooperative control system, which includes a basic cooperative controller, an advanced cooperative controller, a first cooperative feedback controller, a second cooperative feedback controller, and a feedforward compensation module. The specific process of controlling vehicles in a convoy that have established communication and whose sensors and actuators in adjacent vehicles have been attacked includes: Set up vehicles that can communicate and are adjacent. and vehicles Inter-expected safe distance matrix ; Construct the linear matrix inequality of observation gain and solve for the observation gain matrix. First positive definite matrix Second positive definite matrix ;according to and Obtain the feedback gain matrix ; The expression for the linear matrix inequality of observation gain is: ； in, The convergence parameter is greater than zero. For a single vehicle's system matrix, For the input matrix, Indicates the output matrix. Let be the gain estimation constant. It is a 1x1 identity matrix. It is a 2x2 identity matrix. It is a 3x3 identity matrix. For transpose; vehicle The basic collaborative controller, based on , ,vehicle estimated state ,vehicle estimated state Obtain the vehicle Neighborhood error equation And in combination with the set controller parameters To generate vehicles Basic control signals ; For time; vehicle The first collaborative feedback controller, according to Sensor interference estimation first gain matrix Sensor interference estimation second gain matrix Sensor attack estimation matrix ,vehicle estimated state ,vehicle estimated state and from vehicles The sensor's output measurement value Obtain the vehicle First feedback control signal ;in, ， ， ， ； For vehicles The feedback state error of the first collaborative feedback controller, for The derivative of Topology switching signal The vehicles that got off With vehicles Topology parameters between, It is a vehicle The sensor measurements after compensation by the first collaborative feedback controller, For vehicles The estimated value of sensor attacks, , For vehicles With vehicles The expected safe distance between them; vehicle Advanced collaborative controller, based on , , and Constructed Lyapunov parameters ,as well as and Identify the vehicle Cooperative control function ; and then according to and Output vehicles Advanced control signals ; in, , , Represents the L2 norm; Feedforward compensation module, according to , and Generate vehicles compensation signal ; vehicle The second collaborative feedback controller, according to , , and the set actuator interference estimation gain matrix. and vehicles The actual output instructions of the actuator Obtain the vehicle The second feedback control signal ; ;in, Topology switching signal The vehicles that got off With vehicles Topology parameters between, Estimate the gain matrix for actuator disturbances; Feedforward compensation module, according to and Output the final control signal For vehicles The actuators are controlled to achieve vehicle operation. The cooperative control system controls the vehicle's sensors and actuators under attack.

2. The vehicle distributed cooperative control method under dual attacks from sensors and actuators according to claim 1, characterized in that, ， 。 3. The vehicle distributed cooperative control method under dual attacks of sensors and actuators according to claim 1, characterized in that, Generate vehicles Basic control signals The implementation methods include: .

4. The distributed cooperative control method for fleets under dual attacks from sensors and actuators as described in claim 1, characterized in that, Obtain the vehicle Advanced control signals The implementation methods include: 。 5. The distributed cooperative control method for fleets under dual attacks from sensors and actuators according to claim 1, characterized in that, Obtain the vehicle compensation signal The implementation methods include: 。

Images