WO2024065283A1 - 评估车辆风险的方法、装置以及监测攻击的系统 - Google Patents

评估车辆风险的方法、装置以及监测攻击的系统 Download PDF

Info

Publication number
WO2024065283A1
WO2024065283A1 PCT/CN2022/122163 CN2022122163W WO2024065283A1 WO 2024065283 A1 WO2024065283 A1 WO 2024065283A1 CN 2022122163 W CN2022122163 W CN 2022122163W WO 2024065283 A1 WO2024065283 A1 WO 2024065283A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
covert
attack signal
fdi
reachable set
Prior art date
Application number
PCT/CN2022/122163
Other languages
English (en)
French (fr)
Inventor
杨天赐
冯向兵
魏卓
付天福
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2022/122163 priority Critical patent/WO2024065283A1/zh
Publication of WO2024065283A1 publication Critical patent/WO2024065283A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling

Definitions

  • the present application relates to the field of vehicle safety, and more specifically, to a method and device for assessing vehicle risks and a system for monitoring attacks.
  • Cooperative and adaptive cruise control can be used as an automatic vehicle following system, which can enable connected and automated vehicles (CAVs) to travel on the road in a tightly coupled form by sharing the local sensor information of the vehicles based on vehicle-to-vehicle (V2V) wireless communication.
  • V2V vehicle-to-vehicle
  • the sensors and communication systems of CAVs are vulnerable to external attacks.
  • vehicle radars, lidars, cameras, and global navigation satellite systems (GNSS) may be vulnerable to various types of external attacks, and both the in-vehicle network and the out-of-vehicle network can be subject to multiple types of attacks, which allows false information to spread among multiple CAVs, which may cause traffic accidents and even casualties in the worst case.
  • GNSS global navigation satellite systems
  • Embodiments of the present application provide a method and device for evaluating vehicle risk and a system for monitoring attacks.
  • the method for evaluating vehicle risk can evaluate the impact of covert false data injection (FDI) attacks on vehicles, and the system for monitoring attacks can improve the vehicle's ability to resist covert FDI attacks.
  • FDI covert false data injection
  • the method provided in the present application can be applied to vehicles, which are vehicles in a broad sense, and can be means of transportation (such as commercial vehicles, passenger cars, trucks, motorcycles, airplanes, flying cars, trains, ships, etc.), industrial vehicles (such as forklifts, trailers, tractors, etc.), engineering vehicles (such as excavators, bulldozers, cranes, etc.), agricultural equipment (such as mowers, harvesters, etc.), amusement equipment, toy vehicles, etc.
  • the embodiments of the present application do not specifically limit the type of vehicle.
  • a method for assessing vehicle risk comprising: obtaining a covert FDI attack signal directed to a vehicle; determining a first reachable set and a danger set based on the covert FDI attack signal, the first reachable set comprising the dynamic state of the vehicle that changes with the covert FDI attack signal, and the danger set comprising the dynamic state of the vehicle when a dangerous situation occurs; determining the impact of the covert FDI attack signal on the vehicle based on a distance between the first reachable set and the danger set.
  • the first device may be a sensor of the vehicle, such as an inertial measurement unit (IMU), a lidar, a millimeter-wave radar, an ultrasonic radar, or other sensors, which is not specifically limited in the present application.
  • IMU inertial measurement unit
  • lidar a lidar
  • millimeter-wave radar a millimeter-wave radar
  • ultrasonic radar or other sensors, which is not specifically limited in the present application.
  • the covert FDI attack signal may include an attack signal that cannot be detected by the detector (or monitor).
  • the covert FDI attack signal may be an attack signal within the error range allowed by the detector, such as an attack signal hidden in the system interference of the vehicle; or, the covert FDI attack signal may also be an attack signal that the detector does not monitor at all, such as if the detector is not set to monitor abnormalities of the global positioning system (GPS) signal, then all GPS spoofing signals can be regarded as covert FDI attack signals.
  • GPS global positioning system
  • the acquired covert FDI attack signal targeting the vehicle is in the form of a mathematical expression, or may be in other forms.
  • the first reachable set includes all dynamic states of the vehicle that change with the covert FDI attack signal, wherein “all dynamic states” may include the dynamic state of the vehicle under the action of any possible covert FDI attack signal and/or system disturbance signal.
  • the covert FDI attack signal when the distance between the first reachable set and the danger set is less than or equal to zero, it means that the covert FDI attack signal is likely to affect vehicle safety, for example, it may cause the vehicle to exceed speed or deviate; when the distance between the first reachable set and the danger set is greater than zero, it means that the covert FDI attack signal will not cause danger to the vehicle.
  • the method further includes: quantitatively evaluating the impact of the covert FDI attack signal on the vehicle based on the volume of the first reachable set.
  • the volume of the first reachable set can be determined based on the approximate set of the first reachable set; alternatively, the first reachable set can be determined by a simulation method (such as Monte Carlo simulation) and then the volume of the first reachable set can be determined.
  • a simulation method such as Monte Carlo simulation
  • the impact of the covert FDI attack signal on the existing vehicle system can be evaluated by the volume of the first reachable set.
  • the larger the volume of the first reachable set the greater the disturbance effect of the covert attack signal on the vehicle dynamic state, which means that the covert FDI attack signal has a greater impact on the vehicle system.
  • the covert FDI attack signal is determined based on at least one of the following: a residual signal of a monitor, an estimation error of an observer, and sensor noise; wherein the observer is used to determine an estimation value of the sensor at the next moment based on a measurement value of the sensor; and the monitor is used to monitor an attack on the vehicle based on a difference between the measurement value of the sensor and the estimation value of the observer.
  • any attack signal that makes the difference between the sensor's measurement value and the observer's estimate value less than the threshold of the monitor can be identified as a covert FDI attack signal.
  • the covert FDI attack signal is associated with at least one of the following: a vehicle-to-vehicle (V2V) communication network, a global navigation satellite system (GNSS), a millimeter wave radar, an ultrasonic radar, and an in-vehicle communication network.
  • V2V vehicle-to-vehicle
  • GNSS global navigation satellite system
  • millimeter wave radar an ultrasonic radar
  • in-vehicle communication network is associated with at least one of the following: a vehicle-to-vehicle (V2V) communication network, a global navigation satellite system (GNSS), a millimeter wave radar, an ultrasonic radar, and an in-vehicle communication network.
  • V2V vehicle-to-vehicle
  • GNSS global navigation satellite system
  • millimeter wave radar millimeter wave radar
  • ultrasonic radar an ultrasonic radar
  • the covert FDI attack signal is hidden in the noise of the V2V communication network signal, or the noise of the GPS signal, or the noise of the millimeter wave radar signal, or the noise of the ultrasonic radar signal and transmitted to the vehicle.
  • “associated” can also be understood as: the covert FDI attack signal is an attack signal for at least one of the V2V communication network, GNSS, millimeter wave radar, ultrasonic radar, and in-vehicle communication network.
  • determining a first reachable set based on the covert FDI attack signal includes: determining an approximate set of the first reachable set based on the covert FDI attack signal, wherein the approximate set is determined by solving an optimization problem with the vehicle queue chord stability of the vehicle formation and/or the estimation error of the observer as constraints, wherein the vehicle is in the vehicle formation.
  • determining a first reachable set according to the covert FDI attack signal includes: determining the first reachable set by a Monte Carlo simulation method according to the covert FDI attack signal.
  • the first reachable set is determined by a simulation method, which can reduce the computational complexity of the process of determining the first reachable set.
  • the dangerous situation includes at least one of the following: the speed of the vehicle is greater than or equal to a speed threshold, the acceleration of the vehicle is greater than or equal to an acceleration threshold, and the distance between the vehicle and the nearest preceding vehicle is less than or equal to a distance threshold.
  • the speed threshold in different road sections may be the maximum speed allowed in the road section; the acceleration threshold may be determined based on the comfort and safety of the driver and passengers; the distance threshold may be set to 0 (a collision occurs when it is less than or equal to 0), or the distance threshold may also be set to a higher value to eliminate safety hazards.
  • the distance between the vehicle and the most immediately preceding vehicle may be the distance between a rear bumper of the most immediately preceding vehicle and a front bumper of the vehicle.
  • the gain of the monitor and/or observer is determined based on the covert FDI attack signal; wherein the gain is associated with a second reachable set, and the distance between the second reachable set and the danger value is greater than the preset threshold.
  • the preset threshold may be the above-mentioned preset threshold, that is, 0, or may be other values, which is not specifically limited in the embodiments of the present application.
  • the distance between the second reachable set and the danger value is greater than the preset threshold, which means that the covert FDI attack signal will not threaten the safety of the vehicle, for example, it will not cause the vehicle to exceed speed or deviate.
  • the gain corresponds to the second reachable set can be understood as that after the gain is determined by the covert FDI attack signal, the set formed by the dynamic state of the vehicle changing with the covert FDI attack signal is the second reachable set.
  • the gain of the monitor and/or observer can be re-determined according to the covert FDI attack signal to achieve the purpose of reducing the impact of the covert attack signal on the vehicle dynamics; wherein the second reachable set has a smaller volume, and the distance between the second reachable set and the danger value is greater than the preset threshold.
  • a device for assessing vehicle risk comprising: an acquisition unit, for acquiring a covert false information injection (FDI) attack signal directed to the vehicle; a processing unit, for determining a first reachable set and a danger set based on the covert FDI attack signal, the first reachable set comprising the dynamic state of the vehicle that changes with the covert FDI attack signal, and the danger set comprising the dynamic state of the vehicle when a dangerous situation occurs; based on the distance between the first reachable set and the danger set, determining the impact of the covert FDI attack signal on the vehicle.
  • FDI covert false information injection
  • the processing unit is further used to: determine the impact of the covert FDI attack signal on the vehicle based on the volume of the first reachable set.
  • the covert FDI attack signal is determined based on at least one of the following: a threshold of a monitor, an estimation error of an observer, sensor noise, and communication network noise; wherein the observer is used to determine an estimation value of the sensor at the next time step based on a measurement value of the sensor; and the monitor is used to monitor the attack on the vehicle based on a difference between the measurement value of the sensor and the estimation value of the observer.
  • the covert FDI attack signal is associated with at least one of the following: a vehicle-to-vehicle (V2V) communication network, a global navigation satellite system (GNSS), a millimeter wave radar, an ultrasonic radar, and an in-vehicle communication network.
  • V2V vehicle-to-vehicle
  • GNSS global navigation satellite system
  • millimeter wave radar an ultrasonic radar
  • in-vehicle communication network is associated with at least one of the following: a vehicle-to-vehicle (V2V) communication network, a global navigation satellite system (GNSS), a millimeter wave radar, an ultrasonic radar, and an in-vehicle communication network.
  • V2V vehicle-to-vehicle
  • GNSS global navigation satellite system
  • millimeter wave radar millimeter wave radar
  • ultrasonic radar an ultrasonic radar
  • the processing unit is used to: determine an approximate set of the first reachable set based on the covert FDI attack signal, wherein the approximate set is determined by solving a convex optimization problem with the stability of the vehicle queue of the vehicle formation and/or the estimation error of the observer as constraints, wherein the vehicle is in the vehicle formation.
  • the processing unit is used to: determine the first reachable set through a Monte Carlo simulation method according to the covert FDI attack signal.
  • the dangerous situation includes at least one of the following: the speed of the vehicle is greater than or equal to a speed threshold, the acceleration of the vehicle is greater than or equal to an acceleration threshold, the relative distance between the vehicle and other vehicles is less than a relative distance threshold
  • the processing unit is also used to: determine the gain of the monitor and/or observer based on the covert FDI attack signal when the distance between the first reachable set and the danger set is less than or equal to a preset threshold; wherein the gain is associated with a second reachable set, and the distance between the second reachable set and the danger value is greater than the preset threshold.
  • a system for monitoring attacks can be set in a computing platform of a vehicle.
  • the computing platform may include: at least one of an advanced driving assistance system (ADAS), a vehicle control unit (VCU), and an in-vehicle infotainment system (IVI); or may also include other computing platforms, such as an in-car application server (ICAS) controller, a body domain controller (BDC), a special equipment system (SAS), a media graphics unit (MGU), a body super core (BSC), an ADAS super core (ADAS super core), etc., which is not limited in the present application.
  • ICAS may include at least one of the following: a vehicle control server ICAS1, an intelligent driving server ICAS2, an intelligent cockpit server ICAS3, and an infotainment server ICAS4.
  • the system includes: an observer, which is used to determine the second data of the first device at a second moment based on the first data obtained from the first device of the vehicle at a first moment, wherein the second moment is a moment after the first moment; a monitor, which is used to determine the difference between the second data and third data, wherein the third data is the data obtained from the first device at the second moment; when the difference is greater than or equal to a first threshold, outputting an alarm message, wherein the alarm message is used to indicate that the vehicle is under attack; wherein the gain of the monitor is determined according to the FDI attack signal injected with covert false information.
  • the gain of the vehicle's attack monitoring system is updated according to the covert FDI attack signal, which helps to improve the vehicle's ability to resist covert FDI attacks, thereby improving vehicle safety and reducing vehicle risks.
  • the gain of the observer is determined according to the covert FDI attack signal.
  • the system also includes a controller for implementing basic vehicle control performance, and a gain of the controller is determined based on the covert FDI attack signal.
  • the basic vehicle control performance includes the stability of the vehicle tracking error system, the required convergence speed of the vehicle tracking error system, the chord stability of the vehicle platoon, etc.
  • the controller can control the vehicle's motion state by sending the desired motion state of the vehicle to the vehicle's actuator, where the desired motion state of the vehicle may include a desired speed, a desired acceleration, a desired steering wheel angle, etc.
  • the controller In addition to meeting the basic performance requirements of vehicle control, the controller must also ensure low risk of the vehicle.
  • the gain is determined based on a first reachable set and a danger set of the vehicle; wherein the first reachable set is determined based on the initial gains of the observer, the monitor, and the controller, the first reachable set includes the dynamic state of the vehicle that changes with the covert FDI attack signal, and the danger set includes the dynamic state of the vehicle when a dangerous situation occurs, wherein the distance between the first reachable set and the danger set is less than or equal to a second threshold.
  • the second threshold value may be 0, or may be other values, which is not specifically limited in the embodiments of the present application.
  • a distance between the second reachable set and the danger set is greater than the second threshold, and the second reachable set corresponds to the gain.
  • the second reachable set corresponds to the gain can be understood as that after the gain is determined by the covert FDI attack signal, the set formed by all dynamic states of the vehicle changing with the covert FDI attack signal is the second reachable set.
  • the volume of the second reachable set is smaller than the volume of the first reachable set.
  • a device for assessing vehicle risk comprising a processing unit and a storage unit, wherein the storage unit is used to store instructions, and the processing unit executes the instructions stored in the storage unit so that the device executes the method in any possible implementation of the first aspect.
  • the processing unit may include at least one processor, and the storage unit may be a memory, wherein the memory may be a storage unit within the chip (e.g., a register, a cache, etc.), or a storage unit located outside the chip within a mobile carrier (e.g., a read-only memory, a random access memory, etc.).
  • the memory may be a storage unit within the chip (e.g., a register, a cache, etc.), or a storage unit located outside the chip within a mobile carrier (e.g., a read-only memory, a random access memory, etc.).
  • a vehicle comprising a system according to any one of the implementations of the third aspect.
  • a server which includes the device in any implementation of the second aspect or the fourth aspect.
  • a computer program product comprising: a computer program code, when the computer program code is run on a computer, the computer executes the method in any possible implementation of the first aspect.
  • the above-mentioned computer program code can be stored in whole or in part on the first storage medium, wherein the first storage medium can be packaged together with the processor or separately packaged with the processor, and the embodiments of the present application do not specifically limit this.
  • a computer-readable medium stores instructions, and when the instructions are executed by a processor, the processor implements the method in any possible implementation manner of the first aspect.
  • a chip comprising a processor for calling a computer program or computer instructions stored in a memory so that the processor executes a method in any possible implementation of the first aspect.
  • the processor is coupled to the memory via an interface.
  • the chip system also includes a memory, in which a computer program or computer instructions are stored.
  • FIG1 is a functional block diagram of a vehicle provided in an embodiment of the present application.
  • FIG2 is a schematic diagram of a residual-based detector provided in an embodiment of the present application.
  • FIG3 is an exemplary flow chart of a method for assessing vehicle risk provided by an embodiment of the present application.
  • FIG4 is a schematic diagram of a reachable set and its approximate set provided in an embodiment of the present application.
  • FIG5 is a schematic diagram of a relationship between a reachable set and a dangerous set provided in an embodiment of the present application
  • FIG6 is a schematic diagram of the relationship between the reachable set and the dangerous set after gain adjustment provided by an embodiment of the present application.
  • FIG7 is a schematic block diagram of a system for monitoring attacks provided in an embodiment of the present application.
  • FIG8 is a schematic block diagram of a device for assessing vehicle risk provided by an embodiment of the present application.
  • FIG. 9 is a schematic block diagram of a device for assessing vehicle risk provided in an embodiment of the present application.
  • FIG1 is a functional block diagram of a vehicle 100 provided in an embodiment of the present application.
  • the vehicle 100 may include a perception system 120 and a computing platform 150, wherein the perception system 120 may include several sensors for sensing information about the environment around the vehicle 100.
  • the perception system 120 may include a positioning system, and the positioning system may be a GPS, or may be one or more of a Beidou system or other positioning systems, an IMU, a laser radar, a millimeter wave radar, an ultrasonic radar, and a camera device.
  • the computing platform 150 may include processors 151 to 15n (n is a positive integer), and the processor is a circuit with signal processing capability.
  • the processor may be a circuit with instruction reading and execution capability, such as a central processing unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a microprocessor), or a digital signal processor (DSP); in another implementation, the processor may implement certain functions through the logical relationship of a hardware circuit, and the logical relationship of the hardware circuit is fixed or reconfigurable, such as a hardware circuit implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as a field programmable gate array (FPGA).
  • ASIC application-specific integrated circuit
  • PLD programmable logic device
  • the process of the processor loading a configuration document to implement the hardware circuit configuration can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units.
  • it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), etc.
  • the computing platform 150 can also include a memory, the memory is used to store instructions, and some or all of the processors 151 to 15n can call the instructions in the memory and execute the instructions to implement the corresponding functions.
  • the vehicle 100 may include an advanced driving assistant system (ADAS), which utilizes one or more sensors in the perception system 120 (including but not limited to: lidar, millimeter wave radar, camera, ultrasonic sensor, global positioning system, inertial measurement unit) to obtain information from the surroundings of the vehicle, and analyzes and processes the obtained information to achieve functions such as obstacle perception, target recognition, vehicle positioning, automatic vehicle following, path planning, driver monitoring/reminder, etc., thereby improving the safety, automation and comfort of vehicle driving.
  • ADAS advanced driving assistant system
  • CACC can be used as an automatic vehicle following system, which can share the local sensor information of vehicles based on V2V wireless communication, enabling vehicles to drive on the road in a tightly coupled form.
  • V2V wireless communication enabling vehicles to drive on the road in a tightly coupled form.
  • the vehicle's sensors and communication systems are vulnerable to external attacks.
  • the novel residual-based detector can be used to identify sensor failures or external attacks on the communication system of the vehicle.
  • the sensor signal may be attacked by FDI, which in turn affects the safety of the vehicle.
  • the state observer (estimator) is first used to predict the sensor measurement value (hereinafter referred to as the estimated value) of the next time step based on the sensor measurement value that has not been attacked or has not failed and the vehicle mathematical model. Then, at the next time step, the difference between the sensor measurement value and the estimated value provided by the state observer (hereinafter referred to as the observer) is calculated as the residual.
  • the residual will be less than or equal to the preset threshold.
  • the monitor will output an alarm message.
  • the vehicle can take necessary response measures to reduce the further impact of the attack, such as shutting down the engine to prevent overspeeding or even collision with the vehicle in front.
  • the attack signal that cannot be detected by the detector (or monitor) is called a stealth or stealthy FDI attack signal
  • the attack caused by the stealth FDI attack signal is called a stealth FDI attack.
  • the stealth FDI attack signal can be an attack signal within the error range allowed by the detector, such as an attack signal hidden in the system interference of the vehicle.
  • the stealth FDI attack signal will not trigger the monitor to output an alarm message, affecting the driving safety of the vehicle, and further affecting the process of automated platooning. For example, it may cause vehicle speeding or vehicle collision.
  • automated platooning refers to a formation state in which multiple vehicles are driven in a very small distance based on the support of autonomous driving technology and V2V technology.
  • the multiple vehicles in the formation are called a vehicle formation (platoon).
  • the system interference of the vehicle may include the following three categories: sensor noise, network communication noise and vehicle modeling uncertainty.
  • sensor noise can be understood as useless signals in the signal output by the sensor, and its formation factors may include external factors and internal factors.
  • external factors include factors such as artificial or natural interference outside the sensor circuit, such as electromagnetic radiation; internal factors include sensor measurement errors, internal conductive particle discontinuity, semiconductor PN junction (p-n junction) capacitance effect, irregular movement of electrons inside the conductor caused by high temperature, etc.
  • Network communication noise may include: network noise caused by delay, packet loss, quantization, etc. in network communication, random thermal noise inherent in the channel and persisting, and impact noise caused by external factors.
  • Vehicle modeling uncertainty may include: some unpredictable vehicle dynamics inherent in the vehicle itself.
  • the embodiments of the present application provide a method and device for evaluating vehicle risks.
  • By quantifying the impact of covert FDI attack signals on vehicle dynamics it is possible to evaluate the risk level of a vehicle when it is subjected to a covert FDI attack.
  • the system for detecting attack signals can be redesigned based on the covert FDI attack signals.
  • By monitoring attacks through the system it is possible to timely discover and trigger response measures when a vehicle system fails or a random FDI attack occurs. On this basis, the monitoring system can also reduce the impact of covert FDI attack signals on vehicle dynamics and improve driving safety.
  • Covert GPS attack The GPS position is adjusted slightly so that the false navigation route generated after the adjustment matches the shape of the actual road, and physical instructions are triggered to make the vehicle reach the wrong destination indicated by the false navigation route. Since the amplitude of the GPS spoofing signal is small, smaller than the impact of GPS drift, it cannot trigger the monitor to output an alarm message, so this covert attack can bypass the monitor, that is, it will not be detected by the monitor.
  • Covert attack on radar By hiding the attack signal on the radar in the vehicle's system interference, when the attack signal cannot trigger the monitor to output an alarm message, the attack can bypass the monitor. This attack is a covert FDI attack.
  • the speedometers and accelerators of vehicles may be attacked by electromagnetic signals or physical attacks. When these attack signals are hidden in the system interference of the vehicle and cannot trigger the monitor to output alarm information, the attack can bypass the monitor. This attack is a covert FDI attack.
  • Covert attack on V2V channel Assume that for each vehicle in the platoon, the observer and controller in the detector used to detect the attack are designed to achieve the best estimation and control performance without considering safety. In this case, if a vehicle in a platoon sends sensor measurements to another vehicle in the platoon, it can attack the other vehicle by injecting false data into its sensor measurements. If the injected false data cannot trigger the monitor to output an alarm message, the attack can bypass the monitor, and this attack is a covert FDI attack.
  • Covert attack on the in-vehicle communication network The sensor signal is changed during the in-vehicle network transmission process. The amplitude of the signal change cannot trigger the monitor to output an alarm message. The attack can bypass the monitor. This attack is a covert FDI attack.
  • FIG3 shows a schematic flow chart of a method 300 for assessing vehicle risk provided by an embodiment of the present application.
  • the attack monitoring system designed by the method 300 can be applied to the vehicle 100 shown in FIG1 .
  • the method 300 may include:
  • the covert FDI attack signal may include at least one of the following: a covert FDI attack signal introduced through a V2V communication network, a covert FDI attack signal introduced through a GPS signal, a covert FDI attack signal introduced through a millimeter wave radar or an ultrasonic radar, and a covert FDI attack signal introduced through an in-vehicle communication network.
  • the following takes the covert FDI attack signal introduced through the V2V communication network as an example to illustrate the method for determining the covert FDI attack signal.
  • Step (1) Determine the vehicle motion state model of the vehicle.
  • the following system model can be constructed for the longitudinal motion state of the vehicle:
  • xi [ ⁇ iviaiui ⁇ viai -1 ] T , ([ ⁇ ] T represents the transpose of the matrix ), ⁇ i is the tracking error, vi is the actual speed of the ith vehicle, ai and ai -1 are the actual accelerations of the ith and i-1th vehicles respectively, ⁇ vi is the relative speed of the ith and i-1th vehicles, ui and ui -1 are the expected accelerations of the ith and i-1th vehicles respectively, ⁇ i is the abnormal signal of the V2V network, which may be injected by malicious vehicles or caused by abnormalities in the V2V network; ⁇ ui is the disturbance in the V2V network, which may be caused by factors such as network packet loss, delay, quantization, etc., and satisfies is the perturbation boundary, It should be understood that is a set of positive real numbers.
  • K [k p k d ] is the gain of the vehicle controller.
  • h is a time interval constant
  • each vehicle has a millimeter-wave radar, an acceleration sensor, and a velocity sensor, which can directly or indirectly measure ⁇ i , v i , a i , u i , ⁇ v i , the vehicle's sensor signals can be expressed as follows:
  • x i (k+1) A x i (k)+B 1 ui -1 (k)+B 2 ( ui-1 (k)+ ⁇ i (k)+ ⁇ ui (k)),
  • xi (k) is the state value of the vehicle at time k.
  • Step (2) Determine the model description of the vehicle's state observer.
  • the observer model can be expressed as:
  • the specific design method can refer to the design of Luenberger observer.
  • the residual ri (k) is indirectly affected by the abnormal signal ⁇ i (k-1) of the V2V network.
  • ⁇ i (k) is 0, ei (k) approaches 0, ⁇ i (k) approaches 0, and ri (k+1) also approaches 0.
  • Step (3) Determine the model description of the vehicle's monitor.
  • the quadratic form of the residual signal can be considered and defined as in is a positive semidefinite matrix, consider the following form of monitor:
  • the monitor then outputs an alarm message, that is, the value of the alarm signal changes from 0 to 1.
  • the monitor gain ⁇ ensures that the ellipsoid It contains all the dynamic trajectories of the residual system (6) that may be caused by ⁇ ui (k) and ⁇ i (k).
  • the volume of the ellipsoid can be minimized by adjusting ⁇ , making the detection method more sensitive to abnormal signals.
  • the covert FDI attack signal needs to satisfy the following mathematical expression:
  • e i (k) is a bounded signal, That is, ⁇ i (k) can be represented by a series of bounded signals.
  • the first reachable set can be understood as: for the existing design of the vehicle system, that is, the gains of the observer, detector and controller are known, under the influence of the covert FDI attack signal and/or the system disturbance signal, all possible sets of the vehicle's dynamic state that can be achieved.
  • ⁇ i (k) satisfies
  • all possible sets of x i (k) are is the first reachable set at time k.
  • the system disturbance signal may include at least one of the following: an estimation error of an observer, a residual signal, sensor noise (such as radar noise, speed sensor noise, etc.), network communication noise, and disturbance caused by an attack signal.
  • the attack signal includes a fault or random FDI attack signal, and a covert FDI attack signal.
  • Solving the first reachable set may include the following steps (1) to (2):
  • Step (1) Solve The ellipsoid approximation set.
  • Step (2) Solve The ellipsoid approximation set.
  • the matrix P can be decomposed as follows:
  • the relationship between a reachable set and its outer ellipsoid approximation may be as shown in FIG. 4 .
  • the approximate volume of The volume of can be quantitatively characterized by -log(det(P)). Based on this approximate volume as an indicator for assessing vehicle risk, the impact of covert FDI attack signals on vehicle dynamics is determined.
  • the danger set of a vehicle is defined as a set of dynamic states of the vehicle when the vehicle is in violation of traffic rules or danger, such as speeding, or colliding with the vehicle in front, which can be expressed as v i (k)>v max or d i (k) ⁇ 0.
  • v max can be the maximum speed that the i-th vehicle can reach, or can also be the maximum speed limit of the road on which the i-th vehicle is traveling, or can also be the maximum speed allowed to ensure the safety of the vehicle formation during driving, or can also be other speed thresholds, which are not specifically limited in the embodiments of the present application.
  • the specific value of s i varies with the model of the vehicle.
  • s i can be 4.92 feet, or can also be any value between 4.5 and 5.5 feet; for a truck, the value can be greater than 5.5 feet.
  • the distance between the two vehicles may be the distance between a rear bumper of the front vehicle and a front bumper of the rear vehicle.
  • S303 Determine the impact of the covert FDI attack signal on the vehicle according to the distance between the first reachable set and the dangerous set.
  • the distance between the first reachable set and the dangerous set can be determined according to the following formula:
  • the vehicle will not exceed the speed limit or collide under the covert attack, that is, the covert FDI attack signal will not pose a threat to vehicle safety.
  • the vehicle may be in danger of speeding or collision under the covert attack, that is, the covert FDI attack signal may threaten the safety of the vehicle.
  • the covert FDI attack signal will cause the vehicle to be in danger; in the case shown in Figure 5 (c), although However, since the danger set does not intersect with the first reachable set, the vehicle will not be in danger under the influence of the covert FDI attack signal.
  • the method for assessing vehicle risk provided in the embodiment of the present application when applied to vehicle control in automated platooning, when solving the ellipsoid approximation set (or optimization problem), it is also necessary to introduce estimation error, vehicle tracking error system stability, and vehicle queue string stability as constraints. Alternatively, the false alarm rate of the detector can also be introduced as a constraint.
  • the string stability of the vehicle queue can ensure that small disturbances in the vehicle will not be amplified along the vehicle queue, thereby maintaining safety. For example, the string stability of the vehicle queue ensures that the sudden braking of the leading vehicle will not cause a collision with its followers.
  • the first reachable set can also be determined by a Monte Carlo simulation method.
  • a Monte Carlo simulation method Exemplarily, in each simulation, an initial vehicle condition is given, and then a state trajectory can be obtained by adding a covert FDI attack signal and using model iteration.
  • a state trajectory can be obtained by adding a covert FDI attack signal and using model iteration.
  • multiple state trajectories can be obtained, which can be used as a rough approximation of the reachable set.
  • the approximate set of the reachable set can also be determined by other methods other than the ellipsoid approximation.
  • the approximate set of the reachable set can be determined by using a polyhedron approximation or a Zeno polyhedron (zonotopes) approximation; or the reachability analysis can also be performed by a machine learning method, which is not specifically limited in the embodiments of the present application.
  • the embodiment of the present application provides a method for evaluating vehicle risk.
  • risk evaluation indicators the volume of a reachable set and the distance between a reachable set and a dangerous set
  • the impact of covert FDI attacks on vehicle dynamics can be evaluated, which helps guide the redesign of vehicle systems.
  • the system disturbance signal is taken into account, so that the vehicle model is closer to the actual vehicle, and the accuracy of the evaluation results is improved.
  • the existing designs of vehicle systems mainly consider the performance of the vehicle in the absence of abnormalities.
  • the Kalman filter can achieve the best estimation of the vehicle state in the presence of Gaussian white noise, and the vehicle controller needs to meet the best basic performance of vehicle control.
  • the optimal design corresponding to the performance of the vehicle in the absence of abnormalities cannot guarantee safe vehicle control when the vehicle is attacked by covert FDI.
  • the attack monitoring system may include a detector (including an observer, a monitor), or may also include a controller.
  • the observer is used to obtain first data from a first device of the vehicle at a first moment, and determine the second data of the first device at a second moment, wherein the second moment is a moment after the first moment;
  • the monitor is used to determine the difference between the second data and the third data, wherein the third data is the data obtained from the first device at the second moment; when the difference is greater than or equal to the first threshold, an alarm message is output, and the alarm message is used to indicate that the vehicle is under attack;
  • the controller is used to meet the basic performance of vehicle control, and the controller can input the controller output to the observer.
  • the controller output may include the motion state that the controller expects the vehicle to achieve, such as the expected speed, the expected acceleration, the expected steering wheel angle, etc.
  • the above-mentioned expected motion state can be the actual motion state of the vehicle after being executed by the actuator of the vehicle.
  • the controller output may include the expected acceleration; in other intelligent driving systems, the controller output may also be the expected speed, the expected steering wheel angle, etc.
  • the attack monitoring system at least one of the observer gain, the controller gain, and the monitor gain is determined according to the covert FDI attack signal. After the attack monitoring system is redesigned, the vehicle's ability to resist the covert FDI attack can be improved, that is, the impact of the covert FDI attack signal on vehicle safety can be reduced.
  • the distance between the reachable set and the dangerous set changes from the negative distance shown in (a) of Figure 6 to the positive distance shown in (b) of Figure 6. That is, in the case shown in (b) of Figure 6, the vehicle will not be in danger under the action of the covert FDI attack signal.
  • a system for monitoring attacks provided by an embodiment of the present application, since the system for monitoring attacks is redesigned according to a covert FDI attack signal, the impact of the covert FDI attack signal on vehicles using the system for monitoring attacks can be reduced, which helps to improve the safety of the vehicles; and the performance of the system for monitoring attacks can be guaranteed, for example, the observer can ensure a bounded estimation error; the monitor can monitor the attack signal with an acceptable false alarm rate; and the controller can ensure the dynamic stability of the vehicle tracking error and the string stability of the vehicle queue.
  • the attack monitoring system redesigned according to the covert FDI attack signal can be set in the vehicle when the vehicle leaves the factory.
  • the existing attack monitoring system in the vehicle can be updated by software upgrade, for example, the redesigned attack monitoring system can be updated to the vehicle through over the air (OTA) technology.
  • OTA over the air
  • when updating the vehicle's attack monitoring system only one or more gains determined during the redesign process can be updated.
  • the method for assessing vehicle risk and the system for monitoring attacks provided by the present application are not only applicable to CACC driving scenarios, but also to driving scenarios such as adaptive cruise control (ACC), navigation cruise assistant (NCA) or integrated cruise assistant (ICA).
  • ACC adaptive cruise control
  • NCA navigation cruise assistant
  • ICA integrated cruise assistant
  • the above embodiment is described by taking the covert FDI attack signal related to the V2V communication network as an example.
  • covert FDI attack signal related to sensors such as GNSS, millimeter wave radar, ultrasonic radar, and in-vehicle communication networks, and the impact on the vehicle, as well as the scheme for redesigning the system for monitoring attacks based on the above-mentioned covert FDI attack signal, should also be included in the protection scope of the present application.
  • FIG8 shows a schematic block diagram of a device 2000 for assessing vehicle risk provided in an embodiment of the present application.
  • the device 2000 includes an acquisition unit 2010 and a processing unit 2020 .
  • the apparatus 2000 may include units for executing the method in Fig. 3. Moreover, each unit in the apparatus 2000 and the above-mentioned other operations and/or functions are respectively for implementing the corresponding processes of the method embodiment in Fig. 3.
  • the acquisition unit 2010 may be used to execute S301 in the method 300
  • the processing unit 2020 may be used to execute S302 and S303 in the method 300 .
  • the acquisition unit 2010 is used to acquire a covert false information injection FDI attack signal for a vehicle;
  • the processing unit 2020 is used to determine a first reachable set and a danger set according to the covert FDI attack signal, wherein the first reachable set includes the dynamic state of the vehicle that changes with the covert FDI attack signal, and the danger set includes the dynamic state of the vehicle when a dangerous situation occurs; and according to the distance between the first reachable set and the danger set, determine the influence of the covert FDI attack signal on the vehicle. That is, determine whether the covert FDI attack signal will cause danger to the vehicle.
  • the processing unit 2020 is further used to determine the impact of the covert FDI attack signal on the vehicle based on the volume of the first reachable set.
  • the covert FDI attack signal is determined based on at least one of the following: a monitor residual signal, an estimation error of an observer, and sensor noise; wherein the observer is used to determine an estimation value of the sensor at the next time step based on a measurement value of the sensor; and the monitor is used to monitor an attack on the vehicle based on a difference between the measurement value of the sensor and the estimation value of the observer.
  • the covert FDI attack signal is associated with at least one of: a vehicle-to-vehicle (V2V) communication network, a global positioning system (GPS) signal, a millimeter wave radar, an ultrasonic radar, and an in-vehicle communication network.
  • V2V vehicle-to-vehicle
  • GPS global positioning system
  • the processing unit 2020 is used to: determine an approximate set of the first reachable set based on the covert FDI attack signal, wherein the approximate set is determined by solving a convex optimization problem with constraints such as the vehicle queue chord stability of the vehicle formation and/or the estimation error of the observer, wherein the vehicle is in the vehicle formation.
  • the processing unit 2020 is used to: determine the first reachable set through a Monte Carlo simulation method according to the covert FDI attack signal.
  • the dangerous situation includes at least one of the following: the speed of the vehicle is greater than or equal to a speed threshold, the acceleration of the vehicle is greater than or equal to an acceleration threshold, and the distance between the vehicle and the most immediately preceding vehicle is less than or equal to a distance threshold.
  • the processing unit 2020 is also used to: determine the gain of the monitor and/or observer based on the covert FDI attack signal when the distance between the first reachable set and the danger set is less than or equal to a preset threshold; wherein the gain is associated with the second reachable set, and the distance between the second reachable set and the danger value is greater than the preset threshold.
  • the division of the units in the above device is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated.
  • the units in the device can be implemented in the form of a processor calling software; for example, the device includes a processor, the processor is connected to a memory, and instructions are stored in the memory.
  • the processor calls the instructions stored in the memory to implement any of the above methods or realize the functions of the units of the device, wherein the processor is, for example, a general-purpose processor, such as a CPU or a microprocessor, and the memory is a memory in the device or a memory outside the device.
  • the units in the device can be implemented in the form of hardware circuits, and the functions of some or all of the units can be realized by designing the hardware circuits.
  • the hardware circuit can be understood as one or more processors; for example, in one implementation, the hardware circuit is an ASIC, and the functions of some or all of the above units are realized by designing the logical relationship of the components in the circuit; for another example, in another implementation, the hardware circuit can be realized by PLD.
  • FPGA as an example, it can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured through the configuration file, so as to realize the functions of some or all of the above units. All units of the above device may be implemented entirely in the form of a processor calling software, or entirely in the form of a hardware circuit, or partially in the form of a processor calling software and the rest in the form of a hardware circuit.
  • a processor is a circuit with the ability to process signals.
  • the processor may be a circuit with the ability to read and run instructions, such as a CPU, a microprocessor, a GPU, or a DSP; in another implementation, the processor may implement certain functions through the logical relationship of a hardware circuit, and the logical relationship of the hardware circuit is fixed or reconfigurable, such as a hardware circuit implemented by an ASIC or PLD, such as an FPGA.
  • the process of the processor loading a configuration document to implement the configuration of the hardware circuit can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units.
  • it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as an NPU, TPU, DPU, etc.
  • each unit in the above device can be one or more processors (or processing circuits) configured to implement the above method, such as: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.
  • processors or processing circuits
  • the units in the above device can be fully or partially integrated together, or can be implemented independently. In one implementation, these units are integrated together and implemented in the form of a system-on-a-chip (SOC).
  • SOC may include at least one processor for implementing any of the above methods or implementing the functions of each unit of the device.
  • the type of the at least one processor may be different, for example, including a CPU and an FPGA, a CPU and an artificial intelligence processor, a CPU and a GPU, etc.
  • the operations performed by the acquisition unit 2010 and the processing unit 2020 may be performed by the same processor, or may be performed by different processors, for example, respectively by multiple processors.
  • An embodiment of the present application also provides a device, which includes a processing unit and a storage unit, wherein the storage unit is used to store instructions, and the processing unit executes the instructions stored in the storage unit so that the device executes the method or steps executed by the above embodiment.
  • FIG9 is a schematic block diagram of a device for assessing vehicle risk according to an embodiment of the present application.
  • the device 2100 for assessing vehicle risk shown in FIG9 may include: a processor 2110, a transceiver 2120, and a memory 2130.
  • the processor 2110, the transceiver 2120, and the memory 2130 are connected via an internal connection path, the memory 2130 is used to store instructions, the processor 2110 is used to execute the instructions stored in the memory 2130, and the transceiver 2120 receives/sends some parameters.
  • the memory 2130 may be coupled to the processor 2110 via an interface, or may be integrated with the processor 2110.
  • transceiver 2120 may include but is not limited to a transceiver device such as an input/output interface to achieve communication between the device 2100 and other devices or communication networks.
  • the processor 2110 may be a general-purpose CPU, microprocessor, ASIC, GPU or one or more integrated circuits for executing relevant programs to implement the method for assessing vehicle risk in the embodiment of the method of the present application.
  • the processor 2110 may also be an integrated circuit chip with signal processing capabilities.
  • the various steps of the method for assessing vehicle risk in the present application may be completed by hardware integrated logic circuits or software instructions in the processor 2110.
  • the above-mentioned processor 2110 may also be a general-purpose processor, DSP, ASIC, FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
  • the various methods, steps and logic block diagrams disclosed in the embodiments of the present application may be implemented or executed.
  • the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
  • the steps of the method disclosed in the embodiment of the present application may be directly embodied as being executed by a hardware decoding processor, or may be executed by a combination of hardware and software modules in a decoding processor.
  • the software module may be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc.
  • the storage medium is located in the memory 2130, and the processor 2110 reads the information in the memory 2130 and executes the method for assessing vehicle risk in the method embodiment of the present application in combination with its hardware.
  • Memory 2130 can be a read-only memory (ROM), a static storage device, a dynamic storage device or a random access memory (RAM).
  • ROM read-only memory
  • RAM random access memory
  • the transceiver 2120 uses a transceiver device such as, but not limited to, a transceiver to implement communication between the device 2100 and other devices or a communication network.
  • a transceiver device such as, but not limited to, a transceiver to implement communication between the device 2100 and other devices or a communication network.
  • the user location information can be obtained through the transceiver 2120.
  • An embodiment of the present application further provides a server, which may include the above-mentioned device 2000 or the above-mentioned device 2100.
  • the embodiment of the present application further provides a computer program product, which includes: a computer program code, when the computer program code is executed on a computer, the computer executes the method in the above embodiment.
  • An embodiment of the present application also provides a computer-readable storage medium, which stores program code or instructions.
  • the processor implements the method in the above embodiment.
  • An embodiment of the present application also provides a chip, including: at least one processor and a memory, wherein the at least one processor is coupled to the memory and is used to read and execute instructions in the memory to execute the method in the above embodiment.
  • references to "one embodiment” or “some embodiments” etc. described in this specification mean that a particular feature, structure or characteristic described in conjunction with the embodiment is included in one or more embodiments of the present application.
  • the phrases “in one embodiment”, “in some embodiments”, “in some other embodiments”, “in some other embodiments”, etc. appearing in different places in this specification do not necessarily all refer to the same embodiment, but mean “one or more but not all embodiments", unless otherwise specifically emphasized in other ways.
  • the terms “including”, “comprising”, “having” and their variations all mean “including but not limited to”, unless otherwise specifically emphasized in other ways.
  • At least one means one or more
  • plural means two or more.
  • “And/or” describes the association relationship of associated objects, indicating that three relationships may exist.
  • a and/or B can mean: including the existence of A alone, the existence of A and B at the same time, and the existence of B alone, where A and B can be singular or plural.
  • the character “/” generally indicates that the previous and next associated objects are in an “or” relationship.
  • “At least one of the following” or similar expressions refers to any combination of these items, including any combination of single or plural items.
  • At least one of a, b, or c can mean: a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, c can be single or multiple.
  • the disclosed systems, devices and methods can be implemented in other ways.
  • the device embodiments described above are only schematic.
  • the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.
  • Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.
  • the technical solution of the present application can essentially be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application.
  • the aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard drives, ROM, RAM, magnetic disks, or optical disks.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Educational Administration (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种评估车辆风险的方法、装置以及监测攻击的系统，该方法包括：获取针对车辆的隐秘虚假信息注入FDI攻击信号；根据该隐秘FDI攻击信号确定第一可达集，以及危险集，该第一可达集包括随该隐秘FDI攻击信号变化的该车辆的动力学状态，该危险集包括发生危险情况时该车辆的动力学状态；根据该第一可达集与该危险集之间的距离，确定该隐秘FDI攻击信号对该车辆的影响。本申请的方法，可以应用于智能车辆、新能源车辆中，能够评估隐秘FDI攻击信号对车辆安全的影响，以指导车辆的检测攻击的系统的设计，进而提高车辆安全，降低车辆风险。

Description

评估车辆风险的方法、装置以及监测攻击的系统 技术领域
本申请涉及车辆安全领域，更具体地，涉及一种评估车辆风险的方法、装置以及监测攻击的系统。
背景技术
协同自适应巡航控制(cooperative and adaptive cruise control，CACC)可以用作车辆自动跟随系统，基于车对车(vehicle to vehicle，V2V)无线通信共享车辆的本地传感器信息，能够使网联自动驾驶车辆(connected and automated vehicle，CAV)以紧密耦合的形式在路上行驶。然而，CAV的传感器以及通信系统容易受到外部攻击。例如，车辆雷达、激光雷达、摄像头和全球导航卫星系统(global navigation satellite system，GNSS)可能容易受到各种类型的外部攻击，车内网络和车外网络都可以遭受多种类型的攻击，这使得虚假信息在多个CAV之间传播，可能导致交通事故，甚至在最坏的情况下造成人员伤亡。
鉴于此，一种能够提升车辆安全的风险评估以及监测攻击的方案亟待开发。
发明内容
本申请实施例提供了一种评估车辆风险的方法、装置以及监测攻击的系统，该评估车辆风险的方法能够评估隐秘虚假信息注入(false data injection，FDI)攻击对车辆的影响，该监测攻击的系统能够提高车辆抵抗隐秘FDI攻击的能力。
本申请提供的方法可以应用于车辆中，该车辆为广义概念上的车辆，可以是交通工具(如汽车商用车、乘用车、卡车、摩托车、飞机飞行车、火车、轮船等)，工业车辆(如：叉车、挂车、牵引车等)，工程车辆(如挖掘机、推土车、吊车等)，农用设备(如割草机、收割机等)，游乐设备，玩具车辆等，本申请实施例对车辆的类型不作具体限定。
第一方面，提供了一种评估车辆风险的方法，该方法包括：获取针对车辆的隐秘FDI攻击信号；根据该隐秘FDI攻击信号确定第一可达集，以及危险集，该第一可达集包括随该隐秘FDI攻击信号变化的该车辆的动力学状态，该危险集包括发生危险情况时该车辆的动力学状态；根据该第一可达集与该危险集之间的距离，确定该隐秘FDI攻击信号对该车辆的影响。
示例性地，第一设备可以为车辆的传感器，例如可以包括惯性测量单元(inertial measurement unit，IMU)、激光雷达、毫米波雷达、超声雷达中的至少一个，或者还可以为其他传感器，本申请对此不作具体限定。
示例性地，隐秘FDI攻击信号可以包括检测器(或监测器)无法监测到的攻击信号。例如，隐秘FDI攻击信号可以为检测器允许的误差范围内的攻击信号，如隐藏在车辆的系统干扰中的攻击信号；或者，隐秘FDI攻击信号还可以为检测器完全不监测的攻击信号，如未设定检测器对全球定位系统(global positioning system，GPS)信号的异常监测，则所 有的GPS欺骗信号均可以视为隐秘FDI攻击信号。
在一些可能的实现方式中，获取的针对车辆的隐秘FDI攻击信号为数学表达式形式，或者也可以为其他形式。
在一些可能的实现方式中，该第一可达集包括了该车辆随该隐秘FDI攻击信号变化的所有动力学状态。其中，“所有动力学状态”可以包括在任何一种可能的隐秘FDI攻击信号和/或系统扰动信号的作用下，车辆的动力学状态。
在一些可能的实现方式中，在第一可达集与该危险集之间的距离小于或等于零时，代表隐秘FDI攻击信号容易影响车辆安全，例如，可能造成车辆超速或偏航等；在第一可达集与该危险集之间的距离大于零时，代表隐秘FDI攻击信号不会导致车辆发生危险。
在上述技术方案中，通过确定车辆的可达集和危险集，能够评估隐秘FDI攻击信号是否会导致车辆发生危险，以指导车辆的监测攻击的系统(以下有时简称车辆系统)的重设计，进而提高车辆安全。
结合第一方面，在第一方面的某些实现方式中，该方法还包括：根据该第一可达集的体积量化评估该隐秘FDI攻击信号对该车辆的影响。
在一些可能的实现方式中，可以根据第一可达集的近似集确定第一可达集的体积；或者，还可以通过模拟方法(例如蒙特卡洛模拟)确定第一可达集进而确定第一可达集的体积。
在上述技术方案中，可以通过第一可达集的体积，评估隐秘FDI攻击信号对已有车辆系统的影响程度，第一可达集的体积越大，则隐秘攻击信号对车辆动力学状态的扰动作用越大，代表隐秘FDI攻击信号对车辆系统的影响越大。
结合第一方面，在第一方面的某些实现方式中，该隐秘FDI攻击信号为根据如下至少一个确定：监测器的残差信号，观测器的估计误差，以及传感器噪声；其中，该观测器用于根据该传感器的测量值，确定该传感器在下一时刻的估计值；该监测器用于根据该传感器的测量值与该观测器估计值之间的差值，对该车辆的攻击进行监测。
在一些可能的实现方式中，该传感器的测量值与该观测器估计值之间的差值小于监测器的阈值时，监测器即不会发出警报信息。则使得传感器的测量值与该观测器估计值之间的差值小于该监测器的阈值的攻击信号，均可被认定为隐秘FDI攻击信号。
结合第一方面，在第一方面的某些实现方式中，该隐秘FDI攻击信号与如下至少一项相关联：车对车V2V通信网络，全球导航卫星系统GNSS，毫米波雷达，超声波雷达，以及车内通信网络。
其中，“相关联”可以理解为：隐秘FDI攻击信号隐藏在V2V通信网络信号的噪声，或者GPS信号的噪声中，或者毫米波雷达信号的噪声中，或者超声波雷达信号的噪声中传输给车辆。或者，“相关联”也可以理解为：该隐秘FDI攻击信号为针对V2V通信网络，GNSS，毫米波雷达，超声波雷达，以及车内通信网络中至少一个的攻击信号。
在上述技术方案中，可以评估多种情景下隐秘FDI攻击信号对已有车辆系统的影响程度，有助于指导车辆的监测攻击的系统的重设计，进而提高车辆安全。
结合第一方面，在第一方面的某些实现方式中，根据该隐秘FDI攻击信号确定第一可达集，包括：根据该隐秘FDI攻击信号确定第一可达集的近似集，该近似集是以车辆编队的车辆队列弦稳定性，和/或观测器的估计误差为约束求解最优化问题确定，其中，该车 辆处于该车辆编队。
结合第一方面，在第一方面的某些实现方式中，根据该隐秘FDI攻击信号确定第一可达集，包括：根据该隐秘FDI攻击信号通过蒙特卡洛模拟方法确定该第一可达集。
在上述技术方案中，通过模拟的方法确定第一可达集，可以降低确定第一可达集过程的计算复杂度。
结合第一方面，在第一方面的某些实现方式中，该危险情况包括如下至少一项：该车辆的速度大于或等于速度阈值，该车辆的加速度大于或等于加速度阈值，该车辆与最紧邻前车之间的距离小于或等于距离阈值。
示例性地，在不同路段的速度阈值可能为该路段允许的最高时速；加速度阈值可根据驾驶员和乘客的舒适度和安全性确定；该距离阈值可设定为0(小于或等于0时发生碰撞)，或者，该距离阈值也可以设定为更高的数值以排除安全隐患。
示例性地，车辆与最紧邻前车之间的距离可以为最紧邻前车的后保险杠与车辆的前保险杠之间的距离。
结合第一方面，在第一方面的某些实现方式中，在该第一可达集与该危险集之间的距离小于或等于预设阈值时，根据该隐秘FDI攻击信号确定监测器和/或观测器的增益；其中，该增益与第二可达集相关联，该第二可达集与该危险值之间的距离大于该预设阈值。
示例性地，该预设阈值可以为上述预设阈值，即为0，或者也可以为其他数值，本申请实施例对此不作具体限定。
应理解，该第二可达集与该危险值之间的距离大于该预设阈值，代表隐秘FDI攻击信号不会威胁车辆安全，例如，不会造成车辆超速或偏航等。
其中，“该增益与第二可达集相对应”可以理解为通过隐秘FDI攻击信号确定该增益后，该车辆的动力学状态随该隐秘FDI攻击信号变化形成的集合为第二可达集。
在上述技术方案中，可以根据该隐秘FDI攻击信号重新确定监测器和/或观测器的增益来达到减小隐秘攻击信号对车辆动态影响的目的；其中，该第二可达集有更小的体积，并且该第二可达集与该危险值之间的距离大于该预设阈值。这意味着重新确定的监测器和/或观测器的增益降低了隐秘攻击信号对车辆安全的影响，即，车辆风险更低。
第二方面，提供了一种评估车辆风险的装置，该装置包括：获取单元，用于获取针对车辆的隐秘虚假信息注入FDI攻击信号；处理单元，用于根据该隐秘FDI攻击信号确定第一可达集，以及危险集，该第一可达集包括随该隐秘FDI攻击信号变化的该车辆的动力学状态，该危险集包括发生危险情况时该车辆的动力学状态；根据该第一可达集与该危险集之间的距离，确定该隐秘FDI攻击信号对该车辆的影响。
结合第二方面，在第二方面的某些实现方式中，该处理单元还用于：根据该第一可达集的体积确定该隐秘FDI攻击信号对该车辆的影响。
结合第二方面，在第二方面的某些实现方式中，该隐秘FDI攻击信号为根据如下至少一个确定：监测器的阈值，观测器的估计误差，传感器噪声，以及通信网络噪声；其中，该观测器用于根据该传感器的测量值，确定该传感器在下一时间步的估计值；该监测器用于根据该传感器的测量值与该观测器估计值之间的差值，对该车辆的攻击进行监测。
结合第二方面，在第二方面的某些实现方式中，该隐秘FDI攻击信号与如下至少一项相关联：车对车V2V通信网络，全球导航卫星系统GNSS，毫米波雷达，超声波雷达， 以及车内通信网络。
结合第二方面，在第二方面的某些实现方式中，该处理单元用于：根据该隐秘FDI攻击信号确定第一可达集的近似集，该近似集是以车辆编队的车辆队列稳定性，和/或观测器的估计误差为约束求解凸优化问题确定，其中，该车辆处于该车辆编队。
结合第二方面，在第二方面的某些实现方式中，该处理单元用于：根据该隐秘FDI攻击信号通过蒙特卡洛模拟方法确定该第一可达集。
结合第二方面，在第二方面的某些实现方式中，该危险情况包括如下至少一项：该车辆的速度大于或等于速度阈值，该车辆的加速度大于或等于加速度阈值，该车辆与其他车辆的相对距离低于相对距离阈值
结合第二方面，在第二方面的某些实现方式中，该处理单元还用于：在该第一可达集与该危险集之间的距离小于或等于预设阈值时，根据该隐秘FDI攻击信号确定监测器和/或观测器的增益；其中，该增益与第二可达集相关联，该第二可达集与该危险值之间的距离大于该预设阈值。
第三方面，提供了一种监测攻击的系统，本申请提供的监测攻击的系统可以设置于车辆的计算平台中。示例性地，该计算平台可以包括：高级驾驶辅助系统(advanced driving assistance system，ADAS)、整车控制器(vehicle control unit，VCU)、车载信息娱乐系统(in-vehicle infotainment systems,IVI)中的至少一个；或者还可以包括其他计算平台，例如车载应用服务(in-car application-server，ICAS)控制器，车身控制器(body domain controller，BDC)，特殊装备系统(special equipment system，SAS)，媒体图形单元(media graphics unit，MGU)，车身超级核心(body super core，BSC)，ADAS超级核心(ADAS super core)等，本申请对此不做限定。其中，ICAS可以包括如下至少一项：车辆控制服务器ICAS1、智能驾驶服务器ICAS2、智能座舱服务器ICAS3、信息娱乐服务器ICAS4。
该系统包括：观测器，用于根据在第一时刻从车辆的第一设备获取第一数据，确定该第一设备在第二时刻的第二数据，其中，该第二时刻为该第一时刻之后的时刻；监测器，用于确定该第二数据和第三数据之间的差值，该第三数据为在该第二时刻从该第一设备获取的数据；在该差值大于或等于第一阈值时，输出警报信息，该警报信息用于指示该车辆遭受攻击；其中，该监测器的增益为根据隐秘虚假信息注入FDI攻击信号确定。
在上述技术方案中，根据隐秘FDI攻击信号，对车辆的监测攻击的系统进行增益更新，有助于提高车辆抵抗隐秘FDI攻击的能力，进而提高车辆安全，降低车辆风险。
结合第三方面，在第三方面的某些实现方式中，该观测器的增益为根据该隐秘FDI攻击信号确定。
结合第三方面，在第三方面的某些实现方式中，该系统还包括控制器，用于实现车辆控制基本性能，该控制器的增益为根据该隐秘FDI攻击信号确定。
示例性地，车辆控制基本性能包括车辆跟踪误差系统的稳定性，车辆跟踪误差系统符合要求的收敛速度，车辆列队的弦稳定性等。
在一些可能的实现方式中，该控制器可以通过向车辆的执行器发送期望车辆的达到的运动状态实现对车辆运动状态的控制，其中，期望车辆达到的运动状态可以包括期望速度、期望加速度、期望方向盘转角等。该控制器在满足车辆控制基本性能要求以外，还需保证车辆的低风险。
结合第三方面，在第三方面的某些实现方式中，该增益为根据该车辆的第一可达集和危险集确定；其中，该第一可达集为根据该观测器、该监测器和该控制器的初始增益确定，该第一可达集包括随该隐秘FDI攻击信号变化的该车辆的动力学状态，该危险集包括发生危险情况时该车辆的动力学状态，其中，该第一可达集和该危险集之间的距离小于或等于第二阈值。
示例性地，该第二阈值可以为0，或者也可以为其他数值，本申请实施例对此不作具体限定。
结合第三方面，在第三方面的某些实现方式中，第二可达集和该危险集之间的距离大于该第二阈值，该第二可达集与该增益相对应。
其中，“该第二可达集与该增益相对应”可以理解为通过隐秘FDI攻击信号确定该增益后，该车辆的所有动力学状态随该隐秘FDI攻击信号变化形成的集合为第二可达集。
结合第三方面，在第三方面的某些实现方式中，该第二可达集的体积小于该第一可达集的体积。
第四方面，提供了一种评估车辆风险的装置，该装置包括处理单元和存储单元，其中存储单元用于存储指令，处理单元执行存储单元所存储的指令，以使该装置执行第一方面中任一种可能实现方式中的方法。
可选地，上述处理单元可以包括至少一个处理器，上述存储单元可以是存储器，其中存储器可以是芯片内的存储单元(例如，寄存器、缓存等)，也可以是移动载体内位于芯片外部的存储单元(例如，只读存储器、随机存取存储器等)。
第五方面，提供了一种车辆，该车辆包括上述第三方面中任一种实现方式中的系统。
第六方面，提供了一种服务器，该服务器包括上述第二方面或第四方面中任一种实现方式中的装置。
第七方面，提供了一种计算机程序产品，上述计算机程序产品包括：计算机程序代码，当上述计算机程序代码在计算机上运行时，使得计算机执行上述第一方面中任一种可能实现方式中的方法。
需要说明的是，上述计算机程序代码可以全部或部分存储在第一存储介质上，其中第一存储介质可以与处理器封装在一起的，也可以与处理器单独封装，本申请实施例对此不作具体限定。
第八方面，提供了一种计算机可读介质，上述计算机可读介质存储有指令，当上述指令被处理器执行时，使得处理器实现上述第一方面中任一种可能实现方式中的方法。
第九方面，提供了一种芯片，该芯片包括处理器，用于调用存储器中存储的计算机程序或计算机指令，以使得该处理器执行上述第一方面中任一种可能实现方式中的方法。
结合第九方面，在一种可能的实现方式中，该处理器通过接口与存储器耦合。
结合第九方面，在一种可能的实现方式中，该芯片系统还包括存储器，该存储器中存储有计算机程序或计算机指令。
附图说明
图1是本申请实施例提供的车辆的功能性框图示意；
图2是本申请实施例提供的一种基于残差的检测器的示意图；
图3是本申请实施例提供的一种评估车辆风险的方法的示例性流程图；
图4是本申请实施例提供的一种可达集及其近似集的示意图；
图5是本申请实施例提供的一种可达集与危险集之间的关系的示意图；
图6是本申请实施例提供的调整增益后可达集与危险集之间的关系的示意图；
图7是本申请实施例提供的一种监测攻击的系统的示意性框图；
图8是本申请实施例提供的一种评估车辆风险的装置的示意性框图；
图9是本申请实施例提供的一种评估车辆风险的装置的示意性框图。
具体实施方式
下面将结合附图，对本申请实施例中的技术方案进行描述。
图1是本申请实施例提供的车辆100的一个功能框图示意。车辆100可以包括感知系统120和计算平台150，其中，感知系统120可以包括感测关于车辆100周边的环境的信息的若干种传感器。例如，感知系统120可以包括定位系统，定位系统可以是GPS，也可以是北斗系统或者其他定位系统、IMU、激光雷达、毫米波雷达、超声雷达以及摄像装置中的一种或者多种。
车辆100的部分或所有功能可以由计算平台150控制。计算平台150可包括处理器151至15n(n为正整数)，处理器是一种具有信号的处理能力的电路，在一种实现中，处理器可以是具有指令读取与运行能力的电路，例如中央处理单元(central processing unit，CPU)、微处理器、图形处理器(graphics processing unit，GPU)(可以理解为一种微处理器)、或数字信号处理器(digital signal processor，DSP)等；在另一种实现中，处理器可以通过硬件电路的逻辑关系实现一定功能，该硬件电路的逻辑关系是固定的或可以重构的，例如处理器为专用集成电路(application-specific integrated circuit，ASIC)或可编程逻辑器件(programmable logic device，PLD)实现的硬件电路，例如现场可编辑逻辑门阵列(filed programmable gate array，FPGA)。在可重构的硬件电路中，处理器加载配置文档，实现硬件电路配置的过程，可以理解为处理器加载指令，以实现以上部分或全部单元的功能的过程。此外，还可以是针对人工智能设计的硬件电路，其可以理解为一种ASIC，例如神经网络处理单元(neural network processing unit，NPU)、张量处理单元(tensor processing unit，TPU)、深度学习处理单元(deep learning processing unit，DPU)等。此外，计算平台150还可以包括存储器，存储器用于存储指令，处理器151至15n中的部分或全部处理器可以调用存储器中的指令，执行指令，以实现相应的功能。
在一些可能的实现方式中，车辆100可以包括高级驾驶辅助系统(advanced driving assistant system，ADAS)，ADAS利用感知系统120中的一种或多种传感器(包括但不限于：激光雷达、毫米波雷达、摄像装置、超声波传感器、全球定位系统、惯性测量单元)从车辆周围获取信息，并对获取的信息进行分析和处理，实现例如障碍物感知、目标识别、车辆定位、车辆自动跟随、路径规划、驾驶员监控/提醒等功能，从而提升车辆驾驶的安全性、自动化程度和舒适度。
如上所述，CACC可以用作车辆自动跟随系统，基于V2V无线通信共享车辆的本地传感器信息，能够使车辆以紧密耦合的形式在路上行驶。然而，车辆的传感器以及通信系统容易受到外部攻击。
新型基于残差的检测器可用于识别车辆的传感器故障或通信系统遭受的外部攻击。在一些可能情况下，在通信系统遭受外部攻击时，传感器信号可能遭受FDI攻击，继而影响车辆安全。如图2中的(a)所示，在使用新型基于残差的检测器进行故障或随机FDI攻击检测时，首先使用状态观测器(estimator)，根据未遭受攻击或未出现故障的传感器测量值和车辆数学模型预测下一时间步的传感器测量值(下文称为估计值)。然后在下一时间步，计算传感器的测量值和状态观测器(以下简称观测器)提供的估计值之间的差值作为残差。如果传感器未遭受攻击或没有故障，则该残差将小于或等于预设阈值。在残差大于预设阈值时，认为传感器遭受了攻击或出现故障，监测器(monitor)将输出警报信息。当触发监测器输出警报信息时，车辆可采取必要的响应措施降低攻击进一步的影响，例如关闭发动机以防止造成超速，甚至与前车相撞的事故发生。
在本申请中，将检测器(或监测器)无法监测到的攻击信号称为隐秘(stealth或stealthy)FDI攻击信号，由隐秘FDI攻击信号引起的攻击称为隐秘FDI攻击。示例性地，隐秘FDI攻击信号可以为检测器允许的误差范围内的攻击信号，如隐藏在车辆的系统干扰中的攻击信号。如图2中的(b)所示，隐秘FDI攻击信号不会触发监测器输出警报信息，影响车辆行车安全，进而影响自动化列队行驶(platooning)的过程。例如，可能造成车辆超速或车辆碰撞。应理解，本申请实施例中，自动化列队行驶是指多辆车基于自动驾驶技术和V2V技术的支持，以极小的车距尾随行驶的编队状态，该编队行驶的多辆车被称为车辆编队(platoon)。
示例性地，车辆的系统干扰可以包括如下三类：传感器噪声、网络通信噪声和车辆建模不确定性。其中，传感器噪声可以理解为传感器输出的信号中的无用信号，其形成因素可以包括外界因素和内部因素。示例性地，外界因素包括传感器电路外的人为或自然干扰等因素，例如电磁辐射；内部因素包括传感器的测量误差，内部导电微粒不连续、半导体PN结(p-n junction)的电容效应、高温导致的导电体内部电子的无规则运动等。网络通信噪声可以包括：网络通讯中的时延，丢包，量化等带来的网络噪声，信道固有的、持续存在的随机热噪声，以及外界因素导致的冲击噪声。车辆建模不确定性可以包括：车辆本身固有的一些无法预知的车辆动态。
鉴于此，本申请实施例提供一种评估车辆风险的方法和装置，通过量化隐秘FDI攻击信号对车辆动力学的影响，能够评估车辆遭受隐秘FDI攻击时的风险程度，在检测到车辆存在风险时，可以根据隐秘FDI攻击信号重新设计检测攻击信号的系统，通过该系统监测攻击，可以在车辆系统发生故障或者随机FDI攻击时及时发现并触发响应措施，在此基础上，该监测系统还能够降低隐秘FDI攻击信号对车辆动力学的影响，提升行车安全性。
本申请实施例涉及的隐秘FDI攻击的类型，包括但不限于以下几种：
1、隐秘GPS攻击：小幅度调整GPS位置，使调整GPS位置后生成的虚假导航路线与实际道路的形状相匹配，并触发物理指令，以使车辆到达虚假导航路线所指示的错误的目的地。由于GPS欺骗信号的幅度较小，小于GPS漂移带来的影响，无法触发监测器输出警报信息，因此该隐秘攻击可以绕过监测器，即不被监测器检测到。
2、针对雷达的隐秘攻击：通过将对雷达的攻击信号隐藏在车辆的系统干扰中，在攻击信号无法触发监测器输出警报信息时，该攻击可以绕过监测器，这种攻击即为一种隐秘FDI攻击。
3、针对速度计、加速器的隐秘攻击：车辆的速度计、加速器可能受到电磁信号或物理攻击，当这些攻击信号隐藏在车辆的系统干扰中，无法触发监测器输出警报信息时，该攻击可以绕过监测器，这种攻击即为一种隐秘FDI攻击。
4、V2V信道上的隐秘攻击：假设对于车辆编队中的每个车辆，用于检测攻击的检测器中，观测器和控制器被设计为在不考虑安全性的情况下实现最佳估计和控制性能。在这种情况下，如果一个车辆编队中的一辆车在将传感器测量值发送给该车辆编队中的另一辆车时，可以通过向其传感器测量值中注入虚假数据，以实现对该另一辆车的攻击。若注入的虚假数据无法触发监测器输出警报信息，则该攻击可以绕过监测器，这种攻击即为一种隐秘FDI攻击。
5、车内通信网络隐秘攻击：传感器信号在车内网络传输过程被更改，信号更改的幅度无法触发监测器输出警报信息，则该攻击可以绕过监测器，这种攻击即为一种隐秘FDI攻击。
图3示出了本申请实施例提供的一种评估车辆风险的方法300的示意性流程图，通过该方法300设计的监测攻击的系统可以应用于图1所示的车辆100中，该方法300可以包括：
S301，获取针对车辆的隐秘FDI攻击信号。
示例性地，该隐秘FDI攻击信号可以包括如下至少一种：通过V2V通信网络引入的隐秘FDI攻击信号，通过GPS信号引入的隐秘FDI攻击信号，通过毫米波雷达或超声波雷达引入的隐秘FDI攻击信号，以及通过车内通信网络引入的隐秘FDI攻击信号。
以下以通过V2V通信网络引入的隐秘FDI攻击信号为例说明隐秘FDI攻击信号的确定方法。
为了确定针对车辆的隐秘FDI攻击信号，首先需要确定该车辆的车辆运动状态模型，观测器模型以及检测器模型。示例性地，可以包括如下步骤(1)至(3)：
步骤(1)：确定该车辆的车辆运动状态模型。
以该车辆为基于CACC的车辆编队(platoon)中的第i辆车为例，可以对车辆纵向运动状态构建以下系统模型：
Figure PCTCN2022122163-appb-000001
其中，x i＝[ξ iv ia iu iΔv ia i-1] T,([·] T表示矩阵的转置)，ξ i为跟踪误差，v i为第i辆车的实际速度，a i和a i-1分别为第i和第i-1辆车的实际加速度，Δv i是第i和第i-1辆车的相对速度，u i和u i-1分别为第i和第i-1辆车的期望加速度，δ i为V2V网络的异常信号，可能由恶意车辆注入，可能为V2V网络发生异常导致；ω ui为V2V网络中存在的扰动，可 能由网络丢包、时延、量化等因素引起，且满足
Figure PCTCN2022122163-appb-000002
为扰动边界，
Figure PCTCN2022122163-appb-000003
应理解，
Figure PCTCN2022122163-appb-000004
为正实数集合。K＝[k pk d]为车辆的控制器的增益。h为时间间隔常数，τ表示传动系动力学常数，k p和k d也为常数。
假设每辆车有毫米波雷达，加速度传感器，速度传感器，可以直接或间接测量ξ i,v i,a i,u i,Δv i，则车辆的传感器信号可被表示如下：
Figure PCTCN2022122163-appb-000005
其中，ω i为传感器噪声且满足
Figure PCTCN2022122163-appb-000006
由于车辆处于异常状态(如被隐秘攻击)下，接收到的网络信号都是离散的，因此将公式(1)(2)进行离散化，得到如下系统模型：
x i(k+1)＝Ax i(k)+B 1u i-1(k)+B 2(u i-1(k)+δ i(k)+ω ui(k))，
y i(k)＝Cx i(k)+ω i(k)，(3)
其中，
Figure PCTCN2022122163-appb-000007
x i(k)为该车辆在k时刻的状态值。
步骤(2)：确定车辆的状态观测器的模型描述。
该观测器模型可以表达为：
Figure PCTCN2022122163-appb-000008
其中，
Figure PCTCN2022122163-appb-000009
为观测器的增益(
Figure PCTCN2022122163-appb-000010
代表矩阵X为m×n维实矩阵)，
Figure PCTCN2022122163-appb-000011
令观测器的估计误差
Figure PCTCN2022122163-appb-000012
则e i(k)可以由如下式获得：
Figure PCTCN2022122163-appb-000013
其中，
Figure PCTCN2022122163-appb-000014
通过求解线性矩阵不等式(linear matrix inequality，LMI)，可以确定观测器增益L，使得当V2V网络没有异常，即δ i(k)＝0时，满足
Figure PCTCN2022122163-appb-000015
趋近于x i(k)，即估计误差e i(k)趋近于0，具体设计方法可以参考Luenberger观测器的设计。
令
Figure PCTCN2022122163-appb-000016
则可以得到如下公式：
Figure PCTCN2022122163-appb-000017
可以看出，残差r i(k)间接受到V2V网络异常信号δ i(k-1)的影响。当V2V网络没有异常，即δ i(k)为0时，e i(k)趋近于0，ω i(k)趋近于0，则r i(k+1)也趋近于0。
步骤(3)：确定车辆的监测器的模型描述。
为了监测车辆是否有异常，例如是否受到随机FDI攻击，可以考虑残差信号的二次型，定义
Figure PCTCN2022122163-appb-000018
其中
Figure PCTCN2022122163-appb-000019
为半正定矩阵(positive semidefinitematrix)，考虑如下形式的监测器：
如果
Figure PCTCN2022122163-appb-000020
则监测器输出警报信息，即警报信号的值由0变成1。
因此，监测器增益Π保证椭球体
Figure PCTCN2022122163-appb-000021
包含了所有可能由ω ui(k)和ω i(k)所导 致的残差系统(6)的动态轨迹。除此之外，可以通过调整Π使得椭球的体积最小，从而使得检测方法对于异常信号更加敏感。
进一步地，根据隐秘FDI攻击信号的特点(即不会触发监测器发出警报信息)，需要保证隐秘FDI攻击信号在z i(k)≤1的情况下被注入车辆的系统，因此，隐秘FDI攻击信号需满足如下的数学表达式：
Figure PCTCN2022122163-appb-000022
其中，
Figure PCTCN2022122163-appb-000023
且对于所有的k≥0，满足r i(k) T∏r i(k)≤1。由于
Figure PCTCN2022122163-appb-000024
e i(k)为有界信号，
Figure PCTCN2022122163-appb-000025
也就是说，可以通过一系列有界信号表示δ i(k)。
进一步地，在实施本申请的方法评估V2V通信网络相关的隐秘FDI攻击信号对车辆的影响时，需要先获取上述隐秘FDI攻击信号δ i(k)。
S302，根据该隐秘FDI攻击信号确定第一可达集，以及危险集，该第一可达集包括随该隐秘FDI攻击信号变化的该车辆的动力学状态，该危险集包括发生危险情况时该车辆的动力学状态。
示例性地，第一可达集可以理解为：对车辆系统的已有设计，即观测器、检测器和控制器的增益已知，在隐秘FDI攻击信号和/或系统扰动信号的影响下，车辆的动力学状态所能达到的所有可能的集合。即当
Figure PCTCN2022122163-appb-000026
δ i(k)满足
Figure PCTCN2022122163-appb-000027
时，所有x i(k)可能的集合。记
Figure PCTCN2022122163-appb-000028
为在k时刻的第一可达集。需要说明的是，由于
Figure PCTCN2022122163-appb-000029
从数学上无法直接求得，因此可以求解
Figure PCTCN2022122163-appb-000030
的椭球近似集
Figure PCTCN2022122163-appb-000031
通过椭球近似集
Figure PCTCN2022122163-appb-000032
近似表示第一可达集。
示例性地，系统扰动信号可以包括如下至少一项：观测器的估计误差、残差信号、传感器噪声(例如雷达噪声、速度传感器噪声等)、网络通信噪声、攻击信号引起的扰动。其中，该攻击信号包括故障或随机FDI攻击信号，以及隐秘FDI攻击信号。
求解第一可达集可以包括如下步骤(1)至(2)：
步骤(1)：求解
Figure PCTCN2022122163-appb-000033
的椭球近似集。
为了求解第一可达集，可以将系统转化为线性时不变系统，定义延展状态ζ i(k)＝
Figure PCTCN2022122163-appb-000034
则将公式(7)代入公式(3)和(6)可得如下形式的线性时不变系统：
Figure PCTCN2022122163-appb-000035
其中，
Figure PCTCN2022122163-appb-000036
Figure PCTCN2022122163-appb-000037
应理解，该线性时不变系统中，所有扰动信号均为有界信号。
Figure PCTCN2022122163-appb-000038
为系统(8)的第一可达集。由于x i是ζ i的部分，可以先求解
Figure PCTCN2022122163-appb-000039
的近似集
Figure PCTCN2022122163-appb-000040
然后将
Figure PCTCN2022122163-appb-000041
在映射到x i平面得到
Figure PCTCN2022122163-appb-000042
示例性地，通过求解如下线性矩阵不等式(linear matrix inequality)可以获得
Figure PCTCN2022122163-appb-000043
对于给定的a∈(0,1)，如果存在常量a 1，a 2，a 3，a 4和矩阵P，使得如下线性矩阵不等式成立：
Figure PCTCN2022122163-appb-000044
其中，
Figure PCTCN2022122163-appb-000045
W a＝diag[(1-a 1)W 1,…,(1-a 4)W 4]，
Figure PCTCN2022122163-appb-000046
Figure PCTCN2022122163-appb-000047
W 4＝Π，则对于所有的k，有
Figure PCTCN2022122163-appb-000048
且
Figure PCTCN2022122163-appb-000049
若在求解(9)的过程中以最小化-log(det(P))为目标方程(det(P)表示矩阵P的行列式)，则获得的椭球近似集
Figure PCTCN2022122163-appb-000050
为最紧的近似集，即与第一可达集的近似程度最高。
步骤(2)：求解
Figure PCTCN2022122163-appb-000051
的椭球近似集。
示例性地，可以将矩阵P进行如下分解：
Figure PCTCN2022122163-appb-000052
其中，
Figure PCTCN2022122163-appb-000053
则对于所有的k，有
Figure PCTCN2022122163-appb-000054
且
Figure PCTCN2022122163-appb-000055
其中，P x＝P 1-P 2(P 3) -1(P 2) T，且((P 3) -1为矩阵P 3的逆矩阵)。
示例性地，可达集和其椭球外近似之间的关系可以如图4所示。
可选地，还可以进一步根据椭球近似集
Figure PCTCN2022122163-appb-000056
确定
Figure PCTCN2022122163-appb-000057
的近似体积，其中，
Figure PCTCN2022122163-appb-000058
的体积可由-log(det(P))定量刻画。根据该近似体积作为评估车辆风险的一个指标，确定隐秘FDI攻击信号对车辆动力学的影响。
示例性地，车辆的危险集定义为车辆发生违章行为或危险时，如车辆超速行驶，或车辆与前车发生碰撞时车辆的动力学状态集合，可表示为v i(k)>v max或者d i(k)<0。其中，v max可以为第i辆车所能达到的最大速度，或者也可以为第i辆车所行驶道路的最大限速，或者还可以是为保证车辆编队的行驶过程中的安全性所允许的最大速度，或者还可以为其他速度阈值，本申请实施例对此不作具体限定。
在一些可能的实现方式中，车辆的跟踪误差ε i(k)满足ε i(k)＝d i(k)-(s i+hv i(k))，其中，d i(k)为第k时刻第i辆车与第i-1辆车之间的实际距离，s i+hv i(k)为期望的第k时刻第i辆车与第i-1辆车之间的距离，v i(k)为第k时刻第i辆车的速度，h为车辆期望保持一定速度的时间间隔，h的具体数值可以由车辆的用户设定，或者也可以为车辆出厂时的默认数值，本申请实施例对此不作具体限定；s i为期望的车辆停止时两辆车之间的距离(或称停车距离，standstill distance)，s i的具体数值可以为车辆出厂时的默认数值。在一些可能的实现方式中，s i的具体数值随车辆的车型变化，例如，对于小型车，该s i可以取4.92英尺，或者也可以取4.5至5.5英尺中的任一数值；对于货车，该数值可以大于5.5英尺。
示例性地，上述两车之间的距离可以为前车的后保险杠与后车的前保险杠之间的距离。
则当d i(k)<0时，应有ε i(k)+s i+hv i(k)<0。则危险集可表达为：
Figure PCTCN2022122163-appb-000059
其中，
Figure PCTCN2022122163-appb-000060
S303，根据该第一可达集与该危险集之间的距离，确定该隐秘FDI攻击信号对该车辆的影响。
示例性地，可以根据如下公式确定第一可达集和危险集之间的距离：
Figure PCTCN2022122163-appb-000061
示例性地，若
Figure PCTCN2022122163-appb-000062
(如图5中的(a)所示)，则车辆在隐秘攻击下不会发生超速或碰撞，即隐秘FDI攻击信号不会对车辆安全产生威胁。若
Figure PCTCN2022122163-appb-000063
则车辆在隐秘攻击下可能发生超速或者碰撞的危险，即隐秘FDI攻击信号可能会对车辆安全产生威胁。例如，对于图5中(b)所示的情况，由于危险集与第一可达集相交，所以隐秘FDI攻击信号会导致车辆发生危险；对于图5中(c)所示的情况，虽然在
Figure PCTCN2022122163-appb-000064
但由于危险集与第一可达集未相交，所以车辆在隐秘FDI攻击信号的作用下不会发生危险。
在一些可能的实现方式中，在本申请实施例提供的评估车辆风险的方法应用于自动化列队行驶(platooning)中的车辆控制时，在求解椭球近似集(或称最优化问题)时，还需引入估计误差、车辆跟踪误差系统稳定性、车辆队列弦稳定性(string stability)作为约束。或者，还可以引入检测器的误报率作为约束。应理解，车辆队列弦稳定性(string stability)可以保证车辆中的小干扰不会沿车辆队列放大，因此保持了安全性。例如，车辆队列弦稳定性(string stability)保证了领头车辆的突然刹车不会导致其跟随者发生碰撞。
可选地，还可以通过蒙特卡洛模拟方法确定第一可达集。示例性地，在每次仿真中，给定一个初始的车辆条件，然后通过添加隐秘FDI攻击信号，并使用模型迭代，可以获得状态轨迹。通过多个不同的初始的车辆条件并重复上述过程，可以获得多个状态轨迹，该多个状态轨迹可以作为可达集的粗略近似。
可选地，还可以通过除了椭球外近似以外的其他方式确定可达集的近似集。例如，可以使用多面体近似，或者齐诺多面体(zonotopes)近似确定可达集的近似集；或者还可以通过机器学习方法进行可达性分析，本申请实施例对此不作具体限定。
本申请实施例提供的一种评估车辆风险的方法，通过两个风险评估指标(可达集的体积，以及可达集与危险集之间的距离)，可以评估隐秘FDI攻击对车辆动力学的影响，有助于指导车辆系统的重新设计。特别地，在建立车辆模型的过程中，考虑了系统扰动信号，使得车辆模型与实际车辆更接近，提高了评估结果的准确性。
在许多情况下，车辆系统的已有设计(包括检测器，控制器的设计)，主要考虑的都是车辆无异常情况下的性能。例如，卡尔曼滤波器可以在车辆存在高斯白噪声的情况下对车辆的状态实现最优估计，车辆控制器需要满足最佳的车辆控制基本性能等。然而，车辆无异常情况下的性能对应的最优设计，无法保证在车辆遭受隐秘FDI攻击时的安全车控。
因此，在车辆监测攻击的系统的已有设计被评估为存在风险时，可以对其进行重新设计。示例性地，如图7所示，该监测攻击的系统可以包括检测器(包括观测器、监测器)，或者，还可以包括控制器。示例性地，观测器用于根据在第一时刻从车辆的第一设备获取第一数据，确定所述第一设备在第二时刻的第二数据，其中，所述第二时刻为所述第一时刻之后的时刻；监测器用于确定所述第二数据和第三数据之间的差值，所述第三数据为在所述第二时刻从所述第一设备获取的数据；在所述差值大于或等于第一阈值时，输出警报信息，所述警报信息用于指示所述车辆遭受攻击；控制器用于满足车辆控制基本性能，该 控制器可以将控制器输出输入至观测器。其中，控制器输出可以包括控制器期望车辆达到的运动状态，例如期望速度、期望加速度、期望方向盘转角等。应理解，上述期望的运动状态经车辆的执行器执行后，可以为车辆的实际运动状态。示例性地，在CACC系统中，该控制器输出可以包括期望加速度；在其他智能驾驶系统中，该控制器输出还可以为期望速度、期望方向盘转角等。该监测攻击的系统中，观测器的增益、控制器的增益以及监测器的增益中的至少一个，是根据隐秘FDI攻击信号确定的。该监测攻击的系统重新设计后，能够提高车辆抵抗隐秘FDI攻击的能力，即能够降低隐秘FDI攻击信号对车辆安全的影响。
需要说明的是，系统重设计仍是对最优化问题(9)的求解，然而，与评估车辆风险的阶段不同的是，此时的观测器增益L，监测器增益Π，和控制器增益K为未知变量。这使得公式(9)不再是线性矩阵不等式，而是非线性矩阵不等式。为了将观测器增益L，监测器增益Π，和控制器增益K为未知变量情况下的公式(9)线性化，可以在求解公式(9)的过程中令矩阵P具有以下的结构：
Figure PCTCN2022122163-appb-000065
其中，
Figure PCTCN2022122163-appb-000066
均为半正定矩阵。将控制器增益K＝[k pk d]与矩阵A解耦合，定义新变量
Figure PCTCN2022122163-appb-000067
则可以将公式(9)转化为以
Figure PCTCN2022122163-appb-000068
为变量的线性矩阵不等式。
需要说明的是，在求解公式(9)以确定上述增益时，还需满足k p>0，k d>k pτ，从而满足CACC的车辆队列(string)稳定性。由此，可得到新的观测器增益
Figure PCTCN2022122163-appb-000069
新的控制器增益
Figure PCTCN2022122163-appb-000070
进一步地，可以求解得到新的监测器增益Π。
如图6中所示，在调整观测器、控制器和监测器中至少一个的增益后，可达集和危险集之间的距离由图6中的(a)所示的负距离，变为图6中的(b)所示的正距离。也就是说，图6中的(b)所示的情况下，车辆在隐秘FDI攻击信号的作用下不会发生危险。
本申请实施例提供的一种监测攻击的系统中，由于根据隐秘FDI攻击信号对监测攻击的系统进行了重新设计，能够降低隐秘FDI攻击信号对使用该监测攻击的系统的车辆的影响，有助于提升车辆的安全性；并且能够保证监测攻击的系统的性能，例如，观测器能够保证有界估计误差；监测器能够以可接受的误报率监测攻击信号；该控制器能够保证车辆跟踪误差的动态稳定，以及车辆队列的弦稳定性(string stability)。
需要说明的是，根据隐秘FDI攻击信号重新设计的监测攻击的系统，可以在车辆出厂时设置在车辆中。或者，也可以通过软件升级的方式更新车辆中已有的监测攻击的系统，例如，可以通过空中下载技术(over the air，OTA)将重新设计的监测攻击的系统更新至车辆中。在一些可能的实现方式中，在进行车辆的监测攻击的系统更新时，可以将仅更新重新设计过程中确定的一个或多个增益。
还需说明的是，本申请提供的评估车辆风险的方法以及监测攻击的系统，不仅适用于CACC驾驶场景，还适用于自适应巡航控制(adaptive cruise control，ACC)、智驾导航辅助(navigation cruise assistant，NCA)或者智能巡航辅助(integrated cruise assistant，ICA)等驾驶场景。此外，上述实施例以隐秘FDI攻击信号与V2V通信网络相关为例进行说明，应理解，本领域技术人员根据本申请的构思，评估与GNSS，毫米波雷达，超声波雷达等传感器，以及车内通信网络等相关的隐秘FDI攻击信号，对车辆的影响的方案，以及基于 上述隐秘FDI攻击信号对监测攻击的系统重新设计的方案，也应包含在本申请的保护范围之内。
在本申请的各个实施例中，如果没有特殊说明以及逻辑冲突，各个实施例之间的术语和/或描述具有一致性、且可以相互引用，不同的实施例中的技术特征根据其内在的逻辑关系可以组合形成新的实施例。
上文中结合图2至图7详细说明了本申请实施例提供的方法。下面将结合图8和图9详细说明本申请实施例提供的装置。应理解，装置实施例的描述与方法实施例的描述相互对应，因此，未详细描述的内容可以参见上文方法实施例，为了简洁，这里不再赘述。
图8示出了本申请实施例提供的一种评估车辆风险的装置2000的示意性框图，该装置2000包括获取单元2010和处理单元2020。
该装置2000可以包括用于执行图3中的方法的单元。并且，该装置2000中的各单元和上述其他操作和/或功能分别为了实现图3中的方法实施例的相应流程。
其中，当该装置2000用于执行图3中的方法300时，获取单元2010可用于执行方法300中的S301，处理单元2020可用于执行方法300中的S302和S303。
具体地，该获取单元2010，用于获取针对车辆的隐秘虚假信息注入FDI攻击信号；处理单元2020，用于根据该隐秘FDI攻击信号确定第一可达集，以及危险集，该第一可达集包括随该隐秘FDI攻击信号变化的该车辆的动力学状态，该危险集包括发生危险情况时该车辆的动力学状态；根据该第一可达集与该危险集之间的距离，确定该隐秘FDI攻击信号对该车辆的影响。即，确定隐秘FDI攻击信号是否会导致车辆发生危险。
可选地，该处理单元2020还用于：根据该第一可达集的体积确定该隐秘FDI攻击信号对该车辆的影响。
可选地，该隐秘FDI攻击信号为根据如下至少一个确定：监测器残差信号，观测器的估计误差，以及传感器噪声；其中，该观测器用于根据该传感器的测量值，确定该传感器在下一时间步的估计值；该监测器用于根据该传感器的测量值与该观测器估计值之间的差值，对该车辆的攻击进行监测。
可选地，该隐秘FDI攻击信号与如下至少一项相关联：车对车V2V通信网络，全球定位系统GPS信号，毫米波雷达，超声波雷达，以及车内通信网络。
可选地，该处理单元2020用于：根据该隐秘FDI攻击信号确定第一可达集的近似集，该近似集是以车辆编队的车辆队列弦稳定性，和/或观测器的估计误差为约束求解凸优化问题确定，其中，该车辆处于该车辆编队。
可选地，该处理单元2020用于：根据该隐秘FDI攻击信号通过蒙特卡洛模拟方法确定该第一可达集。
可选地，该危险情况包括如下至少一项：该车辆的速度大于或等于速度阈值，该车辆的加速度大于或等于加速度阈值，该车辆与最紧邻前车之间的距离小于或等于距离阈值。
可选地，该处理单元2020还用于：在该第一可达集与该危险集之间的距离小于或等于预设阈值时，根据该隐秘FDI攻击信号确定监测器和/或观测器的增益；其中，该增益与第二可达集相关联，该第二可达集与该危险值之间的距离大于该预设阈值。
应理解，以上装置中各单元的划分仅是一种逻辑功能的划分，实际实现时可以全部或部分集成到一个物理实体上，也可以物理上分开。此外，装置中的单元可以以处理器调用 软件的形式实现；例如装置包括处理器，处理器与存储器连接，存储器中存储有指令，处理器调用存储器中存储的指令，以实现以上任一种方法或实现该装置各单元的功能，其中处理器例如为通用处理器，例如CPU或微处理器，存储器为装置内的存储器或装置外的存储器。或者，装置中的单元可以以硬件电路的形式实现，可以通过对硬件电路的设计实现部分或全部单元的功能，该硬件电路可以理解为一个或多个处理器；例如，在一种实现中，该硬件电路为ASIC，通过对电路内元件逻辑关系的设计，实现以上部分或全部单元的功能；再如，在另一种实现中，该硬件电路为可以通过PLD实现，以FPGA为例，其可以包括大量逻辑门电路，通过配置文件来配置逻辑门电路之间的连接关系，从而实现以上部分或全部单元的功能。以上装置的所有单元可以全部通过处理器调用软件的形式实现，或全部通过硬件电路的形式实现，或部分通过处理器调用软件的形式实现，剩余部分通过硬件电路的形式实现。
在本申请实施例中，处理器是一种具有信号的处理能力的电路，在一种实现中，处理器可以是具有指令读取与运行能力的电路，例如CPU、微处理器、GPU、或DSP等；在另一种实现中，处理器可以通过硬件电路的逻辑关系实现一定功能，该硬件电路的逻辑关系是固定的或可以重构的，例如处理器为ASIC或PLD实现的硬件电路，例如FPGA。在可重构的硬件电路中，处理器加载配置文档，实现硬件电路配置的过程，可以理解为处理器加载指令，以实现以上部分或全部单元的功能的过程。此外，还可以是针对人工智能设计的硬件电路，其可以理解为一种ASIC，例如NPU、TPU、DPU等。
可见，以上装置中的各单元可以是被配置成实施以上方法的一个或多个处理器(或处理电路)，例如：CPU、GPU、NPU、TPU、DPU、微处理器、DSP、ASIC、FPGA，或这些处理器形式中至少两种的组合。
此外，以上装置中的各单元可以全部或部分可以集成在一起，或者可以独立实现。在一种实现中，这些单元集成在一起，以片上系统(system-on-a-chip，SOC)的形式实现。该SOC中可以包括至少一个处理器，用于实现以上任一种方法或实现该装置各单元的功能，该至少一个处理器的种类可以不同，例如包括CPU和FPGA，CPU和人工智能处理器，CPU和GPU等。
在具体实现过程中，上述获取单元2010和处理单元2020所执行的各项操作可以由同一个处理器执行，或者，也可以由不同的处理器执行，例如分别由多个处理器执行。
本申请实施例还提供了一种装置，该装置包括处理单元和存储单元，其中存储单元用于存储指令，处理单元执行存储单元所存储的指令，以使该装置执行上述实施例执行的方法或者步骤。
图9是本申请实施例的一种评估车辆风险的装置的示意性框图。图9所示的评估车辆风险的装置2100可以包括：处理器2110、收发器2120以及存储器2130。其中，处理器2110、收发器2120以及存储器2130通过内部连接通路相连，该存储器2130用于存储指令，该处理器2110用于执行该存储器2130存储的指令，以收发器2120接收/发送部分参数。可选地，存储器2130既可以和处理器2110通过接口耦合，也可以和处理器2110集成在一起。
需要说明的是，上述收发器2120可以包括但不限于输入/输出接口(input/output interface)一类的收发装置，来实现装置2100与其他设备或通信网络之间的通信。
处理器2110可以采用通用的CPU，微处理器，ASIC，GPU或者一个或多个集成电路，用于执行相关程序，以实现本申请方法实施例的评估车辆风险的方法。处理器2110还可以是一种集成电路芯片，具有信号的处理能力。在具体实现过程中，本申请的评估车辆风险的方法的各个步骤可以通过处理器2110中的硬件的集成逻辑电路或者软件形式的指令完成。上述处理器2110还可以是通用处理器、DSP、ASIC、FPGA或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本申请实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器2130，处理器2110读取存储器2130中的信息，结合其硬件执行本申请方法实施例的评估车辆风险的方法。
存储器2130可以是只读存储器(read only memory，ROM)，静态存储设备，动态存储设备或者随机存取存储器(random access memory，RAM)。
收发器2120使用例如但不限于收发器一类的收发装置，来实现装置2100与其他设备或通信网络之间的通信。例如，可以通过收发器2120获取用户位置信息。
本申请实施例还提供一种服务器，该服务器可以包括上述装置2000，或者上述装置2100。
本申请实施例还提供了一种计算机程序产品，该计算机程序产品包括：计算机程序代码，当该计算机程序代码在计算机上运行时，使得计算机执行上述实施例中的方法。
本申请实施例还提供一种计算机可读存储介质，该计算机可读介质存储有程序代码或指令，当该计算机程序代码或指令被计算机的处理器执行时，使得该处理器实现上述实施例中的方法。
本申请实施例还提供一种芯片，包括：至少一个处理器和存储器，该至少一个处理器与该存储器耦合，用于读取并执行该存储器中的指令，以执行上述实施例中的方法。
本申请将围绕包括多个设备、组件、模块等的系统来呈现各个方面、实施例或特征。应当理解和明白的是，各个系统可以包括另外的设备、组件、模块等，并且/或者可以并不包括结合附图讨论的所有设备、组件、模块等。此外，还可以使用这些方案的组合。
另外，在本申请实施例中，“示例的”、“例如”等词用于表示作例子、例证或说明。本申请中被描述为“示例”的任何实施例或设计方案不应被解释为比其它实施例或设计方案更优选或更具优势。确切而言，使用示例的一词旨在以具体方式呈现概念。
本申请实施例中，“相应的(corresponding，relevant)”和“对应的(corresponding)”有时可以混用，应当指出的是，在不强调其区别时，其所要表达的含义是一致的。
在本说明书中描述的参考“一个实施例”或“一些实施例”等意味着在本申请的一个或多个实施例中包括结合该实施例描述的特定特征、结构或特点。由此，在本说明书中的不同之处出现的语句“在一个实施例中”、“在一些实施例中”、“在其他一些实施例中”、“在另外一些实施例中”等不是必然都参考相同的实施例，而是意味着“一个或多个但不是所有的实施例”，除非是以其他方式另外特别强调。术语“包括”、“包含”、“具有”及它们的变形都意味着“包括但不限于”，除非是以其他方式另外特别强调。
本申请中，“至少一个”是指一个或者多个，“多个”是指两个或两个以上。“和/或”，描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：包括单独存在A，同时存在A和B，以及单独存在B的情况，其中A，B可以是单数或者复数。字符“/”一般表示前后关联对象是一种“或”的关系。“以下至少一项(个)”或其类似表达，是指的这些项中的任意组合，包括单项(个)或复数项(个)的任意组合。例如，a，b，或c中的至少一项(个)，可以表示：a，b，c，a-b，a-c，b-c，或a-b-c，其中a，b，c可以是单个，也可以是多个。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (23)

  1. 一种评估车辆风险的方法，其特征在于，包括：
    获取针对车辆的隐秘虚假信息注入FDI攻击信号；
    根据所述隐秘FDI攻击信号确定第一可达集，以及危险集；其中，所述第一可达集包括随所述隐秘FDI攻击信号变化的所述车辆的动力学状态，所述危险集包括发生危险情况时所述车辆的动力学状态；
    根据所述第一可达集与所述危险集之间的距离，确定所述隐秘FDI攻击信号对所述车辆的影响。
  2. 如权利要求1所述的方法，其特征在于，所述方法还包括：
    根据所述第一可达集的体积量化评估所述隐秘FDI攻击信号对所述车辆的影响。
  3. 如权利要求1或2所述的方法，其特征在于，所述隐秘FDI攻击信号为根据如下至少一个确定：
    监测器的残差信号，观测器的估计误差，传感器噪声，以及通信网络噪声；
    其中，所述观测器用于根据所述传感器的测量值，确定所述传感器在下一时刻的估计值；所述监测器用于根据所述传感器的测量值与所述观测器估计值之间的差值，对针对所述车辆的攻击进行监测。
  4. 如权利要求1至3中任一项所述的方法，其特征在于，所述隐秘FDI攻击信号与如下至少一项相关联：
    车对车V2V通信网络，全球导航卫星系统GNSS信号，毫米波雷达，超声波雷达，以及车内通信网络。
  5. 如权利要求1至4中任一项所述的方法，其特征在于，所述危险情况包括如下至少一项：所述车辆的速度大于或等于速度阈值，所述车辆的加速度大于或等于加速度阈值，所述车辆与最紧邻前车之间的距离小于或等于距离阈值。
  6. 如权利要求1至5中任一项所述的方法，其特征在于，所述方法还包括：
    在所述第一可达集与所述危险集之间的距离小于或等于预设阈值时，根据所述隐秘FDI攻击信号确定监测器和/或观测器的增益；
    其中，所述增益与第二可达集相关联，所述第二可达集与所述危险值之间的距离大于所述预设阈值。
  7. 一种评估车辆风险的装置，其特征在于，包括：
    获取单元，用于获取针对车辆的隐秘虚假信息注入FDI攻击信号；
    处理单元，用于根据所述隐秘FDI攻击信号确定第一可达集，以及危险集；其中，所述第一可达集包括随所述隐秘FDI攻击信号变化的所述车辆的动力学状态，所述危险集包括发生危险情况时所述车辆的动力学状态；
    根据所述第一可达集与所述危险集之间的距离，量化评估所述隐秘FDI攻击信号对所述车辆的影响。
  8. 如权利要求7所述的装置，其特征在于，所述处理单元还用于：
    根据所述第一可达集的体积确定所述隐秘FDI攻击信号对所述车辆的影响。
  9. 如权利要求7或8所述的装置，其特征在于，所述隐秘FDI攻击信号为根据如下至少一个确定：
    监测器的残差信号，观测器的估计误差，传感器噪声，以及通信网络噪声；
    其中，所述观测器用于根据所述传感器的测量值，确定所述传感器在下一时刻的估计值；所述监测器用于根据所述传感器的测量值与所述观测器估计值之间的差值，对针对所述车辆的攻击进行监测。
  10. 如权利要求7至9中任一项所述的装置，其特征在于，所述隐秘FDI攻击信号与如下至少一项相关联：
    车对车V2V通信网络，全球导航卫星系统GNSS，毫米波雷达，超声波雷达，以及车内通信网络。
  11. 如权利要求7至10中任一项所述的装置，其特征在于，所述危险情况包括如下至少一项：所述车辆的速度大于或等于速度阈值，所述车辆的加速度大于或等于加速度阈值，所述车辆与最紧邻前车之间的距离小于或等于距离阈值。
  12. 如权利要求7至11中任一项所述的装置，其特征在于，所述处理单元还用于：
    在所述第一可达集与所述危险集之间的距离小于或等于预设阈值时，根据所述隐秘FDI攻击信号确定监测器和/或观测器的增益；
    其中，所述增益与第二可达集相关联，所述第二可达集与所述危险值之间的距离大于所述预设阈值。
  13. 一种监测攻击的系统，其特征在于，包括：
    观测器，用于根据在第一时刻从车辆的第一设备获取第一数据，确定所述第一设备在第二时刻的第二数据，其中，所述第二时刻为所述第一时刻之后的时刻；
    监测器，用于确定所述第二数据和第三数据之间的差值，所述第三数据为在所述第二时刻从所述第一设备获取的数据；在所述差值大于第一阈值时，输出警报信息，所述警报信息用于指示所述车辆遭受攻击；
    其中，所述监测器的增益为根据隐秘FDI攻击信号确定。
  14. 如权利要求13所述的系统，其特征在于，所述观测器的增益为根据所述隐秘FDI攻击信号确定。
  15. 如权利要求13或14所述的系统，其特征在于，所述系统还包括控制器，用于实现车辆控制基本性能，所述控制器的增益为根据所述隐秘FDI攻击信号确定。
  16. 如权利要求13至15中任一项所述的系统，其特征在于，所述增益为根据所述车辆的第一可达集和危险集确定；
    其中，所述第一可达集为根据所述观测器、所述监测器和所述控制器的初始增益确定，所述第一可达集包括随所述隐秘FDI攻击信号变化的所述车辆的动力学状态，所述危险集包括发生危险情况时所述车辆的动力学状态，其中，所述第一可达集和所述危险集之间的距离小于或等于第二阈值。
  17. 如权利要求16所述的系统，其特征在于，第二可达集和所述危险集之间的距离大于所述第二阈值，所述第二可达集与所述增益相对应。
  18. 一种评估车辆风险的装置，其特征在于，包括：
    存储器，用于存储计算机程序；
    处理器，用于执行所述存储器中存储的计算机程序，以使得所述装置执行如权利要求1至6中任一项所述的方法。
  19. 一种车辆，其特征在于，包括如权利要求13至17中任一项所述的系统。
  20. 一种服务器，其特征在于，包括如权利要求7至12中任一项所述的装置，或者如权利要求18所述的装置。
  21. 一种计算机可读存储介质，其特征在于，其上存储有指令，所述指令被处理器执行时，以使得处理器实现如权利要求1至6中任一项所述的方法。
  22. 一种芯片，其特征在于，所述芯片包括处理器与数据接口，所述处理器通过所述数据接口读取存储器上存储的指令，以执行如权利要求1至6中任一项所述的方法。
  23. 一种计算机程序产品，其特征在于，所述计算机程序产品包括：计算机程序代码，当上述计算机程序代码在计算机上运行时，使得计算机执行如权利要求1至6中任一项所述的方法。
PCT/CN2022/122163 2022-09-28 2022-09-28 评估车辆风险的方法、装置以及监测攻击的系统 WO2024065283A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/122163 WO2024065283A1 (zh) 2022-09-28 2022-09-28 评估车辆风险的方法、装置以及监测攻击的系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/122163 WO2024065283A1 (zh) 2022-09-28 2022-09-28 评估车辆风险的方法、装置以及监测攻击的系统

Publications (1)

Publication Number Publication Date
WO2024065283A1 true WO2024065283A1 (zh) 2024-04-04

Family

ID=90475146

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/122163 WO2024065283A1 (zh) 2022-09-28 2022-09-28 评估车辆风险的方法、装置以及监测攻击的系统

Country Status (1)

Country Link
WO (1) WO2024065283A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190113549A1 (en) * 2017-10-13 2019-04-18 Nec Laboratories America, Inc. Detection of false data injection attacks in power systems using multiplex invariant networks and domain knowledge
CN111783845A (zh) * 2020-06-12 2020-10-16 浙江工业大学 一种基于局部线性嵌入和极限学习机的隐匿虚假数据注入攻击检测方法
CN112596387A (zh) * 2020-12-14 2021-04-02 电子科技大学 一种基于扩展观测器的网络化系统安全控制方法
CN113612733A (zh) * 2021-07-07 2021-11-05 浙江工业大学 一种基于孪生网络的少样本虚假数据注入攻击检测方法
CN114268424A (zh) * 2021-12-21 2022-04-01 上海理工大学 电动汽车负载频率控制系统中隐蔽网络攻击的检测方法
CN115032895A (zh) * 2022-05-31 2022-09-09 深圳市三旺通信股份有限公司 攻击检测方法、系统、终端设备以及存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190113549A1 (en) * 2017-10-13 2019-04-18 Nec Laboratories America, Inc. Detection of false data injection attacks in power systems using multiplex invariant networks and domain knowledge
CN111783845A (zh) * 2020-06-12 2020-10-16 浙江工业大学 一种基于局部线性嵌入和极限学习机的隐匿虚假数据注入攻击检测方法
CN112596387A (zh) * 2020-12-14 2021-04-02 电子科技大学 一种基于扩展观测器的网络化系统安全控制方法
CN113612733A (zh) * 2021-07-07 2021-11-05 浙江工业大学 一种基于孪生网络的少样本虚假数据注入攻击检测方法
CN114268424A (zh) * 2021-12-21 2022-04-01 上海理工大学 电动汽车负载频率控制系统中隐蔽网络攻击的检测方法
CN115032895A (zh) * 2022-05-31 2022-09-09 深圳市三旺通信股份有限公司 攻击检测方法、系统、终端设备以及存储介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
YANG TIANCI; MURGUIA CARLOS; LV CHEN: "Risk Assessment for Connected Vehicles Under Stealthy Attacks on Vehicle-to-Vehicle Networks", IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, IEEE, PISCATAWAY, NJ, USA, vol. 24, no. 12, 1 December 2023 (2023-12-01), Piscataway, NJ, USA , pages 13627 - 13638, XP011954766, ISSN: 1524-9050, DOI: 10.1109/TITS.2023.3292188 *

Similar Documents

Publication Publication Date Title
US11354406B2 (en) Physics-based approach for attack detection and localization in closed-loop controls for autonomous vehicles
EP3875907B1 (en) Method, apparatus, computing device and computer-readable storage medium for positioning
US10102761B2 (en) Route prediction device
CN110884490B (zh) 一种车辆侵入判断及辅助行驶的方法、系统、车辆及存储介质
CN109739230B (zh) 驾驶轨迹生成方法、装置及存储介质
US20240046798A1 (en) Method and apparatus for automated vehicle platooning
US11586212B2 (en) Vehicle device localization
Bernsteiner et al. Radar sensor model for the virtual development process
US20180203446A1 (en) Object tracking by unsupervised learning
US11423162B2 (en) Systems and methods for message assurance in vehicle systems
US11167754B2 (en) Systems and methods for trajectory based safekeeping of vehicles
US11731661B2 (en) Systems and methods for imminent collision avoidance
US11072326B2 (en) Systems and methods for trajectory based safekeeping of vehicles
Simon et al. Motion planning for collision mitigation via fem–based crash severity maps
Zhang et al. Cyber-attack detection for autonomous driving using vehicle dynamic state estimation
Lakhal et al. Risk management for intelligent vehicles based on interval analysis of ttc
WO2024065283A1 (zh) 评估车辆风险的方法、装置以及监测攻击的系统
EP4141483A1 (en) Target detection method and apparatus
WO2024119401A1 (zh) 异常检测的方法、装置和车辆
US11891061B2 (en) Target vehicle detection
Lee et al. Development of an emergency control algorithm for a fail-safe system in automated driving vehicles
EP4017772A1 (en) Systems and methods for trajectory based safekeeping of vehicles
Bithar et al. Application of collision probability estimation to calibration of advanced driver assistance systems
SE543781C2 (en) Method and control unit for predicting a collision between a vehicle and a mobile object
CN114148344B (zh) 一种车辆行为预测方法、装置及车辆

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22959929

Country of ref document: EP

Kind code of ref document: A1