WO2025103243A1 - 一种入侵检测方法、装置以及车辆 - Google Patents

一种入侵检测方法、装置以及车辆 Download PDF

Info

Publication number
WO2025103243A1
WO2025103243A1 PCT/CN2024/131072 CN2024131072W WO2025103243A1 WO 2025103243 A1 WO2025103243 A1 WO 2025103243A1 CN 2024131072 W CN2024131072 W CN 2024131072W WO 2025103243 A1 WO2025103243 A1 WO 2025103243A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
data
characteristic data
feature data
intrusion detection
Prior art date
Legal status (The legal status 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 status listed.)
Pending
Application number
PCT/CN2024/131072
Other languages
English (en)
French (fr)
Inventor
吕麒
刘生林
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Yinwang Intelligent Technology Co Ltd
Original Assignee
Shenzhen Yinwang Intelligent Technology Co Ltd
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 Shenzhen Yinwang Intelligent Technology Co Ltd filed Critical Shenzhen Yinwang Intelligent Technology Co Ltd
Publication of WO2025103243A1 publication Critical patent/WO2025103243A1/zh
Anticipated expiration legal-status Critical
Pending legal-status Critical Current

Links

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W50/02—Ensuring safety in case of control system failures, e.g. by diagnosing, circumventing or fixing failures
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W50/08—Interaction between the driver and the control system
    • B60W50/14—Means for informing the driver, warning the driver or prompting a driver intervention
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W60/00—Drive control systems specially adapted for autonomous road vehicles

Definitions

  • the present application relates to the field of vehicle technology, and in particular to an intrusion detection method, device and vehicle.
  • vehicles achieve autonomous driving through the internal autonomous driving system that calculates the path based on the map and vehicle environment information.
  • the safe driving of autonomous vehicles depends not only on the perception of surrounding obstacles, but also on the centimeter-level positioning of the vehicle on the map by the global satellite navigation system. If the positioning is wrong, it will directly cause the vehicle to run off the road or drive in the wrong direction, resulting in a vehicle accident.
  • the present application provides an intrusion detection method, device and vehicle for detecting attacks on vehicle positioning signals to ensure the safe driving of autonomous driving vehicles.
  • the present application provides an intrusion detection method, which can be performed by an intrusion detection device, which can be an independent device, a chip or component in a device, or software.
  • the intrusion detection device can be deployed on a vehicle, which can be a vehicle in a fully manual driving mode, or a vehicle in a fully automatic driving mode, or the vehicle can be configured as a vehicle in a partially automatic driving mode.
  • a vehicle in a partially automatic driving mode for example, means that the vehicle can control itself while in the automatic driving mode, and can determine the current state of the vehicle and the surrounding environment through human operation, determine the possible behavior of at least one other vehicle in the surrounding environment, and control the vehicle based on the determined information.
  • the vehicle can be set to operate without human interaction.
  • the present application does not limit the product form and deployment method of the intrusion detection device.
  • the method includes: obtaining first characteristic data, the first characteristic data describing information of static traffic objects related to the area where the first vehicle is located; obtaining second characteristic data, the second characteristic data describing information of dynamic traffic objects related to the area where the first vehicle is located; determining the occurrence of an intrusion event based on the first characteristic data, the second characteristic data and an intrusion detection rule, the intrusion detection rule describing a detection method for an intrusion event associated with a service to be detected of the first vehicle; and sending an alarm message to the intelligent driving system of the first vehicle, the alarm message indicating the intrusion event.
  • the service to be detected includes a positioning service.
  • both automatic driving and intelligent driving refer to the driving behavior performed by a vehicle under the control of an intelligent driving system, and are not distinguished.
  • the intrusion detection device can perform intrusion detection based on the information of static traffic objects and dynamic traffic objects to prevent attackers from blocking the vehicle's position sensor and then forging the position information to perform counterfeit attacks.
  • the feature data relied on for intrusion detection does not come from the position sensor, and can effectively detect a wide range of position signal counterfeit attacks without incurring additional hardware costs.
  • obtaining the first characteristic data includes: filtering the first characteristic data from N types of third characteristic data, wherein the N types of third characteristic data are obtained after perceptual processing of N types of perception data, and the N types of perception data are original perception data collected by N sensors associated with the first vehicle, N is greater than or equal to 1, and the N sensors do not include position sensors.
  • the characteristic data relied on for intrusion detection does not come from the position sensor but from other sensor devices installed on the vehicle, which can effectively detect a wide range of position signal counterfeiting attacks.
  • the perception processing includes multi-target detection processing
  • the N types of third feature data include labels of identified traffic objects
  • filtering the first feature data from the N types of third feature data includes: filtering the first feature data from the N types of third feature data according to the label, wherein the label associated with the first feature data is used to describe the static traffic object.
  • the vehicle's various sensors can be used to obtain specific static features of the surrounding environment and road conditions for intrusion detection. This is to prevent attackers from blocking the global positioning system (GPS) signal receiving device and then forging GPS signals to carry out counterfeit attacks.
  • GPS global positioning system
  • the dynamic traffic objects include other vehicles within a first range around the first vehicle
  • obtaining the second characteristic data includes: obtaining the second characteristic data based on N types of third characteristic data, wherein the second characteristic data includes the number of the other vehicles, and the N types of third characteristic data are obtained after perception processing of N types of perception data, and the N types of perception data are original perception data collected by N sensors associated with the first vehicle, and N is greater than or equal to 1.
  • the vehicle's various sensors can be used to obtain specific dynamic characteristics of the surrounding environment and road conditions for intrusion detection, so as to prevent attackers from forging GPS signals based on locations with similar environmental characteristics.
  • the service to be detected is associated with the location information of the first vehicle
  • the intrusion detection rule includes analyzing the difference in information obtained by different means
  • the method also includes: obtaining the first location information of the first vehicle through the positioning component of the first vehicle; obtaining fourth feature data of the first vehicle from a map server, the fourth feature data being information of surrounding vehicles of the first vehicle provided by the map server; determining the occurrence of an intrusion event based on the first feature data, the second feature data and the intrusion detection rule, including: when the difference between the location information determined according to the first feature data and the first location information meets the first condition, and the difference between the second feature data and the fourth feature data meets the second condition, determining that a positioning intrusion event has occurred.
  • the GPS signal detection feature does not come from the GPS signal itself, and can effectively detect a wide range of GPS signal spoofing attacks. At the same time, it only needs to rely on the sensor equipment and computing equipment already installed in the autonomous driving vehicle, without relying on other hardware support, and will not incur additional hardware costs.
  • the first condition includes: the similarity between the location information determined based on the first feature data and the first location information is less than or equal to a first threshold, and the method further includes: inputting the feature data associated with different sensors in the first feature data into M location prediction models to obtain M location prediction values, where M is greater than 1; weighting the M location prediction values to obtain the location information determined based on the first feature data; wherein each of the M location prediction models is used to predict a location information based on feature data obtained based on a sensor, and the M location prediction models are obtained by training based on samples in a feature library associated with GPS services.
  • the M location prediction models include a supervised learning model corresponding to at least one of the following sensors: a camera, a lidar sensor, a millimeter-wave radar sensor, or an ultrasonic radar sensor.
  • multiple location prediction models can be pre-trained to predict the vehicle's location information based on the original perception data collected by different sensors, and the predicted location information can be compared with the location information collected by the location sensor to determine whether GPS spoofing exists.
  • the fourth characteristic data includes a reference number of other vehicles within a first range around the first vehicle; the second condition includes: the similarity between the number of other vehicles indicated by the second characteristic data and the reference number of other vehicles indicated by the fourth characteristic data is less than or equal to a second threshold.
  • the present application provides an intrusion detection device, including: a first acquisition unit, used to acquire first characteristic data, the first characteristic data describes information about static traffic objects related to the area where the first vehicle is located; a second acquisition unit, used to acquire second characteristic data, the second characteristic data describes information about dynamic traffic objects related to the area where the first vehicle is located; a determination unit, used to determine the occurrence of an intrusion event based on the first characteristic data, the second characteristic data and an intrusion detection rule, the intrusion detection rule describes a detection method for an intrusion event associated with a service to be detected of the first vehicle; a communication unit, used to send an alarm message to the intelligent driving system of the first vehicle, the alarm message indicating the intrusion event.
  • the first acquisition unit is used to: filter the first feature data from N types of third feature data, wherein the N types of third feature data are obtained after perception processing of N types of perception data, and the N types of perception data are original perception data collected by N sensors associated with the first vehicle, and N is greater than or equal to 1.
  • the perception processing includes multi-target detection processing
  • the N types of third feature data include labels of identified traffic objects
  • the first acquisition unit filters the first feature data from the N types of third feature data, including: filtering the first feature data from the N types of third feature data according to the label, wherein the label associated with the first feature data is used to describe the static traffic object.
  • the dynamic traffic object includes other vehicles within a first range around the first vehicle
  • the second acquisition unit is used to: acquire the second feature data according to N types of third feature data, wherein the second feature The data includes the number of the other vehicles, the N types of third characteristic data are obtained after perception processing of the N types of perception data, the N types of perception data are original perception data collected by N sensors associated with the first vehicle, and N is greater than or equal to 1.
  • the service to be detected is associated with the location information of the first vehicle
  • the intrusion detection rule includes analyzing the difference in information obtained by different means
  • the determination unit is also used to: obtain the first location information of the first vehicle through the positioning component of the first vehicle; obtain the fourth feature data of the first vehicle from the map server, and the fourth feature data is the information of the surrounding vehicles of the first vehicle provided by the map server; when the difference between the location information determined according to the first feature data and the first location information meets the first condition, and the difference between the second feature data and the fourth feature data meets the second condition, it is determined that an intrusion event has occurred.
  • the first condition includes: the similarity between the location information determined according to the first feature data and the first location information is less than or equal to a first threshold, and the determination unit is used to: input the feature data associated with different sensors in the first feature data into M location prediction models to obtain M location prediction values, where M is greater than 1; weighted according to the M location prediction values to obtain the location information determined according to the first feature data; wherein each of the M location prediction models is used to predict a location information for feature data obtained based on a sensor, and the M location prediction models are trained based on samples in a feature library associated with the GPS service.
  • the M position prediction models include a supervised learning model corresponding to at least one of the following sensors: a camera, a lidar sensor, a millimeter-wave radar sensor, or an ultrasonic radar sensor.
  • the fourth characteristic data includes a reference number of other vehicles within a first range around the first vehicle; the second condition includes: the similarity between the number of other vehicles indicated by the second characteristic data and the reference number of other vehicles indicated by the fourth characteristic data is less than or equal to a second threshold.
  • the present application provides a communication device comprising at least one processor and an interface circuit, wherein the interface circuit is used to provide data or code instructions to the at least one processor, and the at least one processor is used to implement the method described in the first aspect and any possible design of the first aspect through a logic circuit or by executing code instructions.
  • the present application provides a computer-readable storage medium, wherein the computer-readable medium stores a program code, and when the program code runs on a computer, the computer executes the method described in the first aspect and any possible design of the first aspect.
  • the present application provides a computer program product, which, when executed on a computer, enables the computer to execute the method described in the first aspect and any possible design of the first aspect.
  • the present application provides a chip comprising a processor, wherein the processor is coupled to a memory and is used to execute a computer program or instructions stored in the memory.
  • the computer program or instructions are executed, the method described in the first aspect and any possible design of the first aspect is implemented.
  • an embodiment of the present application provides a vehicle, comprising a module for implementing the method described in the first aspect and any possible design of the first aspect, or the intrusion detection device described in the second aspect and any possible design of the second aspect.
  • FIG1 is a schematic diagram showing an application scenario to which the present application is applicable
  • FIG2 shows a schematic structural diagram of a vehicle of the present application
  • FIG3 shows a schematic diagram of the process of the intrusion detection method of the present application
  • FIG4 shows a schematic diagram of the process of generating a feature library and training a model in the present application
  • FIG5 shows a schematic diagram of the principle of model training of the present application
  • FIG6 shows a schematic diagram of the process of the intrusion detection method of the present application
  • FIG. 7 shows a schematic diagram of the prediction process based on the location prediction model of the present application.
  • FIG8 shows a schematic diagram of the structure of a communication device of the present application.
  • FIG9 shows a schematic structural diagram of a communication device of the present application.
  • the present invention provides an intrusion detection method, device and vehicle for detecting attacks on vehicle positioning signals to ensure Autonomous driving vehicles drive safely.
  • the method and the device are based on the same technical concept. Since the principles of solving problems by the method and the device are similar, the implementation of the device and the method can refer to each other, and the repeated parts will not be repeated.
  • the terms and/or descriptions between the various embodiments are consistent and can be referenced to each other.
  • the technical features in different embodiments can be combined to form a new embodiment according to their internal logical relationship.
  • the intrusion detection scheme of the present application can be applied to the Internet of Vehicles, such as vehicle to everything (V2X), long term evolution-vehicle (LTE-V), vehicle to vehicle (V2V), etc.
  • V2X vehicle to everything
  • LTE-V long term evolution-vehicle
  • V2V vehicle to vehicle
  • the other devices include but are not limited to: other sensors such as vehicle-mounted terminals, vehicle-mounted control units, vehicle-mounted modules, vehicle-mounted modules, vehicle-mounted components, vehicle-mounted chips, vehicle-mounted units, vehicle-mounted radars or vehicle-mounted cameras.
  • the vehicle can implement the intrusion detection method provided in the embodiment of the present application through the vehicle-mounted terminal, vehicle-mounted control unit, vehicle-mounted module, vehicle-mounted module, vehicle-mounted components, vehicle-mounted chips, vehicle-mounted units, vehicle-mounted radars or vehicle-mounted cameras.
  • the intrusion detection scheme in the embodiment of the present application can also be used in other intelligent terminals with mobile control functions other than vehicles, or be set in other intelligent terminals with mobile control functions other than vehicles, or be set in components of the intelligent terminal.
  • the intelligent terminal can be an intelligent transportation device, an intelligent home device, a robot, etc. And include but are not limited to smart terminals or control units, chips, other sensors such as radars or cameras, and other components in smart terminals.
  • At least one refers to one or more
  • plural refers to two or more.
  • And/or describes the association relationship of associated objects, indicating that three relationships may exist.
  • a and/or B can represent: A exists alone, A and B exist at the same time, and B exists 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 represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c can be single or multiple.
  • ordinal numbers such as “first” and “second” mentioned in the embodiments of the present application are used to distinguish between multiple objects and are not used to limit the priority or importance of multiple objects.
  • FIG1 shows a schematic diagram of an application scenario to which an embodiment of the present application is applicable.
  • a vehicle 100 may be included.
  • the application scenario may also include a cloud server 200, and the vehicle 100 and the cloud server 200 may communicate via a network.
  • the cloud server 200 may be implemented by a virtual machine.
  • the computing platform 150 may include at least one processor 151, which may execute instructions 153 stored in a non-transitory computer-readable medium such as a memory 152. In some embodiments, the computing platform 150 may also be a plurality of computing devices that control individual components or subsystems of the vehicle 100 in a distributed manner.
  • the processor 151 may be any conventional processor, such as a central processing unit (CPU). Alternatively, the processor 151 may also include a graphics processor (GPU), a field programmable gate array (FPGA), a system on chip (SoC), an application-specific integrated circuit (ASIC), or a combination thereof.
  • GPU graphics processor
  • FPGA field programmable gate array
  • SoC system on chip
  • ASIC application-specific integrated circuit
  • the vehicle 100 may be a car, a truck, a motorcycle, a bus, a ship, an airplane, a helicopter, a lawn mower, an amusement vehicle, an amusement park vehicle, construction equipment, a tram, a golf cart, a train, etc., which is not particularly limited in the present embodiment.
  • the vehicle 100 may be a new energy vehicle.
  • FIG. 1 The structure of the vehicle in FIG. 1 should not be understood as limiting the embodiments of the present application.
  • the intrusion detection method of the embodiment of the present application can be implemented by an intrusion detection device, which can be an independent device, a chip or component in the vehicle 100 shown in Figure 1, or a software module, which can be deployed on the relevant on-board equipment of the vehicle 100.
  • the intrusion detection scheme of the embodiment of the present application will be introduced by taking the intrusion detection device as a processor of the computing platform 150 integrated in the aforementioned vehicle 100 as an example.
  • the intrusion detection scheme can also be integrated in other electronic control units (electronic control units, ECU) of the vehicle, such as an intelligent driving domain control unit, or a vehicle control unit (vehicle control unit, VCU), etc.
  • ECU electronicelectronic control units
  • VCU vehicle control unit
  • the embodiment of the present application does not limit the product form or deployment method of the intrusion detection device.
  • the sensing device can obtain the sensing data of the environment in which the vehicle is located, and constantly monitor and collect data about the environment around the vehicle.
  • the sensing device may include at least one sensor in the vehicle's sensing system, and the sensing system may include but is not limited to: a camera, light detection and ranging (LIDAR), millimeter-wave radar (RADAR), ultrasonic radar sensor or other sensor devices.
  • the sensing data obtained by the sensing system may include: the vehicle's current position; An image or video of the surrounding environment, a target object (such as a traffic object) identified based on the image or video, a distance (or interval) between the vehicle and the identified target object, etc.
  • the sensing system may also include a position sensor, which may be, for example, a GPS receiver, which may be used to receive the position information of the vehicle.
  • the sensing device may provide the acquired multiple sensing data to the intrusion detection device.
  • the sensing device may send the multiple sensing data to the intrusion detection device via an in-vehicle communication network (eg, an in-vehicle bus, not shown in the figure).
  • an in-vehicle communication network eg, an in-vehicle bus, not shown in the figure.
  • the intrusion detection device itself may have the function of performing perception processing on the original perception data, may obtain a variety of original perception data from various sensors, and perform perception processing on the obtained multiple original perception data to obtain a variety of feature data.
  • the intrusion detection device may filter out the feature data required for performing intrusion detection analysis from the obtained multiple feature data, for example, including first feature data and second feature data, the first feature data describing information about static traffic objects related to the area where the vehicle is located, and the second feature data describing information about dynamic traffic objects related to the area where the vehicle is located.
  • the intrusion detection device may perform intrusion detection analysis based on the filtered first feature data and second feature data to determine whether an intrusion event has occurred.
  • the intrusion detection device itself may not have the function of perceiving and processing the original perception data, but may obtain the original perception data from various sensors with the help of other devices (such as the fusion perception module of the intelligent driving system, not shown in FIG2 ), and perform perception processing on the original perception data.
  • the fusion perception module can provide the corresponding feature data obtained after the perception processing to the intrusion detection device.
  • the fusion perception module can send the corresponding feature data obtained after the perception processing to the intrusion detection device through the in-vehicle communication network.
  • the intrusion detection device can filter out the feature data required to perform intrusion detection analysis from the multiple feature data obtained, for example, including first feature data and second feature data, the first feature data describing the information of static traffic objects related to the area where the vehicle is located, and the second feature data describing the information of dynamic traffic objects related to the area where the vehicle is located.
  • the intrusion detection device can perform intrusion detection analysis based on the filtered first feature data and second feature data to determine whether an intrusion event has occurred.
  • traffic objects include any traffic participants other than the vehicle in the area where the vehicle is located, including but not limited to the current lane of the vehicle and the lane line of the current lane; other vehicles around the vehicle in the current lane (including the front and rear vehicles); adjacent lanes of the current lane and the lane lines of adjacent lanes; several other vehicles on adjacent lanes; pedestrians in the current lane, adjacent lanes, or road intersections; various traffic facilities: such as crosswalks, road side units (RSUs), traffic lights (such as motor vehicle lights, non-motor vehicle lights, crosswalk lights, direction indicators, lane lights, flashing lights, road and railway level intersection lights, etc.), fences, lighting facilities, sight guidance signs, highway reflectors, highway information boards, etc.; buildings or mountains around the area, etc.
  • RSUs road side units
  • traffic lights such as motor vehicle lights, non-motor vehicle lights, crosswalk lights, direction indicators, lane lights, flashing lights, road and railway level intersection lights, etc.
  • fences lighting facilities, sight guidance signs, highway reflector
  • dynamic traffic objects include dynamic traffic participants such as vehicles and pedestrians, and these dynamic traffic objects will move according to traffic rules over time.
  • Static objects include static traffic participants such as lane lines, traffic facilities, buildings, and mountains. These static traffic objects are in a state of no change (for example, their position remains unchanged) within a certain time range. That is, they are relatively fixed traffic participants with a low frequency of change and will not move over time like dynamic traffic objects.
  • the above-mentioned perception processing may include, but is not limited to, semantic segmentation processing and multi-target detection processing of the original perception data.
  • the characteristic data of different traffic objects around the area where the vehicle is located may be obtained, for example, represented as third characteristic data.
  • the intrusion detection device may filter out the above-mentioned first characteristic data and/or second characteristic data from a variety of third characteristic data, and perform intrusion detection analysis based on the filtered first characteristic data and second characteristic data to determine whether an intrusion event has occurred.
  • an intrusion detection rule may be pre-set in the intrusion detection device, and the intrusion detection rule describes a detection method for an intrusion event associated with the vehicle's service to be detected.
  • the intrusion detection device may use the filtered feature data and the intrusion detection rule to analyze whether an intrusion event associated with the service to be detected has occurred. For example, the intrusion detection rule describes the conditions that different features corresponding to the intrusion event should meet. If the filtered feature data meets the corresponding conditions described in the intrusion detection rule, the intrusion detection device may consider that the intrusion event has occurred. If the filtered feature data does not meet the corresponding conditions described in the intrusion detection rule, the intrusion detection device may consider that the intrusion event has not occurred.
  • the detection method described in the intrusion detection rule may be a method for analyzing whether an intrusion event has occurred by comparing the above-mentioned first feature data or the second feature data with the corresponding reference information.
  • the reference information corresponding to the first feature data may include the vehicle location information obtained by the vehicle positioning component (e.g., a position sensor), and the reference information corresponding to the second feature data may include feature data obtained from a map server (e.g., the number of surrounding vehicles).
  • the first condition includes, for example, that the similarity of the location information obtained by different methods is less than or equal to the first threshold
  • the second condition includes, for example, that the similarity of the number of other surrounding vehicles obtained by different methods is less than or equal to the second threshold.
  • the intrusion detection device can send an alarm message to the vehicle's execution device, and the alarm message can indicate
  • the execution device can receive the warning information from the intrusion detection device, and execute vehicle-related control processing in combination with the warning information to ensure the safe driving of the vehicle.
  • the service to be detected may include the intelligent driving service of the vehicle
  • the execution device may include the intelligent driving system of the vehicle
  • the intrusion event may include an event in which an attacker forges the location information of the vehicle or forges the information of other surrounding vehicles to "cheat" the intelligent driving system of the vehicle.
  • an alarm message may be sent to the intelligent driving system of the vehicle through the vehicle communication network (such as a gateway) to indicate the intrusion event, so that the intelligent driving system can make intelligent driving decisions (such as path planning, vehicle control decisions, etc.) in combination with the alarm information, and send control commands to the ECU related to the vehicle control to reduce or even eliminate the interference caused by the intrusion event to the intelligent driving control decision, assist in controlling the safe driving of the vehicle, and thus ensure the safe driving of the vehicle.
  • the vehicle communication network such as a gateway
  • the execution device may also include a smart cockpit of the vehicle, which may include a central control display, a head-up display device, an audio system, seat vibration, lights, etc.
  • the smart cockpit may receive warning information from the intrusion detection device, and may output the warning information through the central control display, the head-up display device, the audio system, the seat vibration, lights, etc., for reference by the occupants in the vehicle (including the driver or other occupants), thereby assisting the driver of the vehicle to control the safe driving of the vehicle, or enabling other occupants to provide driving assistance to the driver of the vehicle according to the warning information.
  • the intrusion detection device may also send warning information to a peripheral device associated with the vehicle (such as a user's mobile smart device) through a gateway to indicate an intrusion event to the user, so that the user can be informed of the intrusion event in a timely manner and make corresponding solutions.
  • a peripheral device associated with the vehicle such as a user's mobile smart device
  • the intrusion detection device may also send warning information to other vehicle components or associated devices to timely discover and resolve intrusion events, which will not be repeated here.
  • the intrusion detection rules may be pre-saved on a storage medium accessible to the intrusion detection device, which may be a local storage device of the vehicle, or may be a cloud server, which is not limited in the embodiment of the present application.
  • the operator may pre-design at least one intrusion detection rule for different attack methods based on crowdsourcing data and experience.
  • the intrusion detection device may obtain and save at least one intrusion detection rule from a cloud server as needed, or may manually save at least one intrusion detection rule on a local storage device of the vehicle, which is not limited in the embodiment of the present application.
  • S310 The intrusion detection device obtains first characteristic data.
  • the dynamic traffic objects include dynamic traffic participants related to the area where the first vehicle is located, such as other vehicles and pedestrians around the first vehicle.
  • the information of the dynamic traffic objects includes, for example, attribute information of the dynamic traffic objects themselves, such as location information, shape information, size information, etc. of other vehicles.
  • the information of the dynamic traffic objects may also include, for example, information about the dynamic traffic objects relative to the first vehicle, such as distance information and azimuth information of the dynamic traffic objects relative to the first vehicle.
  • the embodiments of the present application do not limit the content of the information of the dynamic traffic objects.
  • an intrusion event may, for example, include an event in which an attacker forges the location information of the vehicle or forges the information of other surrounding vehicles to "cheat" the intelligent driving system of the vehicle.
  • the detection method described in the intrusion detection rule may be a method of comparing the above-mentioned first feature data or the second feature data with their respective corresponding reference information to analyze whether an intrusion event has occurred.
  • the reference information corresponding to the first characteristic data may include the first position information of the first vehicle obtained by the positioning component of the first vehicle, and the first characteristic data can be used to determine the position information of the first vehicle, for example, represented as the second position information.
  • the intrusion detection device can compare the first position information with the second position information to determine whether an intrusion event has occurred.
  • the reference information corresponding to the second characteristic data may include the fourth characteristic data of the first vehicle obtained from the map server, and the fourth characteristic data is information about the surrounding vehicles of the first vehicle provided by the map server.
  • the second characteristic data is information obtained based on perception and used to determine the surrounding vehicles of the first vehicle.
  • the intrusion detection device can be used to compare the second characteristic data with the fourth characteristic data to determine whether an intrusion event has occurred.
  • the intrusion detection device may further perform the following steps:
  • the intrusion detection device sends an alarm message to the intelligent driving system of the first vehicle, where the alarm message indicates the intrusion event.
  • the intelligent driving system can receive the warning information from the intrusion detection device and perform vehicle-related control processing in combination with the warning information to ensure the safe driving of the vehicle.
  • the intelligent driving system can send control commands to the vehicle control-related ECU to enable the vehicle control-related ECU to perform braking, steering, acceleration, deceleration, etc. to assist in controlling the safe driving of the vehicle, thereby ensuring the safe driving of the vehicle.
  • the intrusion detection device can generate a feature library associated with the GPS service, and obtain M position prediction models based on sample training in the feature library.
  • the M position prediction models can be used to predict the second position information based on the first feature data when implementing the above S330.
  • the second position information can be used to compare with the first position information of the first vehicle obtained by the positioning component of the first vehicle to analyze whether an intrusion event has occurred.
  • the generation process of the feature library associated with the GPS service and the corresponding model training process may include the following steps:
  • N sensors associated with the first vehicle collect N types of original perception data, and provide the N types of original perception data to a fusion perception module of the intelligent driving system, where N is greater than or equal to 1.
  • the fusion perception module of the intelligent driving system performs perception processing on the N types of original perception data obtained, such as semantic segmentation processing, multi-target detection processing, etc., and provides N types of third feature data obtained after the perception processing to the intrusion detection device.
  • the third feature data is a feature matrix, including sensor feature vectors and labels, expressed as (X, Y), where X represents the sensor feature vector, and Y represents the label of the identified traffic object, such as street lights, traffic signs, ramps, etc.
  • the intrusion detection device selects and saves the required feature data from the N third feature data according to the tag, that is, generates a feature library associated with the GPS service.
  • the feature data describes information about static traffic objects related to the area where the first vehicle is located, and feature data other than the feature data can be discarded.
  • the feature matrix obtained after screening can be concatenated with the corresponding location information (such as GPS longitude and latitude data) into a new matrix, expressed as (X, Y, P), where X represents the sensor feature vector, Y represents the label of the identified traffic object, and P represents the GPS longitude and latitude data.
  • the feature matrix (X, Y, P) corresponding to different sensor feature vectors can be used to train location prediction models corresponding to different types of sensors.
  • the intrusion detection device trains position prediction models corresponding to different types of sensors based on feature matrices (X, Y, P) corresponding to different sensors.
  • the M position prediction models trained include supervised learning models corresponding to at least one of the following sensors: a camera, a laser radar sensor, a millimeter wave radar sensor, or an ultrasonic radar sensor.
  • the supervised learning model as a deep learning model as an example, during training, the feature matrix corresponding to different types of sensors obtained after screening and splicing can be input into the corresponding initial deep learning model.
  • position prediction models corresponding to different types of sensors can be obtained, such as a position prediction model corresponding to a camera, a position prediction model corresponding to a laser radar sensor, a position prediction model corresponding to a millimeter wave radar sensor, or an ultrasonic radar sensor.
  • Each of the M position prediction models is used to predict a position information based on feature data obtained from a sensor.
  • the screening and training process can be implemented through an interface call, similar to the get_models_from_sensors(sensor_name_list) interface, which can be used to train a deep model based on the road condition data identified from the sensor as input and output the trained model. Similar to sensor_id_list, it represents the sensor ID list of the data input participating in the model training.
  • the implementation of this interface is similar to the formal description of the logic as follows:
  • t1, t2, t3, and t4 represent the API interfaces of the deep learning models to be trained corresponding to different sensors
  • S1, S2, S3, and S4 represent the feature matrices provided by different types of sensors to the corresponding deep learning models
  • Xi represents the input feature vector corresponding to the sensor
  • Yi represents the label vector
  • M1, M2, M3, and M4 represent the trained deep learning models respectively.
  • ti will filter the input feature vector according to the label vector Yi, so as to filter out the required information Xi describing the static traffic object, and discard the unnecessary feature vectors, which will not participate in the model training process.
  • S405 Incrementally synchronize the M location prediction models obtained through training with the model data of the cloud server as needed.
  • model data for location prediction stored on the cloud server is updated to the model data corresponding to the M location prediction models obtained through training.
  • the method shown in FIG. 3 can be implemented using the M position prediction models. As shown in FIG. 6 , the method may include the following steps:
  • N sensors associated with the first vehicle collect N types of original perception data and provide the N types of original perception data to a fusion perception module of the intelligent driving system, where N is greater than or equal to 1.
  • the fusion perception module of the intelligent driving system performs perception processing on the N types of original perception data obtained, such as semantic segmentation processing, multi-target detection processing, etc., and provides N types of third feature data obtained after the perception processing to the intrusion detection device.
  • the above prediction process can also be implemented by calling an interface, similar to the get_pos_from_sensors(sensor_name_list) interface, which means that the road condition data identified by the sensor is used as input data, input into the pre-trained deep learning model, and outputs the corresponding GPS position prediction value. Similar to sensor_id_list, it represents the sensor ID list involved in the position calculation.
  • the implementation of this interface is similar to the formal description of the logic as follows:
  • the intrusion detection device obtains the fourth characteristic data of the first vehicle from the map server, and analyzes the difference between the fourth characteristic data and the second characteristic data.
  • the intrusion detection device can use a similar get_pos_range_cars(longitude, latitude, range, time) interface provided by a map service provider to obtain the number of other vehicles around the current location of the first vehicle.
  • Longitude and latitude represent the current real-time longitude and latitude information of the first vehicle, which is GPS data received by the GPS signal receiving device.
  • Range represents the perception range corresponding to the longitude and latitude, such as a range of 20m around the first vehicle, and time represents the request time.
  • the information returned by the map server is a collection of longitude and latitude information, which is expressed as follows:
  • s1 ⁇ (x1,y1),(x2,y2) ⁇ , where x and y represent the longitude and latitude information respectively.
  • the intrusion detection device can use a similar get_pos_range_cars interface to implement a similar get_cars_from_map_service(longitude, latitude, range, time, service_id_list) interface to obtain information about other vehicles around the first vehicle from the map server.
  • the service_id_list represents the ID list of the map service provider.
  • the get_cars_from_map_service(longitude,latitude,range,time,service_id_list) interface is called periodically in time series.
  • the time value is [202303020605,202303020606,202303020607]
  • the get_cars_from_sensors(range) interface is called periodically in time series to indicate the number of other vehicles around the first vehicle identified based on the original sensing data, i.e., the second feature data.
  • the sliding window algorithm used in the above S605 and S606 may include but is not limited to a piecewise linear representation (PLR) algorithm, a dynamic time warping (DTW) algorithm, etc., and the embodiment of the present application does not limit the implementation of this algorithm.
  • the similarity algorithm may include but is not limited to a Euclidean distance algorithm, a cosine similarity algorithm, a Minkowski distance algorithm, etc., and the embodiment of the present application does not limit the implementation of this algorithm.
  • the intrusion detection device determines that GPS spoofing exists based on the results of S604 and S605, that is, an intrusion event exists.
  • the intrusion detection device sends an alarm message to the intelligent driving system, where the alarm message indicates the intrusion event.
  • the intelligent driving system can implement subsequent control processing in combination with the alarm message to ensure safe driving of the vehicle.
  • the intrusion detection device sends an alarm message to the smart cockpit, and the alarm message indicates the intrusion event.
  • the smart cockpit can receive the alarm message from the intrusion detection device, and can output the alarm message through the central control display, head-up display device, audio system, seat vibration, lighting, etc., for reference by the occupants in the vehicle (including the driver or other occupants), thereby assisting the vehicle driver to control the safe driving of the vehicle, or enabling other occupants to provide driving assistance to the vehicle driver according to the alarm message.
  • the vehicle's various sensors are used to obtain specific static features of the surrounding environment and road conditions for intrusion detection to prevent attackers from blocking the GPS signal receiving device and then forging GPS signals for counterfeiting attacks.
  • the vehicle's various sensors are used to obtain specific dynamic features of the surrounding environment and road conditions for intrusion detection to prevent attackers from forging GPS signals based on locations with similar environmental features.
  • the GPS signal detection features do not come from the GPS signal itself, and can effectively detect a wide range of GPS signal counterfeiting attacks. At the same time, it only needs to rely on the sensor equipment and computing equipment already installed by the autonomous driving function, without the need for Relying on other hardware support will not incur additional hardware costs.
  • the above-mentioned intrusion detection method can also be used as a sub-function of other functions of the vehicle.
  • the above-mentioned method is implemented to analyze whether an intrusion event has occurred. If an intrusion event has occurred, the GPS signal received by the GPS signal receiving device cannot be directly used as real-time positioning information for recommendation, so as to avoid recommending wrong items to the user.
  • the above-mentioned location prediction value identified based on the original perception data of multiple sensors can be temporarily used as reliable location information for recommendation to the user. After the intrusion event is resolved, the GPS signal received by the GPS signal receiving device can be reused as real-time positioning information for recommendation.
  • the embodiment of the present application also provides an intrusion detection device for executing the method executed by the intrusion detection device in the above method embodiment.
  • the relevant features can be found in the above method embodiment and will not be repeated here.
  • the intrusion detection device 800 may include: a first acquisition unit 801, used to acquire first characteristic data, the first characteristic data describing information of static traffic objects related to the area where the first vehicle is located; a second acquisition unit 802, used to acquire second characteristic data, the second characteristic data describing information of dynamic traffic objects related to the area where the first vehicle is located; a determination unit 803, used to determine the occurrence of an intrusion event based on the first characteristic data, the second characteristic data and an intrusion detection rule, the intrusion detection rule describing a detection method of an intrusion event associated with the service to be detected of the first vehicle; a communication unit 804, used to send an alarm message to the intelligent driving system of the first vehicle, the alarm message indicating the intrusion event.
  • a first acquisition unit 801 used to acquire first characteristic data, the first characteristic data describing information of static traffic objects related to the area where the first vehicle is located
  • a second acquisition unit 802 used to acquire second characteristic data, the second characteristic data describing information of dynamic traffic objects related to the area where the first vehicle is located
  • 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 central processing unit (CPU) or a microprocessor, and the memory is a memory inside the device or a memory outside the device.
  • CPU central processing unit
  • microprocessor a microprocessor
  • the units in the device may be implemented in the form of hardware circuits, and the functions of some or all of the units may be implemented by designing the hardware circuits, and the hardware circuits may be understood as one or more processors; for example, in one implementation, the hardware circuit is an application-specific integrated circuit (ASIC), and the functions of some or all of the above units may be implemented by designing the logical relationship of the components in the circuit; for another example, in another implementation, the hardware circuit may be implemented by a programmable logic device (PLD), and a field programmable gate array (FPGA) may be used as an example, which may include a large number of logic gate circuits, and the connection relationship between the logic gate circuits may be configured by a configuration file, so as to implement the functions of some or all of the above units. All units of the above devices may be implemented in the form of a processor calling software, or in the form of hardware circuits, or in part by a processor calling software, and the rest by hardware circuits.
  • ASIC application-specific integrated circuit
  • FPGA field programm
  • the processor is a circuit with signal processing capability.
  • the processor may be a circuit with instruction reading and running capability, such as a CPU, a microprocessor, a graphics processing unit (GPU) (which may be understood as a microprocessor), or a digital signal processor (DSP), etc.; 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 may be 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 hardware circuit configuration may be understood as the process of the processor loading instructions to implement the functions of some or all of the above units.
  • it may also be a hardware circuit designed for artificial intelligence, which may 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.
  • NPU neural network processing unit
  • TPU tensor processing unit
  • 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 SoC.
  • the 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 CPU and FPGA, CPU and artificial intelligence processor, CPU and GPU, etc.
  • the device 900 shown in Fig. 9 includes at least one processor 910 and a communication interface 930.
  • a memory 920 may also be included.
  • the processor 910 may perform data transmission through the communication interface 930 when communicating with other devices.
  • the processor 910 in FIG. 9 can call the computer-executable instructions stored in the memory 920 so that the device 900 can execute any of the above method embodiments.
  • the present application also provides a vehicle, the structure of which is shown in FIG. 2 , wherein the intrusion detection device has the structure shown in FIG. 8 or FIG. 9 .
  • An embodiment of the present application also relates to a chip system, which includes a processor for calling a computer program or computer instructions stored in a memory so that the processor executes the method of any of the above embodiments.
  • the processor may be coupled to the memory through an interface.
  • the chip system may also directly include a memory, in which a computer program or computer instructions are stored.
  • the memory may be a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memory.
  • the nonvolatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory.
  • the volatile memory may be a random access memory (RAM), which is used as an external cache.
  • RAM direct rambus RAM
  • SRAM static RAM
  • DRAM dynamic RAM
  • SDRAM synchronous DRAM
  • DDR SDRAM double data rate SDRAM
  • ESDRAM enhanced SDRAM
  • SLDRAM synchlink DRAM
  • DR RAM direct rambus RAM
  • An embodiment of the present application also relates to a processor, which is used to call a computer program or computer instruction stored in a memory so that the processor executes the method described in any of the above embodiments.
  • the processor is an integrated circuit chip with signal processing capabilities.
  • the processor can be an FPGA, a general-purpose processor, a DSP, an ASIC or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, a SoC, a CPU, a network processor (network processor, NP), a microcontroller (micro controller unit, MCU), a PLD or other integrated chip, which can implement or execute the methods, steps and logic block diagrams disclosed in the embodiment of the present application.
  • the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.
  • the steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to perform, or the hardware and software modules in the decoding processor can be combined to perform.
  • the software module can 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 a memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.
  • an embodiment of the present application provides a computer-readable storage medium, which stores a program code.
  • the program code runs on the computer, the computer executes the above method embodiment.
  • an embodiment of the present application provides a computer program product.
  • the computer program product When the computer program product is run on a computer, the computer is enabled to execute the above method embodiment.
  • the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware.
  • the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
  • These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.
  • These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Human Computer Interaction (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种入侵检测方法、装置以及车辆，该方法包括：获取第一特征数据，所述第一特征数据描述第一车辆所在区域相关的静态交通对象的信息；获取第二特征数据，所述第二特征数据描述所述第一车辆所在区域相关的动态交通对象的信息；根据所述第一特征数据、所述第二特征数据和入侵检测规则，确定发生入侵事件，所述入侵检测规则描述所述第一车辆的待检测业务关联的入侵事件的检测方法；向所述第一车辆的智能驾驶系统发送告警信息，所述告警信息指示所述入侵事件。该方法可用于检测对车辆定位信号的攻击，以保障自动驾驶车辆安全行驶。

Description

一种入侵检测方法、装置以及车辆
相关申请的交叉引用
本申请要求在2023年11月14日提交中华人民共和国国家知识产权局、申请号为202311519004.6、申请名称为“一种入侵检测方法、装置以及车辆”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及车辆技术领域，特别涉及一种入侵检测方法、装置以及车辆。
背景技术
目前，车辆是由内部的自动驾驶系统根据地图以及车辆环境信息进行路径计算来实现自动驾驶的。自动驾驶车辆的安全行驶既依靠对周边障碍物的感知，也依赖全球卫星导航系统对车辆在地图上进行厘米级定位，若出现定位错误的情况，则会直接导致车辆冲出路面或驶向错误方向，导致车辆事故。
攻击者通过伪造定位信号，来“欺骗”自动驾驶车辆的智能驾驶系统，是车辆定位错误造成车辆失控的原因之一。因此，如何检测这些攻击手段，以保障自动驾驶车辆安全行驶，仍为亟需解决的重要问题。
发明内容
本申请提供一种入侵检测方法、装置以及车辆，用于检测对车辆定位信号的攻击，以保障自动驾驶车辆安全行驶。
第一方面，本申请提供了一种入侵检测方法，该方法可由入侵检测装置执行，该入侵检测装置可以是独立设备，也可以是设备中的芯片或部件，还可以是软件。该入侵检测装置可以部署在车辆上，该车辆可以是处于完全人工驾驶模式的车辆，或完全自动驾驶模式的车辆，或者，车辆可以配置为部分地自动驾驶模式的车辆。其中，部分地自动驾驶模式的车辆，例如是指，车辆可以在处于自动驾驶模式中同时控制自身，并且可以通过人为操作来确定车辆以及周边环境的当前状态，确定周边环境中的至少一个其它车辆的可能行为，并基于所确定的信息控制车辆。在车辆处于完全自动驾驶模式中时，可以将车辆置为在没有和人交互的情况下操作。本申请对该入侵检测装置的产品形态以及部署方式不做限定。
该方法包括：获取第一特征数据，所述第一特征数据描述第一车辆所在区域相关的静态交通对象的信息；获取第二特征数据，所述第二特征数据描述所述第一车辆所在区域相关的动态交通对象的信息；根据所述第一特征数据、所述第二特征数据和入侵检测规则，确定发生入侵事件，所述入侵检测规则描述所述第一车辆的待检测业务关联的入侵事件的检测方法；向所述第一车辆的智能驾驶系统发送告警信息，所述告警信息指示所述入侵事件。所述待检测业务包括定位业务。本申请中，自动驾驶及智能驾驶，均是指车辆在智能驾驶系统的控制下所执行的驾驶行为，且不予以区分。
通过上述方法，入侵检测装置可以根据静态交通对象的信息以及动态交通对象的信息来进行入侵检测，以防止攻击者对车辆的位置传感器进行阻塞后再伪造位置信息进行仿冒攻击，该方法中，进行入侵检测所依赖的特征数据并非来自于位置传感器，可有效检测大范围的位置信号仿冒攻击，同时不会带来额外的硬件成本。
结合第一方面，在一种可能的设计中，所述获取第一特征数据，包括：从N种第三特征数据中筛选所述第一特征数据，其中，所述N种第三特征数据为对N种感知数据进行感知处理后得到的，所述N种感知数据为通过与所述第一车辆关联的N个传感器采集到的原始感知数据，N大于等于1，所述N个传感器不包括位置传感器。
通过上述方法，进行入侵检测所依赖的特征数据并非来自于位置传感器而是来自于车辆上装配的其他传感设备，可有效检测大范围的位置信号仿冒攻击。
结合第一方面，在一种可能的设计中，所述感知处理包括多目标检测处理，所述N种第三特征数据包括识别到的交通对象的标签，所述从N种第三特征数据中筛选所述第一特征数据，包括：根据所述标签，从N种第三特征数据中筛选所述第一特征数据，其中，所述第一特征数据关联的标签用于描述所述静态交通对象。
通过上述方法，可以利用车辆的多种传感器获取周围环境和路况的特定静态特征来进行入侵检测， 以防止攻击者对全球定位系统(global positioning system，GPS)信号接收装置进行阻塞后再伪造GPS信号进行仿冒攻击。
结合第一方面，在一种可能的设计中，所述动态交通对象包括所述第一车辆周围第一范围内的其他车辆，所述获取第二特征数据，包括：根据N种第三特征数据获取所述第二特征数据，其中，所述第二特征数据包括所述其他车辆的数量，所述N种第三特征数据为对N种感知数据进行感知处理后得到的，所述N种感知数据为通过与所述第一车辆关联的N个传感器采集到的原始感知数据，N大于等于1。
通过上述方法，可以利用车辆的多种传感器获取周围环境和路况的特定动态特征来进行入侵检测，以防止攻击者基于具有相似环境特征的位置伪造GPS信号。
结合第一方面，在一种可能的设计中，所述待检测业务关联所述第一车辆的位置信息，所述入侵检测规则包括分析通过不同方式获得的信息的差异，所述方法还包括：通过所述第一车辆的定位部件获取所述第一车辆的第一位置信息；从地图服务器获取所述第一车辆的第四特征数据，所述第四特征数据为所述地图服务器提供的所述第一车辆的周边车辆的信息；所述根据所述第一特征数据、所述第二特征数据和入侵检测规则，确定发生入侵事件，包括：当根据所述第一特征数据确定的位置信息和所述第一位置信息的差异满足第一条件，以及所述第二特征数据和所述第四特征数据的差异满足第二条件时，确定发生定位入侵事件。
通过上述方法，GPS信号检测特征不来自于GPS信号本身，可以有效检测大范围的GPS信号仿冒攻击，同时只需依赖自动驾驶车辆已装配的传感设备和计算设备即可，无需依赖其它硬件支持，也不会带来额外的硬件成本。
结合第一方面，在一种可能的设计中，所述第一条件包括：所述根据所述第一特征数据确定的位置信息与所述第一位置信息的相似度小于或等于第一阈值，所述方法还包括：将所述第一特征数据中关联于不同传感器的特征数据输入M个位置预测模型，获得M个位置预测值，M大于1；根据所述M个位置预测值加权得到所述根据所述第一特征数据确定的位置信息；其中，所述M个位置预测模型中的每个位置预测模型用于对基于一种传感器获得的特征数据预测一个位置信息，所述M个位置预测模型是根据GPS业务相关联的特征库中的样本训练得到的。示例性地，所述M个位置预测模型包括以下至少一种传感器对应的有监督学习模型：摄像头、激光雷达传感器、毫米波雷达传感器或者超声波雷达传感器。
通过上述方法，可以预训练多个位置预测模型，以便根据不同传感器采集到的原始感知数据，预测车辆的位置信息，并将预测到的位置信息与位置传感器采集到的位置信息进行比对，以确定是否存在GPS欺骗。
结合第一方面，在一种可能的设计中，所述第四特征数据包括所述第一车辆周围第一范围内的其他车辆的参考数量；所述第二条件包括：所述第二特征数据指示的其他车辆的数量与所述第四特征数据指示的其他车辆的参考数量的相似度小于或等于第二阈值。
第二方面，本申请提供了一种入侵检测装置，包括：第一获取单元，用于获取第一特征数据，所述第一特征数据描述第一车辆所在区域相关的静态交通对象的信息；第二获取单元，用于获取第二特征数据，所述第二特征数据描述所述第一车辆所在区域相关的动态交通对象的信息；确定单元，用于根据所述第一特征数据、所述第二特征数据和入侵检测规则，确定发生入侵事件，所述入侵检测规则描述所述第一车辆的待检测业务关联的入侵事件的检测方法；通信单元，用于向所述第一车辆的智能驾驶系统发送告警信息，所述告警信息指示所述入侵事件。
结合第二方面，在一种可能的设计中，所述第一获取单元用于：从N种第三特征数据中筛选所述第一特征数据，其中，所述N种第三特征数据为对N种感知数据进行感知处理后得到的，所述N种感知数据为通过与所述第一车辆关联的N个传感器采集到的原始感知数据，N大于等于1。
结合第二方面，在一种可能的设计中，所述感知处理包括多目标检测处理，所述N种第三特征数据包括识别到的交通对象的标签，所述第一获取单元从N种第三特征数据中筛选所述第一特征数据，包括：根据所述标签，从N种第三特征数据中筛选所述第一特征数据，其中，所述第一特征数据关联的标签用于描述所述静态交通对象。
结合第二方面，在一种可能的设计中，所述动态交通对象包括所述第一车辆周围第一范围内的其他车辆，所述第二获取单元用于：根据N种第三特征数据获取所述第二特征数据，其中，所述第二特征 数据包括所述其他车辆的数量，所述N种第三特征数据为对N种感知数据进行感知处理后得到的，所述N种感知数据为通过与所述第一车辆关联的N个传感器采集到的原始感知数据，N大于等于1。
结合第二方面，在一种可能的设计中，所述待检测业务关联所述第一车辆的位置信息，所述入侵检测规则包括分析通过不同方式获得的信息的差异，所述确定单元还用于：通过所述第一车辆的定位部件获取所述第一车辆的第一位置信息；从地图服务器获取所述第一车辆的第四特征数据，所述第四特征数据为所述地图服务器提供的所述第一车辆的周边车辆的信息；当根据所述第一特征数据确定的位置信息和所述第一位置信息的差异满足第一条件，以及所述第二特征数据和所述第四特征数据的差异满足第二条件时，确定发生入侵事件。
结合第二方面，在一种可能的设计中，所述第一条件包括：所述根据所述第一特征数据确定的位置信息与所述第一位置信息的相似度小于或等于第一阈值，所述确定单元用于：将所述第一特征数据中关联于不同传感器的特征数据输入M个位置预测模型，获得M个位置预测值，M大于1；根据所述M个位置预测值加权得到所述根据所述第一特征数据确定的位置信息；其中，所述M个位置预测模型中的每个位置预测模型用于对基于一种传感器获得的特征数据预测一个位置信息，所述M个位置预测模型是根据GPS业务相关联的特征库中的样本训练得到的。
结合第二方面，在一种可能的设计中，所述M个位置预测模型包括以下至少一种传感器对应的有监督学习模型：摄像头、激光雷达传感器、毫米波雷达传感器或者超声波雷达传感器。
结合第二方面，在一种可能的设计中，所述第四特征数据包括所述第一车辆周围第一范围内的其他车辆的参考数量；所述第二条件包括：所述第二特征数据指示的其他车辆的数量与所述第四特征数据指示的其他车辆的参考数量的相似度小于或等于第二阈值。
第三方面，本申请提供了一种通信装置，包括至少一个处理器和接口电路，所述接口电路用于为所述至少一个处理器提供数据或者代码指令，所述至少一个处理器用于通过逻辑电路或执行代码指令实现如上述第一方面以及第一方面任一可能设计所述的方法。
第四方面，本申请提供了一种计算机可读存储介质，所述计算机可读介质存储有程序代码，当所述程序代码在计算机上运行时，使得计算机执行如上述第一方面以及第一方面任一可能设计所述的方法。
第五方面，本申请提供了一种计算机程序产品，当所述计算机程序产品在计算机上运行时，使得所述计算机执行如上述第一方面以及第一方面任一可能设计所述的方法。
第六方面，本申请提供了一种芯片，包括处理器，所述处理器与存储器耦合，用于执行所述存储器中存储的计算机程序或指令，当所述计算机程序或指令被执行时，实现如上第一方面以及第一方面任一可能设计所述的方法。
第七方面，本申请实施例提供了一种车辆，包括用于实现如上第一方面以及第一方面任一可能设计所述的方法的模块，或第二方面以及第二方面任一可能设计所述的入侵检测装置。
本申请实施例在上述各方面提供的实现的基础上，还可以进行进一步组合以提供更多实现。
上述第二方面至第七方面中任一方面中的任一可能实现方式可以达到的技术效果，可以相应参照上述第一方面中任一方面中的任一可能实现方式可以达到的技术效果描述，重复之处不予论述。
附图说明
图1示出了本申请适用的应用场景的示意图；
图2示出了本申请的车辆的结构示意图；
图3示出了本申请的入侵检测方法的流程示意图；
图4示出了本申请的生成特征库和训练模型的流程示意图；
图5示出了本申请的模型训练的原理示意图；
图6示出了本申请的入侵检测方法的流程示意图；
图7示出了本申请的基于位置预测模型的预测过程示意图；
图8示出了本申请的通信装置的结构示意图；
图9示出了本申请的通信设备的结构示意图。
具体实施方式
本申请实施例提供了一种入侵检测方法、装置以及车辆，用于检测对车辆定位信号的攻击，以保障 自动驾驶车辆安全行驶。其中，方法和装置是基于同一技术构思的，由于方法及装置解决问题的原理相似，因此装置与方法的实施可以相互参见，重复之处不再赘述。并且，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。
本申请的入侵检测方案可以应用于车联网，如车-万物(vehicle to everything，V2X)、车间通信长期演进技术(long term evolution-vehicle，LTE-V)、车辆-车辆(vehicle to vehicle，V2V)等。例如可以应用于车辆，或者车辆中的其它装置。该其它装置包括但不限于：车载终端、车载控制单元、车载模块、车载模组、车载部件、车载芯片、车载单元、车载雷达或车载摄像头等其他传感器，车辆可通过该车载终端、车载控制单元、车载模块、车载模组、车载部件、车载芯片、车载单元、车载雷达或车载摄像头，实施本申请实施例提供的入侵检测方法。当然，本申请实施例中的入侵检测方案还可以用于除了车辆之外的其它具有移动控制功能的智能终端，或设置在除了车辆之外的其它具有移动控制功能的智能终端中，或设置于该智能终端的部件中。该智能终端可以为智能运输设备、智能家居设备、机器人等。且包括但不限于智能终端或智能终端内的控制单元、芯片、雷达或摄像头等其它传感器、以及其它部件等。
本申请实施例中，“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：单独存在A，同时存在A和B，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和b,a和c,b和c,或a和b和c，其中a,b,c可以是单个，也可以是多个。
以及，除非有特别说明，本申请实施例提及“第一”、“第二”等序数词是用于对多个对象进行区分，不用于限定多个对象的优先级或者重要程度。
下面结合附图，对本申请实施例进行介绍。
图1示出了本申请实施例适用的应用场景的示意图。在该应用场景中，可以包括车辆100。在一种可能的实现方式中，该应用场景中还可以包括云端服务器200，车辆100和云端服务器200可以通过网络通信。在一个实施例中，该云端服务器200可以通过虚拟机来实现。
车辆100的部分或所有功能受计算平台150(或称为计算机系统)控制。计算平台150可包括至少一个处理器151，处理器151可以执行存储在例如存储器152这样的非暂态计算机可读介质中的指令153。在一些实施例中，计算平台150还可以是采用分布式方式控制车辆100的个体组件或子系统的多个计算设备。处理器151可以是任何常规的处理器，诸如中央处理单元(central processing unit，CPU)。替选地，处理器151还可以包括诸如图像处理器(graphic processing unit，GPU)，现场可编程门阵列(field programmable gate array，FPGA)、片上系统(system on chip，SoC)、专用集成芯片(application-specific integrated circuit，ASIC)或它们的组合。
可选地，上述车辆100可以为轿车、卡车、摩托车、公共汽车、船、飞机、直升飞机、割草机、娱乐车、游乐场车辆、施工设备、电车、高尔夫球车、火车等，本申请实施例不做特别的限定。在一种可能的实现方式中，该车辆100可以为新能源汽车。
图1中车辆的结构不应理解为对本申请实施例的限制。
本申请实施例的入侵检测方法可由入侵检测装置实现，该入侵检测装置可以为独立装置，也可以为图1所示的车辆100中的芯片或部件，还可以是软件模块，可以部署在车辆100的相关车载设备上。下文中，为了便于理解和描述，将以入侵检测装置为集成在前述车辆100中的计算平台150的处理器为例对本申请实施例的入侵检测方案进行介绍。在其它实施例中，该入侵检测方案还可以集成在车辆的其它电子控制单元(electronic control unit，ECU)，例如智能驾驶域控制单元、或者整车控制单元(vehicle control unit，VCU)等，本申请实施例对该入侵检测装置的产品形态或者部署方式均不做限定。
图2示出了车辆100的一种可能的结构示意图。车辆100可以包括感知装置、入侵检测装置以及执行装置。
参阅图2所示，感知装置可以获取车辆所在环境的感知数据，时刻保持对车辆周围环境的监测与数据采集。其中，感知装置可以包括车辆的传感系统中的至少一个传感器，该传感系统可以包括但不限于：摄像头(camera)、激光雷达(light detection and ranging，LIDAR)、毫米波雷达(millimeter-wave radar，RADAR)、超声波雷达传感器或者其它传感器件。通过传感系统获取的感知数据可以包括：车辆当前在 所处的周围环境的图像或视频、基于该图像或视频识别到的目标物体(例如交通对象)、车辆与识别到的目标物体之间的距离(或者间隔)等。可选的，该传感系统还可以包括位置传感器，该位置传感器例如可以是GPS接收器，可用于接收车辆的位置信息。
感知装置可以将获取到的多种感知数据提供至入侵检测装置。例如，感知装置可以通过车内通信网络(例如车载总线，图中未示出)将多种感知数据发送至入侵检测装置。
在一个示例中，入侵检测装置自身可以具备对原始感知数据进行感知处理的功能，可以从各种传感器获取多种原始感知数据，并对获取到的多种原始感知数据进行感知处理获得多种特征数据。入侵检测装置可以从所获得的多种特征数据中筛选出执行入侵检测分析所需要的特征数据，例如包括第一特征数据和第二特征数据，该第一特征数据描述车辆所在区域相关的静态交通对象的信息，该第二特征数据描述车辆所在区域相关的动态交通对象的信息。入侵检测装置可以根据筛选出的第一特征数据和第二特征数据来进行入侵检测分析，以确定是否发生入侵事件。
在另一个示例中，入侵检测装置自身可以不具备对原始感知数据进行感知处理的功能，而是可以借助于其它装置(例如智能驾驶系统的融合感知模块，图2未示出)从各种传感器获取原始感知数据，并对原始感知数据进行感知处理。融合感知模块可以将经过感知处理后获得相应特征数据提供至入侵检测装置。例如融合感知模块可以通过车内通信网络将经过感知处理后获得相应特征数据发送至入侵检测装置。入侵检测装置可以从所获得的多种特征数据中筛选出执行入侵检测分析所需的特征数据，例如包括第一特征数据和第二特征数据，该第一特征数据描述车辆所在区域相关的静态交通对象的信息，该第二特征数据描述车辆所在区域相关的动态交通对象的信息。入侵检测装置可以根据筛选出的第一特征数据和第二特征数据来进行入侵检测分析，以确定是否发生入侵事件。
本申请实施例中，交通对象包括车辆所在区域中除自车以外的任何交通参与者，包括但不限于自车的当前车道、及当前车道的车道线；当前车道中本车的周边其他车辆(包括前车和后车)；当前车道的相邻车道、及相邻车道的车道线；相邻车道上的若干个其他车辆；当前车道、或相邻车道、或道路交汇口的行人；各种交通设施：例如人行横道线、路侧单元(road side unit，RSU)、交通信号灯(例如机动车信号灯、非机动车信号灯、人行横道信号灯、方向指示灯、车道信号灯、闪光灯信号灯、道路与铁道平面交叉道路信号灯等)、栅栏、照明设施、视线诱导标志、公路反射镜、公路情报板等；区域周围的建筑物或者山体等。其中，动态交通对象包括车辆、行人等动态的交通参与者，这些动态交通对象会随着时间推移，根据交通规则移动。静态对象包括车道线、交通设施、建筑物、山体等静态的交通参与者，这些静态交通对象在一定时间范围内处于不发生变化的状态(例如位置不变)，即相对固定、变化频率小的交通参与者，不会像动态交通对象那样随着时间移动。
上述感知处理例如可以包括但不限于对原始感知数据进行语义分割处理、多目标检测处理等。经过感知处理，可以获取到车辆所在区域周围的不同交通对象的特征数据，例如表示为第三特征数据。入侵检测装置可以从多种第三特征数据中筛选出上述第一特征数据和/或第二特征数据，并根据筛选出的第一特征数据和第二特征数据进行入侵检测分析，以确定是否发生入侵事件。
其中，入侵检测装置中可以预置有入侵检测规则，该入侵检测规则描述车辆的待检测业务关联的入侵事件的检测方法。入侵检测装置可以利用筛选出的特征数据以及该入侵检测规则，来分析是否发生待检测业务关联的入侵事件。例如，入侵检测规则描述入侵事件对应的不同特征所应满足的条件，若筛选出的特征数据满足入侵检测规则描述的相应条件，则入侵检测装置可以认为发生该入侵事件。若筛选出的特征数据未满足入侵检测规则描述的相应条件，则入侵检测装置可以认为未发生该入侵事件。
作为示例，入侵检测规则所描述的检测方法可以是将上述第一特征数据或者第二特征数据分别与各自对应的参考信息进行比较来分析是否发生入侵事件的方法。其中，第一特征数据对应的参考信息可以包括通过车辆的定位部件(例如位置传感器)获取到的车辆位置信息，第二特征数据对应的参考信息可以包括从地图服务器获取到的特征数据(例如周边车辆的数量)，若根据第一特征数据确定的位置信息与通过车辆的定位部件获取到的车辆位置信息之间的差异满足第一条件、以及若第二特征数据和从地图服务器获取到的特征数据之间的差异满足第二条件，则可以确定发生入侵事件。其中，第一条件例如包括：通过不同方式获得的位置信息的相似度小于或等于第一阈值，第二条件例如包括通过不同方式获得的周边其他车辆的数量的相似度小于或等于第二阈值。下文中将结合方法实施例对此分析过程进行详细介绍，在此暂不赘述。
在确定发生入侵事件后，入侵检测装置可以向车辆的执行装置发送告警信息，该告警信息可以指示 该入侵事件。相应地，执行装置可以接收来自入侵检测装置的告警信息，并结合该告警信息执行车辆相关的控制处理，以保障车辆的安全行驶。
示例性地，待检测业务可以包括车辆的智能驾驶业务，执行装置可以包括车辆的智能驾驶系统，入侵事件可以包括攻击者伪造车辆的位置信息或者伪造周边其他车辆的信息来“欺骗”车辆的智能驾驶系统的事件。若入侵检测装置根据所获得的特征数据进行分析后确定发生入侵事件，可以通过车辆通信网络(例如网关)向车辆的智能驾驶系统发送告警信息来指示该入侵事件，以便智能驾驶系统可以结合告警信息进行智能驾驶决策(例如路径规划、车控决策等)，并向车控相关的ECU发送控制命令，以减少甚至是排除入侵事件对智能驾驶控制决策造成的干扰，辅助控制车辆安全行车，从而保障车辆的安全行驶。
其中，车控相关的ECU可以包括电机控制单元(motor control unit，MCU)，MCU例如可以包括前轴电机控制单元(表示为MCU_F)或者后轴电机控制单元(表示为MCU_R)。或者例如，车控相关的ECU可以包括车身电子稳定性控制系统(electronic stability controller，ESC)。或者例如，车控相关的ECU可以包括车辆的制动系统，具体例如是集成电动制动(integrated power brake，IPB)，该IPB例如可以集成防抱死系统(antilock brake system，ABS)、ESP、驱动轮防滑系统(acceleration slip regulation，ASR)(又称为牵引力控制系统)等，本申请实施例对此车控相关的ECU的具体实现不做限定。
在一种可选的实施方式中，该执行装置还可以包括车辆的智能座舱，该智能座舱可以包括车辆的中控显示器、抬头显示器件、音响系统、座椅震动、灯光等。该智能座舱可以接收来自入侵检测装置的告警信息，并可以通过中控显示器、抬头显示器件、音响系统、座椅震动、灯光等输出该告警信息，以供车辆内乘员(包括驾驶员或者其他乘员)参考，从而辅助车辆驾驶员控制车辆安全行驶，或者使得其他乘员根据该告警信息为车辆驾驶员提供驾驶辅助。
在另一种可选的实施方式中，入侵检测装置还可以通过网关向车辆关联的外围设备(例如用户的可移动智能设备)发送告警信息，以向用户指示入侵事件，使得用户可以及时获知此入侵事件，并做出相应的解决方案。此处对执行装置或外围设备的介绍为示例而非任何限定，在其它实施例中，入侵检测装置还可以将告警信息发送至其它车辆部件或者关联设备，以便及时发现和解决入侵事件，在此不再赘述。
本申请实施例中，入侵检测规则可以预先保存在入侵检测装置可访问的存储介质上，该存储介质可以是车辆本地的存储设备，或者可以是云端服务器，本申请实施例对此不做限定。在一种可选的实施方式中，运营商可以预先根据众包数据和经验，为不同的攻击手段设计至少一个入侵检测规则。入侵检测装置可以根据需要从云端服务器获取至少一个入侵检测规则并保存，也可以由人工在车辆本地的存储设备上保存至少一个入侵检测规则，本申请实施例对该至少一个入侵检测规则的获取方式不做限定。
图2中，不同模块之间的双向箭头仅用于表示相应模块之间可以通信，并不限定任何通信方式或信息格式。图2中示出的车辆中的其它模块仅是示例，虚线框仅表示相应模块是可选模块，该车辆中可以不包含图2所示的部分模块，也可以包括除图2所示的部分模块以外的其它模块，或将图2中的部分模块替换为未示出的其它模块，在此不再赘述。在一些设计中，车辆的传感系统也可以集成在MDC、VCU或整车域控制单元(vehicle domain controller，VDC)中的任一项上，本申请实施例对车辆的不同模块的产品形态或者集成方式不做限定。
结合图1和图2所示的系统架构可以实施本申请实施例的入侵检测方法，参阅图3所示，该入侵检测方法可以包括以下步骤：
S310：入侵检测装置获取第一特征数据。
本申请实施例中，第一特征数据描述第一车辆所在区域相关的静态交通对象的信息。
其中，第一车辆为前文介绍的自车。静态交通对象包括第一车辆所在区域相关的静态的交通参与者，例如车道线、交通设施、建筑物、山体等。静态交通对象的信息例如可以包括静态交通对象自身的属性信息，例如位置信息、尺寸信息、形状信息等。静态交通对象的信息例如也可以包括静态交通对象相对于第一车辆的信息，例如静态交通对象相对于第一车辆的距离信息、方位角信息等。本申请实施例对此静态交通对象的信息的内容不做限定。
在实施S310时，入侵检测装置可以是从N种第三特征数据中筛选该第一特征数据。其中，该N种第三特征数据可以为对N种感知数据进行感知处理后得到的，该N种感知数据为通过与第一车辆关联的N个传感器采集到的路况信息，该路况信息例如为原始感知数据，N大于等于1。
其中，可以是入侵检测装置对N种感知数据进行感知处理，也可以是其它装置对N种感知数据进 行感知处理，本申请实施例对此不做限定。该感知处理可以包括但不限于是对原始感知数据进行语义分割处理、多目标检测处理等。以原始感知数据包括图像为例，感知处理后得到的N种第三特征数据例如可以包括图像的RGB矩阵向量以及识别到图像中的交通对象对应的标签，入侵检测装置可以根据标签，从N种第三特征数据中筛选第一特征数据，所述第一特征数据关联的标签用于描述静态交通对象，例如车道线特征、交通设施的特征、建筑物的特征、山体的特征等。
S320：入侵检测装置获取第二特征数据。
本申请实施例中，第二特征数据描述第一车辆所在区域相关的动态交通对象的信息。
其中，动态交通对象包括第一车辆所在区域相关的动态的交通参与者，例如第一车辆周边的其他车辆、行人等。动态交通对象的信息例如包括动态交通对象自身的属性信息，例如其他车辆的位置信息、形状信息、尺寸信息等。动态交通对象的信息例如也可以包括动态交通对象相对于第一车辆的信息，例如动态交通对象相对于第一车辆的距离信息、方位角信息等。本申请实施例对该动态交通对象的信息的内容不做限定。
在实施S320时，入侵检测装置可以是根据N种第三特征数据获取第二特征数据。其中，该N种第三特征数据可以为对N种感知数据进行感知处理后得到的，该N种感知数据为通过与第一车辆关联的N个传感器采集到的路况信息，该路况信息例如为原始感知数据，N大于等于1。
其中，可以是入侵检测装置对N种感知数据进行感知处理，也可以是其它装置对N种感知数据进行感知处理，本申请实施例对此不做限定。该感知处理可以包括但不限于是对原始感知数据进行语义分割处理、多目标检测处理等。以动态交通对象包括第一车辆周边的其他车辆为例，该第二特征数据例如可以包括其他车辆的数量。
S330：入侵检测装置根据所述第一特征数据、所述第二特征数据和入侵检测规则，确定发生入侵事件。
本申请实施例中，入侵检测规则描述第一车辆的待检测业务关联的入侵事件的检测方法。
以该待检测业务为车辆的智能驾驶业务、执行装置包括智能驾驶系统为例，入侵事件例如可以包括攻击者伪造车辆的位置信息或者伪造周边其他车辆的信息来“欺骗”车辆的智能驾驶系统的事件，入侵检测规则描述的检测方法可以是将上述第一特征数据或者第二特征数据分别与各自对应的参考信息进行比较来分析是否发生入侵事件的方法。
其中，第一特征数据对应的参考信息可以包括通过第一车辆的定位部件获取到的第一车辆的第一位置信息，第一特征数据可以用于确定第一车辆的位置信息，例如表示为第二位置信息。入侵检测装置可以将该第一位置信息与第二位置信息进行比较来确定是否发生入侵事件。第二特征数据对应的参考信息可以包括从地图服务器获取到的第一车辆的第四特征数据，该第四特征数据为地图服务器提供的第一车辆的周边车辆的信息，第二特征数据为基于感知的方式获得的且用于确定第一车辆的周边车辆的信息，入侵检测装置可以用于将第二特征数据与第四特征数据进行比较来确定是否发生入侵事件。
若确定发生入侵事件，入侵检测装置还可以执行以下步骤：
S340：入侵检测装置向第一车辆的智能驾驶系统发送告警信息，所述告警信息指示所述入侵事件。
相应地，智能驾驶系统可以接收来自入侵检测装置的告警信息，并结合该告警信息执行车辆相关的控制处理，以保障车辆的安全行驶。例如，智能驾驶系统可以向车控相关的ECU发送控制命令，以使得车控相关的ECU执行刹车、转向、加速、减速等，以辅助控制车辆安全行车，从而保障车辆的安全行驶。
为了便于理解，下面以待检测业务为智能驾驶业务为例，结合附图及实施例对S310-S340的详细实现进行介绍。
入侵检测装置在实施S310之前可以生成GPS业务相关联的特征库，并基于该特征库中的样本训练得到M个位置预测模型，该M个位置预测模型可用于在实施上述S330时，根据第一特征数据预测得到第二位置信息，该第二位置信息可用于与通过第一车辆的定位部件获取的第一车辆的第一位置信息进行比较，来分析是否发生入侵事件。
如图4所示，以通过智能驾驶系统的融合感知模块对原始感知数据进行感知处理为例，GPS业务相关联的特征库的生成过程以及相应的模型训练过程可以包括以下步骤：
S401：与第一车辆关联的N个传感器采集N种原始感知数据，并将N种原始感知数据提供至智能驾驶系统的融合感知模块，N大于等于1。
S402：智能驾驶系统的融合感知模块对获取到的N种原始感知数据进行感知处理，比如语义分割处理、多目标检测处理等，将经过感知处理后得到N种第三特征数据提供至入侵检测装置。
经过感知处理后得到N种第三特征数据，该第三特征数据为特征矩阵，包括传感器特征向量与标签，表示为(X,Y)，其中，X表示传感器特征向量，Y表示识别到的交通对象的标签，例如路灯、交通标识牌、匝道口等。
S403：入侵检测装置根据标签从N种第三特征数据中筛选所需的特征数据并保存，即生成GPS业务相关联的特征库。该特征数据描述第一车辆所在区域相关的静态交通对象的信息，该特征数据以外的特征数据可丢弃。
筛选后获得的特征矩阵可以与相应的位置信息(例如GPS经纬度数据)拼接为一个新的矩阵，表示为(X，Y，P)，其中，X表示传感器特征向量，Y表示识别到的交通对象的标签，P表示GPS经纬度数据，不同传感器特征向量对应的特征矩阵(X，Y，P)可用于训练不同类型的传感器对应的位置预测模型。
S404：入侵检测装置基于不同传感器对应的特征矩阵(X，Y，P)训练不同类型的传感器对应的位置预测模型。
如图5所示，所训练的M个位置预测模型包括以下至少一种传感器对应的有监督学习模型：摄像头、激光雷达传感器、毫米波雷达传感器或者超声波雷达传感器。以该有监督学习模型为深度学习模型为例，在训练时，可以将经过筛选和拼接后得到的不同类型传感器对应的特征矩阵输入到相应的初始深度学习模型，经过基于位置向量P的有监督学习的训练过程，可以得到不同类型传感器对应的位置预测模型，例如摄像头对应的位置预测模型、激光雷达传感器对应的位置预测模型、毫米波雷达传感器对应的位置预测模型或者超声波雷达传感器。M个位置预测模型中的每个位置预测模型用于对基于一种传感器获得的特征数据预测一个位置信息。
其中，该筛选和训练的过程可以通过接口调用的方式实现，类似于get_models_from_sensors(sensor_name_list)接口，可用来根据从传感器识别的路况数据作为输入进行深度模型训练，并输出训练好的模型。类似于sensor_id_list表示参与模型训练的数据输入的传感器id列表。该接口的实现类似逻辑的形式化描述如下：
以t1，t2，t3，t4分别代表不同传感器对应待训练的深度学习模型的API接口，以S1，S2，S3，S4分别代表不同类型的传感器向相应的深度学习模型提供的特征矩阵，Xi表示传感器对应的输入特征向量，Yi表示标签向量，Pi表示位置对应的经纬数据特征向量，如下所示：
S1＝(X1,Y1,P1),S2＝(X2,Y2,P2),S3＝(X3,Y3,P3),S3＝(X4,Y4,P4)；
则：
M1＝t1(S1)，M2＝t2(S2)，M3＝t3(S3)，M4＝t4(S4)；
其中，M1,M2,M3,M4分别表示训练好的深度学习模型。
在具体实施时，ti会根据标签向量Yi对输入的特征向量进行过滤，从而筛选出所需的描述静态交通对象的信息Xi，而将不需要的特征向量丢弃，不参与模型训练过程。
S405(可选步骤)：根据需要将训练得到的M个位置预测模型与云端服务器的模型数据进行增量同步。
即将云端服务器上存储的用于位置预测的模型数据更新为训练得到的M个位置预测模型对应的模型数据。
基于图4和图5训练得到M个位置预测模型后，即可利用该M个位置预测模型实施图3所示的方法。如图6所示，该方法可以包括以下步骤：
S601：与第一车辆关联的N个传感器采集N种原始感知数据，并将N种原始感知数据提供至智能驾驶系统的融合感知模块，N大于等于1。
S602：智能驾驶系统的融合感知模块对获取到的N种原始感知数据进行感知处理，比如语义分割处理、多目标检测处理等，将经过感知处理后得到N种第三特征数据提供至入侵检测装置。
经过感知处理后得到N种第三特征数据，该第三特征数据为特征矩阵，包括传感器特征向量与标签，表示为(X,Y)，其中，X表示传感器特征向量，Y表示识别到的交通对象的标签，例如路灯、交通标识牌、匝道口等。
S603：入侵检测装置根据标签从N种第三特征数据中筛选所需的第一特征数据，第一特征数据描 述第一车辆所在区域相关的静态交通对象的信息，和/或，入侵检测装置根据N种第三特征数据获取第二特征数据，所述第二特征数据包括所述其他车辆的数量。
S604：入侵检测装置将第一特征数据中关联于不同传感器的特征数据输入M个位置预测模型，获得M个位置预测值，并根据M个位置预测值加权得到所述根据所述第一特征数据确定的位置信息，即平均位置预测值，表示为第二位置信息，M大于1。
如图7所示，以(X,Y)表示筛选后获得的特征矩阵，X表示传感器特征向量，Y表示识别到的交通对象的标签，例如路灯、交通标识牌、匝道口等。将不同传感器关联的(X,Y)分别输入到M个位置预测模型，每个位置预测模型输出一个位置预测值，根据M个位置预测值进行加权平均处理可以得到平均GPS预测值，即第二位置信息。
其中，上述预测过程也可以通过接口调用的方式实现，类似于get_pos_from_sensors(sensor_name_list)接口，表示将从传感器识别的路况数据作为输入数据，输入到预训练好的深度学习模型，并输出相应的GPS位置的预测值。类似于sensor_id_list表示参与位置计算的传感器id列表。该接口的实现类似逻辑的形式化描述如下：
以f1，f2，f3，f4分别表示不同传感器对应预训练好的深度学习模型对应的函数接口，以S1，S2，S3，S4分别代表不同类型的传感器向相应的深度学习模型提供的特征矩阵，Xi表示传感器对应的输入特征向量，Yi表示标签向量，Pi表示位置对应的经纬数据特征向量，如下所示：
S1＝(X1,Y1),S2＝(X2,Y2),S3＝(X3,Y3),S3＝(X4,Y4)；
则：
P1＝f1(S1),P2＝f2(S2),P3＝f3(S3),P4＝f4(S4)；
其中fi会根据标签向量Yi对特征向量Xi进行过滤，从而筛选出所需的描述静态交通对象的信息Xi，而将不需要的特征向量丢弃，不参与模型预测过程。
其中，P1,P2,P3,P4分别表示不同模型的对当前位置(例如经纬度数据)的预测值，对M个位置预测值进行的加权平均处理时，位置预测值可以用如下表达式表示：
本申请实施例中，可以周期性调用上述接口生成包括时间和位置信息的序列，并采用滑动窗口(sliding window)算法，在时间窗口内使用相似度算法计算位置预测值与基于第一车辆的定位部件(例如GPS信号接收器)接收到的GPS序列的相似度。
例如，按时间序列周期性调用get_pos_from_sensors(sensor_name_list)接口，在一个时间窗口内，比如时间的取值为[202303020605,202303020606,202303020607]时，可以生成一个时间和位置信息(例如经纬度数据)的序列，表示如下：
sensor_position_timing_num＝[(202303020605,22.83239514238987,114.08831784625996),
(202303020605,22.819089228869782,114.10772361277796),(202303020605,22.825155625708656,114.14154185306363)]。
调用get_pos_from_gps()接口，可以用来获取从位置传感器(例如GPS信号接收装置)获取到的位置信息。时间序列周期性的调用get_pos_from_gps()接口，在一个时间窗口内，如时间的取值为[202303020605,202303020606,202303020607]时，可以生成一个时间和位置的经纬序列，表示如下：
gps_position_timing_num＝[(202303020605,22.83239514238997,114.08831784625956),
(202303020605,22.819089228869792,114.10772361277706),(202303020605,22.825155625708636,114.14154185306323)]。
调用get_pos_distance(cloud_cars_timing_num,sensor_cars_timing_num)接口，用来计算时间序列窗口内的相识度。该接口返回两个时序的差异度timing_pos_distance，如果timiing_pos_distance>POS_DISTANCE_THRESHOLD，即相似度小于或等于第一阈值，则可认为存在GPS欺骗。
S605：入侵检测装置从地图服务器获取第一车辆的第四特征数据，并分析第四特征数据与第二特征数据的差异。
其中，第四特征数据为所述地图服务器提供的所述第一车辆的周边车辆的信息，例如第四特征数据包括所述第一车辆周围第一范围内的其他车辆的参考数量。若第二特征数据指示的其他车辆的数量与所述第四特征数据指示的其他车辆的参考数量的相似度小于或等于第二阈值，则可认为存在GPS欺骗。
该过程也可以通过接口调用的方式实现，周期性调用相关接口生成时间和车辆数量的序列，并采用滑动窗口算法，在时间窗口内使用相似度算法计算通过感知数据确定的其他车辆的数量，与从地图服务器获得的其他车辆的数量的相似度。
例如，入侵检测装置可使用地图服务商提供的类似get_pos_range_cars(longitude,latitude,range,time)接口，用来获取第一车辆当前所在位置周围的其他车辆的数量，longitude和latitude表示进行第一车辆的当前实时经纬度信息，该经纬度信息是GPS信号接收装置接收到的GPS数据，range代表该经纬度对应的感知范围，例如第一车辆周围20m的范围,time表示请求时间。地图服务器返回的信息为经纬度信息的集合，表示如下：
s1＝{(x1,y1),(x2,y2)},其中x，y分别代表经纬度信息。
入侵检测装置可使用类似get_pos_range_cars接口实现类似get_cars_from_map_service(longitude,latitude,range,time,service_id_list)接口，用来从地图服务器获取第一车辆周边的其他车辆的信息的。其中service_id_list代表地图服务商的id列表。该接口实现逻辑可以形式化描述如下：
以s1，s2，s3代表不同地图服务商提供的调用接口的返回值：
s1＝{(x1,y1),(x2,y2)}；
s2＝{(x2,y2),(x3,y3)}；
s3＝{(x1,y1),(x2,y2),(x4,y4)}；
则：
s＝s1∪s2∪s3＝{(x1,y1),(x2,y2),(x3,y3),(x4,y4)}；
经过并集处理后，该接口返回从地图服务商获取的车辆位置集合，则cars_from_cloud＝s.size()＝4。也即第一车辆周边的其他车辆的数量为4。
其中，按时间序列周期性调用get_cars_from_map_service(longitude,latitude,range,time,service_id_list)接口，在一个时间窗口内，如时间的取值为[202303020605,202303020606,202303020607]，可以生成一个时间和车辆数量的序列，表示如下：
cloud_cars_timing_num＝[(202303020605,3),(202303020605,4),(202303020605,4)]。
按时间序列周期性调用get_cars_from_sensors(range)接口，表示根据原始感知数据识别到的第一车辆周边的其他车辆的数量，即第二特征数据。在一个时间窗口内，如时间的取值为[202303020605,202303020606,202303020607],则可以生成一个时间和车辆数量的序列，表示如下：
sensor_cars_timing_num＝[(202303020605,3),(202303020605,4),(202303020605,4)]。
调用类似计算时间序列窗口内的相识度接口get_car_num_distance(cloud_cars_timing_num,sensor_cars_timing_num)，该接口返回两个时序的差异度timing_num_distance，如果timiing_num_distance>NUM_DISTANCE_THRESHOLD，即相似度小于或等于第二阈值，则认为存在GPS欺骗。
本申请实施例中，上述S605、S606中使用到的滑动窗口算法可以包括但不限于是分段线性表示(piecewise linear represent，PLR)算法、动态时间归整(dynamic time warping，DTW)算法等，本申请实施例对此算法实现不做限定。相似度算法可以包括但不限于如欧式距离算法、余弦相似度算法、闵可夫斯基距离算法等，本申请实施例对此算法实现也不做限定。
S606：入侵检测装置根据S604和S605的结果确定存在GPS欺骗，即存在入侵事件。
S607：入侵检测装置向智能驾驶系统发送告警信息，所述告警信息指示所述入侵事件。智能驾驶系统可以结合该告警信息实施后续的控制处理，以保障车辆的安全行驶。
S608(可选步骤)：入侵检测装置向智能座舱发送告警信息，所述告警信息指示所述入侵事件。智能座舱可以接收来自入侵检测装置的告警信息，并可以通过中控显示器、抬头显示器件、音响系统、座椅震动、灯光等输出该告警信息，以供车辆内乘员(包括驾驶员或者其他乘员)参考，从而辅助车辆驾驶员控制车辆安全行驶，或者使得其他乘员根据该告警信息为车辆驾驶员提供驾驶辅助。
由此，通过上述方法，一方面，利用车辆的多种传感器获取周围环境和路况的特定静态特征来进行入侵检测，以防止攻击者对GPS信号接收装置进行阻塞后再伪造GPS信号进行仿冒攻击，另一方面，利用车辆的多种传感器获取周围环境和路况的特定动态特征来进行入侵检测，以防止攻击者基于具有相似环境特征的位置伪造GPS信号。该方法中，GPS信号检测特征不来自于GPS信号本身，可以有效检测大范围的GPS信号仿冒攻击，同时只需依赖自动驾驶功能已装配的传感设备和计算设备即可，无需 依赖其它硬件支持，也不会带来额外的硬件成本。
在其它实施例中，上述入侵检测方法还可以作为车辆的其它功能的子功能，例如在基于位置信息的推荐系统中，通过实施上述方法以分析是否发生入侵事件，若发生入侵事件，则不可直接使用GPS信号接收装置收到的GPS信号作为实时的定位信息进行推荐，以免向用户推荐错误的项目。作为可选的实施方式，上述基于多种传感器的原始感知数据识别到的位置预测值，可以暂时用作可靠的位置信息，用于向用户进行推荐。在解决入侵事件后，可再利用GPS信号接收装置收到的GPS信号作为实时的定位信息进行推荐。
本申请实施例还提供了一种入侵检测装置，用于执行上述方法实施例中入侵检测装置所执行的方法，相关特征可以参见上述方法实施例，在此不再赘述。
如图8所示，该入侵检测装置800可以包括：第一获取单元801，用于获取第一特征数据，所述第一特征数据描述第一车辆所在区域相关的静态交通对象的信息；第二获取单元802，用于获取第二特征数据，所述第二特征数据描述所述第一车辆所在区域相关的动态交通对象的信息；确定单元803，用于根据所述第一特征数据、所述第二特征数据和入侵检测规则，确定发生入侵事件，所述入侵检测规则描述所述第一车辆的待检测业务关联的入侵事件的检测方法；通信单元804，用于向所述第一车辆的智能驾驶系统发送告警信息，所述告警信息指示所述入侵事件。具体实现方式，请参考上述方法实施例中入侵检测装置所实现的方法步骤，这里不再赘述。
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，例如中央处理单元(Central Processing Unit，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，该硬件电路为专用集成电路(application-specific integrated circuit，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可编程逻辑器件(programmable logic device，PLD)实现，以现场可编程门阵列(Field Programmable Gate Array，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、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 singnal 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。
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，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。
在一个简单的实施例中，本领域的技术人员可以想到上述实施例中的入侵检测装置均可采用图9所示的形式。
如图9所示的装置900，包括至少一个处理器910和通信接口930。在一种可选的设计中，还可以包括存储器920。
本申请实施例中不限定上述处理器910以及存储器920之间的具体连接介质。
在如图9的装置中，处理器910在与其他设备进行通信时，可以通过通信接口930进行数据传输。
当通信装置采用图9所示的形式时，图9中的处理器910可以通过调用存储器920中存储的计算机执行指令，使得装置900可以执行上述任一方法实施例。
本申请还提供了一种车辆，该车辆结构如图2所示，其中的入侵检测装置具有图8或图9所示的结构。
本申请实施例还涉及一种芯片系统，该芯片系统包括处理器，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任一实施例的方法。
在一种可能的实现方式中，该处理器可以通过接口与存储器耦合。
在一种可能的实现方式中，该芯片系统还可以直接包括存储器，该存储器中存储有计算机程序或计算机指令。
示例地，存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的RAM可用，例如静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(dynamic RAM，DRAM)、同步动态随机存取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(synchlink DRAM，SLDRAM)和直接内存总线随机存取存储器(direct rambus RAM，DR RAM)。
本申请实施例还涉及一种处理器，该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任一实施例所述的方法。
示例地，在本申请实施例中，处理器是一种集成电路芯片，具有信号的处理能力。例如，该处理器可以是FPGA，可以是通用处理器、DSP、ASIC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，SoC，CPU，网络处理器(network processor，NP)，微控制器(micro controller unit，MCU)，PLD或其他集成芯片，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，处理器读取存储器中的信息，结合其硬件完成上述方法的步骤。
应明白，本申请的实施例可提供为方法、系统、或计算机程序产品。
在一种可能的实现方式中，本申请实施例提供了一种计算机可读存储介质，所述计算机可读存储介质存储有程序代码，当所述程序代码在所述计算机上运行时，使得计算机执行上述方法实施例。
在一种可能的实现方式中，本申请实施例提供了一种计算机程序产品，当所述计算机程序产品在计算机上运行时，使得所述计算机执行上述方法实施例。
因此，本申请可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
显然，本领域的技术人员可以对本申请实施例进行各种改动和变型而不脱离本申请实施例范围。这样，倘若本申请实施例的这些修改和变型属于本申请权利要求及其等同技术的范围之内，则本申请也意图包含这些改动和变型在内。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。

Claims (12)

  1. 一种入侵检测方法，其特征在于，包括：
    获取第一特征数据，所述第一特征数据描述第一车辆所在区域相关的静态交通对象的信息；
    获取第二特征数据，所述第二特征数据描述所述第一车辆所在区域相关的动态交通对象的信息；
    根据所述第一特征数据、所述第二特征数据和入侵检测规则，确定发生入侵事件，所述入侵检测规则描述所述第一车辆的待检测业务关联的入侵事件的检测方法；
    向所述第一车辆的智能驾驶系统发送告警信息，所述告警信息指示所述入侵事件。
  2. 根据权利要求1所述的方法，其特征在于，所述获取第一特征数据，包括：
    从N种第三特征数据中筛选所述第一特征数据，其中，所述N种第三特征数据为对N种感知数据进行感知处理后得到的，所述N种感知数据为通过与所述第一车辆关联的N个传感器采集到的原始感知数据，N大于等于1。
  3. 根据权利要求2所述的方法，其特征在于，所述感知处理包括多目标检测处理，所述N种第三特征数据包括识别到的交通对象的标签，所述从N种第三特征数据中筛选所述第一特征数据，包括：
    根据所述标签，从N种第三特征数据中筛选所述第一特征数据，其中，所述第一特征数据关联的标签用于描述所述静态交通对象。
  4. 根据权利要求1-3中任一项所述的方法，其特征在于，所述动态交通对象包括所述第一车辆周围第一范围内的其他车辆，所述获取第二特征数据，包括：
    根据N种第三特征数据获取所述第二特征数据，其中，所述第二特征数据包括所述其他车辆的数量，所述N种第三特征数据为对N种感知数据进行感知处理后得到的，所述N种感知数据为通过与所述第一车辆关联的N个传感器采集到的原始感知数据，N大于等于1。
  5. 根据权利要求1-4中任一项所述的方法，其特征在于，所述待检测业务关联所述第一车辆的位置信息，所述入侵检测规则包括分析通过不同方式获得的信息的差异，所述方法还包括：
    通过所述第一车辆的定位部件获取所述第一车辆的第一位置信息；
    从地图服务器获取所述第一车辆的第四特征数据，所述第四特征数据为所述地图服务器提供的所述第一车辆的周边车辆的信息；
    所述根据所述第一特征数据、所述第二特征数据和入侵检测规则，确定发生入侵事件，包括：
    当根据所述第一特征数据确定的位置信息和所述第一位置信息的差异满足第一条件，以及所述第二特征数据和所述第四特征数据的差异满足第二条件时，确定发生定位入侵事件。
  6. 根据权利要求5所述的方法，其特征在于，所述第一条件包括：所述根据所述第一特征数据确定的位置信息与所述第一位置信息的相似度小于或等于第一阈值，所述方法还包括：
    将所述第一特征数据中关联于不同传感器的特征数据输入M个位置预测模型，获得M位置预测值，M大于1；
    根据所述M个位置预测值加权得到所述根据所述第一特征数据确定的位置信息；
    其中，所述M个位置预测模型中的每个位置预测模型用于对基于一种传感器获得的特征数据预测一个位置信息。
  7. 根据权利要求6所述的方法，其特征在于，所述M个位置预测模型包括以下至少一种传感器对应的有监督学习模型：
    摄像头、激光雷达传感器、毫米波雷达传感器或者超声波雷达传感器。
  8. 根据权利要求5-7中任一项所述的方法，其特征在于：
    所述第四特征数据包括所述第一车辆周围第一范围内的其他车辆的参考数量；
    所述第二条件包括：所述第二特征数据指示的其他车辆的数量与所述第四特征数据指示的其他车辆的参考数量的相似度小于或等于第二阈值。
  9. 一种入侵检测装置，其特征在于，包括：
    第一获取单元，用于获取第一特征数据，所述第一特征数据描述第一车辆所在区域相关的静态交通对象的信息；
    第二获取单元，用于获取第二特征数据，所述第二特征数据描述所述第一车辆所在区域相关的动态交通对象的信息；
    确定单元，用于根据所述第一特征数据、所述第二特征数据和入侵检测规则，确定发生入侵事件，所述入侵检测规则描述所述第一车辆的待检测业务关联的入侵事件的检测方法；
    通信单元，用于向所述第一车辆的智能驾驶系统发送告警信息，所述告警信息指示所述入侵事件。
  10. 一种通信装置，其特征在于，包括至少一个处理器和接口电路，所述接口电路用于为所述至少一个处理器提供数据或者代码指令，所述至少一个处理器用于通过逻辑电路或执行代码指令实现如权利要求1-8中任一项所述的方法。
  11. 一种车辆，其特征在于，包括用于实现如权利要求1-8中任一项所述的方法的模块或包括如权利要求9所述的入侵检测装置。
  12. 一种计算机可读存储介质，其特征在于，所述计算机可读介质存储有程序代码，当所述程序代码在计算机上运行时，使得计算机执行如权利要求1-8中任一项所述的方法。
PCT/CN2024/131072 2023-11-14 2024-11-08 一种入侵检测方法、装置以及车辆 Pending WO2025103243A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202311519004.6A CN117755334A (zh) 2023-11-14 2023-11-14 一种入侵检测方法、装置以及车辆
CN202311519004.6 2023-11-14

Publications (1)

Publication Number Publication Date
WO2025103243A1 true WO2025103243A1 (zh) 2025-05-22

Family

ID=90311172

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2024/131072 Pending WO2025103243A1 (zh) 2023-11-14 2024-11-08 一种入侵检测方法、装置以及车辆

Country Status (2)

Country Link
CN (1) CN117755334A (zh)
WO (1) WO2025103243A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117755334A (zh) * 2023-11-14 2024-03-26 华为技术有限公司 一种入侵检测方法、装置以及车辆

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111829545A (zh) * 2020-09-16 2020-10-27 深圳裹动智驾科技有限公司 自动驾驶车辆及其运动轨迹的动态规划方法及系统
US20210114606A1 (en) * 2020-12-23 2021-04-22 Intel Corporation Systems and methods for intrusion detection in vehicle systems
CN112822684A (zh) * 2021-02-04 2021-05-18 中汽创智科技有限公司 车辆入侵检测方法及防御系统
CN115348091A (zh) * 2022-08-17 2022-11-15 广州小鹏自动驾驶科技有限公司 基于自动驾驶的入侵检测方法、装置和电子设备
CN116708014A (zh) * 2023-07-25 2023-09-05 南京项尚车联网技术有限公司 一种智能驾驶车辆系统的信息安全方法
CN116866033A (zh) * 2023-07-10 2023-10-10 一汽解放汽车有限公司 车辆的安全防护方法、装置、计算机设备和存储介质
CN117755334A (zh) * 2023-11-14 2024-03-26 华为技术有限公司 一种入侵检测方法、装置以及车辆

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111829545A (zh) * 2020-09-16 2020-10-27 深圳裹动智驾科技有限公司 自动驾驶车辆及其运动轨迹的动态规划方法及系统
US20210114606A1 (en) * 2020-12-23 2021-04-22 Intel Corporation Systems and methods for intrusion detection in vehicle systems
CN112822684A (zh) * 2021-02-04 2021-05-18 中汽创智科技有限公司 车辆入侵检测方法及防御系统
CN115348091A (zh) * 2022-08-17 2022-11-15 广州小鹏自动驾驶科技有限公司 基于自动驾驶的入侵检测方法、装置和电子设备
CN116866033A (zh) * 2023-07-10 2023-10-10 一汽解放汽车有限公司 车辆的安全防护方法、装置、计算机设备和存储介质
CN116708014A (zh) * 2023-07-25 2023-09-05 南京项尚车联网技术有限公司 一种智能驾驶车辆系统的信息安全方法
CN117755334A (zh) * 2023-11-14 2024-03-26 华为技术有限公司 一种入侵检测方法、装置以及车辆

Also Published As

Publication number Publication date
CN117755334A (zh) 2024-03-26

Similar Documents

Publication Publication Date Title
CN111354187B (zh) 用于辅助交通工具的驾驶员的方法和驾驶员辅助系统
CN110949376B (zh) 车辆控制装置、车辆控制方法及存储介质
CN113741485A (zh) 车路协同自动驾驶的控制方法、装置、电子设备及车辆
EP3895950A1 (en) Methods and systems for automated driving system monitoring and management
CN113320541B (zh) 车辆控制装置、车辆控制方法及存储介质
CN110371114A (zh) 车辆控制装置、车辆控制方法及存储介质
CN107949875B (zh) 用于确定具有交互可能性的交通参与者的方法和系统
CN112639913B (zh) 车辆控制装置及其方法、自动驾驶车开发系统、存储介质
CN108399792A (zh) 一种无人驾驶车辆避让方法、装置和电子设备
KR102788663B1 (ko) 규칙집 궤적 생성을 위한 그래프 탐색
CN115265537B (zh) 具有交通状态检测机制的导航系统及其操作方法
CN111009146A (zh) 服务器、信息处理方法和存储程序的非暂时性存储介质
JP6903598B2 (ja) 情報処理装置、情報処理方法、情報処理プログラム、および移動体
US12277769B2 (en) Method for deep neural network functional module deduplication
CN118053321A (zh) 车辆碰撞威胁评估
US12602931B2 (en) Identification of unknown traffic objects
WO2025103243A1 (zh) 一种入侵检测方法、装置以及车辆
KR20250085763A (ko) 모션 추론을 통해 신호등 라벨링을 위한 방법 및 시스템
KR20250088548A (ko) 원격 차량 지원을 위한 궤적들의 안내된 생성
KR102901210B1 (ko) 종속 가능성 구조체로의 다중 타겟 추적
CN112249016A (zh) 用于自动驾驶车辆的掉头控制系统和方法
CN113771845B (zh) 预测车辆轨迹的方法、装置、车辆和存储介质
US20250002056A1 (en) Dynamic risk adaptation in a vehicle
KR20250093331A (ko) 차량 환경에서 새로운 클래스의 대상체의 식별
KR20250047338A (ko) 맵 및 그룹 파라미터로부터 대상체 배향 결정

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: 24890599

Country of ref document: EP

Kind code of ref document: A1