Method and device for verifying the security of intelligent control system

Abstract

The embodiment of the present invention provides a method and device for verifying the safety of an intelligent control system, and relates to the technical field of safety verification methods and devices. The device comprises: a modeling subsystem, a three-dimensional display subsystem, an environment modeling subsystem, a physical interaction subsystem and an auxiliary test subsystem of an intelligent control system; the modeling subsystem of the intelligent control system comprises: a task generation unit, control software and AirSim simulation software; the task generation unit generates a task and sends it to the control software, the control software generates a task instruction according to the task and sends it to the AirSim simulation software, the AirSim simulation software generates a system instruction according to the task instruction and sends it to the physical interaction subsystem, and receives system status data returned by the physical interaction subsystem; the AirSim simulation software sends model power data to the three-dimensional display subsystem and receives virtual environment data sent by the three-dimensional display subsystem.

Description

Technical Field

[0001] The present invention relates to the technical field of security verification methods and devices, and in particular to a method and device for verifying the security of an intelligent control system. Background Art

[0002] In the trend of the modern information age, control systems are developing in the direction of unmanned and autonomous. Intelligent control systems with functions such as reconnaissance, patrol, etc. have become increasingly mature, such as drones and unmanned vehicles. However, the means of attacking sensors, control systems, and networking communications of intelligent control systems are changing with each passing day. Among them, one attack method that can effectively strike intelligent control systems and destroy their mission execution capabilities to the greatest extent is to build a denial environment (Denied Environments) to interfere with and suppress satellite positioning, real-time communication, and sensor signals that intelligent control systems rely on. These factors have posed new challenges to the security technology of intelligent control system equipment.

[0003] In the research on the means of protection against denial environments, building a virtual simulation platform with a denial environment for verifying the security of intelligent control systems can provide great convenience. The research on intelligent control system simulation has been very common and relatively mature. Common intelligent control system simulation systems are mainly divided into software-in-the-loop (SITL) simulation and hardware-in-the-loop (HITL) simulation. The system firmware code of software-in-the-loop simulation runs on the same computer where the virtual environment runs, or on another computer on the same network, while hardware-in-the-loop simulation uses a real-time processor to run a simulation model to simulate the operating state of the controlled object, and is connected to the object under test through an I / O interface.

[0004] At present, the intelligent control system simulation environment widely used in secondary development is Gazebo. Gazebo has a powerful 3D simulation environment, which is particularly suitable for testing object obstacle avoidance and computer vision. It is currently used for multi-vehicle simulation abroad and can be used with tools such as ROS for automatic control. Gazebo currently supports simulation vehicles for intelligent control systems including: fixed-wing, multi-rotor, hybrid-wing drones, unmanned vehicles, unmanned submarines, etc.

[0005] A semi-physical simulation system based on the Gazebo virtual environment that supports PX4 flight control hardware has a flight control board hardware with in-loop mode activated. In this mode, no sensors on the flight control board are started, and all sensor data comes from the simulation. The Gazebo virtual environment is connected to the flight control board via USB, and to the ground station via UDP and transmits data via MAVLink. The joystick / game controller can be connected to the simulation loop through the ground station via the serial port.

[0006] In the data processing flow of Gazebo simulation, the flight simulation data generated by the simulation environment is first input into the plug-in attached to the simulation. The plug-in here refers to the control program for the drone model in the simulation environment, such as the DroneKit drone autonomous flight program. After being processed by the plug-in, the data is packaged according to the Mavlink protocol and sent to the flight control board with the hardware-in-the-loop mode enabled, where the drone flight control stack (such as PX4) generates flight instructions and control signals. Building a denial area in the Gazebo simulation environment depends on the development of custom plug-ins. At present, a more feasible solution is to use the ROS system to create a program node that can simulate the denial effect as a simulation plug-in to process the simulation data in the Gazebo environment. Although Gazebo simulation has a technical route to build a denial simulation environment, there are still many problems in its implementation:

[0007] (1) Most physical models in Gazebo simulation are simple geometric bodies, which cannot build realistic and complex 3D scenes, and are not conducive to the simulation and reproduction of real scenes. The virtual environment in Gazebo simulation is theoretically a program node in the ROS system. Its communication with the Gazebo environment depends on ROS message packets, which will bring some problems in data real-time performance.

[0008] (2) Existing security testing methods are mainly conducted in real environments, which consumes a lot of resources and is also limited by the size of the testing site. Summary of the invention

[0009] In order to solve the above technical problems, the present invention proposes a method and a device for verifying the security of an intelligent control system.

[0010] The present invention adopts the following technical solutions:

[0011] A device for verifying the safety of an intelligent control system, comprising:

[0012] Intelligent control system modeling subsystem, 3D display subsystem, environment modeling subsystem, physical interaction subsystem and auxiliary testing subsystem;

[0013] The modeling subsystem of the intelligent control system includes: a task generation unit, control software and AirSim simulation software;

[0014] The task generation unit generates a task and sends it to the control software. The control software generates a task instruction according to the task and sends it to the AirSim simulation software. The AirSim simulation software generates a system instruction according to the task instruction and sends it to the physical interaction subsystem, and receives the system status data returned by the physical interaction subsystem. The AirSim simulation software sends the model power data to the three-dimensional display subsystem and receives the virtual environment data sent by the three-dimensional display subsystem.

[0015] The three-dimensional display subsystem includes a workstation and a UE simulation environment, and the workstation is equipped with and runs the UE simulation environment;

[0016] The UE simulation environment sends the virtual environment data to the AirSim simulation software and the physical interaction subsystem respectively, and receives the model power data and the system status data respectively;

[0017] The environment modeling subsystem includes a UE development engine and an environment modeling unit; the UE development engine sends an environment material package to the environment modeling unit, the environment modeling unit generates the UE simulation environment using the environment material package, and sends the rejection environment type information to the UE simulation environment;

[0018] The physical interaction subsystem collects raw data, receives the virtual environment data sent by the UE simulation environment, receives the system instructions and the system status data, and analyzes the raw data to obtain parsed data. The physical interaction subsystem processes the parsed data into the system status data and communication data to be analyzed, and sends the system status data to the UE simulation environment and the AirSim simulation software, and sends the communication data to be analyzed to the auxiliary test subsystem;

[0019] The auxiliary test subsystem receives the communication data to be analyzed sent by the control hardware, and analyzes and processes the communication data to be analyzed.

[0020] The physical interaction subsystem includes a sensor, an onboard computer, control hardware and a communication encryption module; the sensor collects raw data and sends the raw data to the onboard computer, the onboard computer analyzes the raw data to obtain parsed data, the onboard computer sends the parsed data to the control hardware, the control hardware processes the parsed data into the system status data and the communication data to be analyzed, and sends the system status data to the communication encryption module, and sends the communication data to be analyzed to the auxiliary test subsystem; the control hardware receives the virtual environment data sent by the UE simulation environment;

[0021] The communication encryption module receives the system instructions and the system status data, encrypts the system status data and the system instructions to obtain the system status data; and sends the system status data to the AirSim simulation software and the UE simulation environment respectively.

[0022] The auxiliary test subsystem includes: an oscilloscope, a spectrum analyzer, a programmable power supply, a signal generator, a wireless signal analyzer and a vector network analyzer; the environmental modeling unit includes tools, which include: a resource deployment tool, a terrain modification tool and a physical environment configuration tool; the tools are used to complete the resource deployment of the virtual battlefield scene and the modification of the denial environment factors.

[0023] The environment modeling unit generates the UE simulation environment by using the tool, and obtains the virtual environment data by simulation; and configures the UE simulation environment according to the security test requirements;

[0024] The UE simulation environment modifies the configuration file of the AirSim simulation software by sending the virtual environment data to the AirSim simulation software; the configuration file includes: simulation vehicle, view mode and placement time.

[0025] The modeling subsystem of the intelligent control system is used to virtualize the intelligent control system into an intelligent control system model;

[0026] The interaction process between the intelligent control system modeling subsystem and the UE simulation environment includes:

[0027] When the intelligent control system model enters the virtual denied environment, the task generation unit configures the denied environment type, converts the denied environment type into the task, and sends it to the control software;

[0028] The control software converts the task into the task instruction and sends it to the Airsim simulation software. After receiving the task instruction, the AirSim simulation software sends the model power data to the UE simulation environment, so that the UE simulation environment triggers the UE development engine through the simulation interface;

[0029] The UE development engine calls a search function to obtain the model power data, sends the denied environment type information of the virtual denied environment to the UE simulation environment through the simulation interface, and the UE simulation environment responds according to the denied environment type information.

[0030] After the UE simulation environment responds according to the denied environment type information, the Airsim simulation software responds to different strategies according to the denied environment type information and executes the corresponding tasks; and verifies the security of the intelligent control system according to the task status of the intelligent control system model.

[0031] The environmental material package includes: the virtual battlefield scenes of ocean, mountain and city and the denied environmental factors; the communication data to be analyzed includes: pulse width modulation signal and controller area network bus signal.

[0032] The virtual denied environment adopts a hardware-in-the-loop mode to ensure that the software working on the control hardware receives the virtual environment data sent by the UE simulation environment, processes the virtual environment data as real external environment data, and calculates the corresponding system status data of the real external environment.

[0033] A method for verifying the security of an intelligent control system, the method adopts the device for verifying the security of an intelligent control system, and the method comprises the following steps:

[0034] First, the deployment and configuration of the UE simulation system are completed using the environment modeling unit;

[0035] Then, simulating a virtual environment using the UE simulation environment, and configuring setting items of the UE simulation environment;

[0036] The intelligent control system model enters a virtual denied environment, and the UE development engine configures the denied environment type information of the UE simulation environment;

[0037] After the intelligent control system model enters the virtual denied environment, the UE development engine is triggered, the UE development engine calls a search function to obtain the model power data, the UE development engine sends the denied environment type information to the UE simulation environment through a simulation interface, and the UE simulation environment responds according to the denied environment type information;

[0038] The intelligent control system model responds with different strategies according to the denied environment type information and executes corresponding tasks;

[0039] Finally, according to the task status of the intelligent control system: verify the security of the intelligent control system.

[0040] In a method and device for verifying the security of an intelligent control system provided by an embodiment of the present invention, the device for verifying the security of an intelligent control system includes: an intelligent control system modeling subsystem, a three-dimensional display subsystem, an environmental modeling subsystem, a physical interaction subsystem and an auxiliary test subsystem. The intelligent control system modeling subsystem is used to simulate a real intelligent control system to obtain an intelligent control system model; the three-dimensional display subsystem is used to display a three-dimensional virtual rejection environment, the position and posture of the intelligent control system model, etc., so as to facilitate a more intuitive observation of the system security test process; the environmental modeling subsystem is used to generate a UE simulation environment in the three-dimensional display subsystem, including the rejection environment therein; the physical interaction subsystem realizes a security test environment in the form of semi-physical simulation by connecting the three-dimensional display subsystem, the hardware equipment of the real intelligent control system and the auxiliary test subsystem; the auxiliary test subsystem is used to analyze the communication data in the operation of the control hardware of the intelligent control system, and obtain information data used to evaluate the system security during the system security test process. The workflow of the method and device for verifying the security of an intelligent control system is as follows: security test environment configuration: first, the environment modeling unit of the environment modeling subsystem is used to complete the deployment and configuration of the UE simulation environment of the three-dimensional display subsystem, and then, the UE simulation environment is used to simulate the rejection environment, and the setting items of the UE simulation environment are configured according to the security test requirements; then, the three-dimensional display subsystem, the real hardware equipment of the intelligent control system and the auxiliary test subsystem are connected through the physical interaction subsystem to realize a security test environment in the form of semi-physical simulation; security test process: the intelligent control system model enters the virtual rejection environment, and its task generation unit configures the rejection type of the UE simulation environment; After the control system model enters the denied environment, it triggers the trigger in the denied electromagnetic network set by the UE development engine in the UE simulation environment, and then calls the search function to enable the UE development engine to obtain the role identity information of the intelligent control system and the simulation interface in the role identity information. The UE development engine transmits the type information of the denied environment to the UE simulation environment through the simulation interface, and then the UE simulation environment responds according to the denied environment type information; then, the intelligent control system model responds with different strategies according to the denied type and executes corresponding tasks; the security capability of the intelligent control system is verified according to the task status of the denied environment of the intelligent control system and the test result information data output by the auxiliary test subsystem. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of the composition of a device for verifying the security of an intelligent control system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The specific implementation modes of the present invention are described in detail below with reference to the accompanying drawings.

[0043] Denial environment: refers to an electronic technology means that can set up a hostile environment within a domain to interfere with and suppress the satellite positioning, real-time communications and sensor signals that intelligent control systems rely on. The goal is to make it impossible for intelligent control systems entering the area to complete their tasks correctly or lose control capabilities.

[0044] Semi-physical simulation of intelligent control systems: also known as hardware-in-the-loop simulation, refers to a simulation mode in which the intelligent control system firmware runs on a real intelligent control system controller board and interacts with the simulation environment via a communication link.

[0045] Simulated denial environment: refers to a virtual denial area constructed in a simulation system, which simulates the impact of a real denial environment on the intelligent control system controller by modifying environmental data through programming.

[0046] The method and device for verifying the security of an intelligent control system provided by the embodiments of the present invention mainly solve the following problems:

[0047] For the physical security of intelligent control systems, complex physical virtual scenarios are constructed. Based on the Airsim platform, for common terrains such as oceans, mountains, and cities, virtual environments with a certain area size, a certain degree of complexity, complete natural environmental factors, and visible objects such as buildings, vehicles, and ships are constructed to meet the basic needs of typical physical security verification scenarios and to achieve rapid simulation of different physical scenarios.

[0048] A denial environment for real-time data transmission is built for the cyberspace security of intelligent control systems. In the virtual battlefield of the simulation system, custom programs are used to modify simulation data to create virtual communication obstruction, positioning interference deception, sensor interference and other operational denial environments, breaking through various limitations of physical experiments.

[0049] The data processing module of the denial environment is integrated into the simulation program, with strict real-time operation sequence, which enables convenient migration of the simulation environment under the Windows operating system and reduces the difficulty of operating the simulation program for users.

[0050] See also Figure 1 An embodiment of the present invention provides a device for verifying the safety of an intelligent control system, which includes five subsystems: a modeling subsystem of the intelligent control system, a three-dimensional display subsystem, an environment modeling subsystem, a physical interaction subsystem, and an auxiliary testing subsystem.

[0051] Among them, the three-dimensional display subsystem can visualize the three-dimensional simulation environment, the position and posture of the intelligent control system, the dynamics of the intelligent control system in the physical environment, etc., which is convenient for more intuitive observation and grasp of the overall situation during the intelligent control system development and system testing process. The hardware environment of the three-dimensional display subsystem is a high-performance development workstation, and the software environment is a UE simulation system.

[0052] The environment modeling subsystem virtually reconstructs the physical environment as much as possible, including the virtual construction of the environment's three-dimensional terrain, electromagnetic environment, and network environment, and finally generates a UE simulation environment. This subsystem includes a UE development engine and a variety of environment material packages under the UE development engine, and builds an environment modeling unit based on the UE development engine and the environment material package.

[0053] The modeling subsystem of the intelligent control system virtualizes the intelligent control system while retaining the actual operation control and communication mechanism of the intelligent control system. The virtual intelligent control system model can obtain various variables of the external environment, be affected by the environment, and make different feedbacks, thereby reflecting the dynamic changes of the real intelligent control system during flight operation as much as possible.

[0054] The modeling subsystem of the intelligent control system includes a task generation unit and a self-developed control system software, AirSim simulation software and its material library, blueprints and environment variables, the task generation unit generates tasks and sends them to the control software, the control software generates task instructions and sends them to the AirSim simulation software based on the secondary development of Airsim, and then obtains the model power data and sends them to the UE simulation environment. The AirSim simulation software represents a real intelligent control system. The model power data is the real motion state information of the real intelligent control system. When the AirSim simulation software is a drone, the model power data is the motion state data of the drone, for example, the flight speed, flight altitude, flight attitude angle, etc. of the drone. The UE development engine obtains the model power data by calling the search function, forms the denied environment type information, sends the denied environment type information to the UE simulation environment through the simulation interface, and the UE simulation environment responds according to the denied environment type information.

[0055] The physical interaction subsystem connects the UE simulation environment, intelligent control system model, onboard computer, sensor, communication encryption module and other hardware entities into a whole, providing real communication, navigation, identification and other physical support for the semi-physical simulation of intelligent control system. The communication data to be analyzed include: control signals of various buses, such as PWM (pulse width modulation) signal and CAN (controller area network) bus signal.

[0056] The auxiliary test subsystem includes: an oscilloscope, a spectrum analyzer, a programmable power supply, a signal generator, a wireless signal analyzer and a vector network analyzer. The auxiliary test subsystem receives the communication data to be analyzed sent by the control hardware, and analyzes and processes the communication data to be analyzed.

[0057] The model structure of the intelligent control system in the simulation environment is mainly divided into two parts: one is the role model for data interaction with the simulation environment, and the other is the communication model for data interaction with the physical flight controller. Both are uniformly scheduled by the simulation system and expose the necessary parameters and functional interfaces to the simulation system. The intelligent control system role model is responsible for controlling the behavior of the intelligent control system role in the 3D engine. Its main members include various types of information required for interaction with the simulation environment:

[0058] 1) Basic information: Contains the identity information of this role that the simulation environment recognizes.

[0059] 2) Denial environment information: Contains information about the types of denial interference that exist in the simulation environment where the character is located.

[0060] 3) Environmental information: Contains physical information such as wind resistance, air pressure, temperature, etc. in the simulated environment where the character is located.

[0061] 4) Physical structure: Contains the physical structure information of the intelligent control system corresponding to the character, such as wheelbase, torque, weight, etc.

[0062] 5) Viewpoint information: Contains information about the character's observation viewpoint, that is, the angle from which the user can observe the simulation model.

[0063] 6) Event type: includes special event types generated by characters in the simulation environment, such as collision, loss of control, etc.

[0064] 7) Functional interface: that is, the method functions that the role model opens to the simulation system. The simulation system controls the behavior of the role model in the simulation environment through various method functions.

[0065] The communication model of the intelligent control system transmits sensor information in the UE simulation environment to the controller through the communication protocol, and receives the control signal sent by it. Its main members include various types of information required for data exchange with the controller:

[0066] 1) Basic information: Contains the identity information of the intelligent control system that interacts between the simulation system and the controller hardware.

[0067] 2) Denied environment information: Contains the denied area information for interaction between the simulation system and the controller hardware

[0068] 3) Link information: Contains the communication link information between the simulation system and the controller hardware.

[0069] 4) Policy information: Contains task strategy information for interaction between the simulation system and the controller hardware.

[0070] 5) Control information: Contains controller control information sent by the controller hardware to the simulation system.

[0071] 6) Sensor information: Contains simulated sensor information sent by the simulation system to the controller hardware.

[0072] The virtual environment construction process includes:

[0073] 1) Build a virtual scene. The main tools needed to build a virtual scene include: resource deployment tools, terrain modification tools, physical environment configuration tools, etc. They can be used to complete the resource deployment of the virtual scene, and modify the terrain, etc.

[0074] Methods for deploying resources (such as geometry, triggers, objects) in the scene: basic models can be dragged into the scene from the "Mode" edit bar on the left, and imported models can be dragged into the scene from the "Resource Browser" below.

[0075] 2) Migrate the virtual scene. The method of migrating an existing scene to a newly created UE4 project is to copy the resources in the "content" folder of the original scene project directly to the "content" folder of the new project, and then find the corresponding map file in the "Content Browser" column in the UE4 development interface to open the scene. Compared with the ROS system that only supports deployment and operation in the Linux operating system, the UE4-based virtual scene system in the present invention can achieve convenient migration and deployment of the simulation environment under Windows through simple steps of copying directories and loading project files.

[0076] 3) Deployment and configuration of the simulation system. The simulation system meets the requirements of control system firmware software, hardware-in-the-loop, and physical and visual realistic simulation, and configures the simulation system settings according to the corresponding safety test requirements.

[0077] 4) Modify the configuration file. Modifying the AirSim simulation software configuration file can change system parameters including simulation vehicle (intelligent control system), view mode, placement time, etc.

[0078] The virtual environments of different scenarios are all configured during the development phase, and the rejection factors are preset in specific areas of the scenarios. They have the characteristics of fast startup, fast generation, and automatic connection to flight control equipment, realizing rapid simulation of complex physical scenarios.

[0079] The interaction process between the intelligent control system and the virtual simulation platform is the interaction process from the task generation unit to the UE simulation environment, including:

[0080] First, the deployment and configuration of the UE simulation system are completed using the environment modeling unit;

[0081] Then, simulating a virtual environment using the UE simulation environment, and configuring setting items of the UE simulation environment;

[0082] The intelligent control system model enters the virtual virtual denied environment, and the UE development engine configures the denied environment type information of the UE simulation environment;

[0083] After the intelligent control system model enters the denied environment, the UE development engine is triggered. The UE development engine is set in the denied electromagnetic network, and the search function is called. The UE development engine obtains the model power data. The UE development engine sends the denied environment type information to the UE simulation environment through the simulation interface, and the UE simulation environment responds according to the denied environment type information;

[0084] The intelligent control system model responds with different strategies according to the denied environment type information and executes corresponding tasks;

[0085] Finally, according to the task status of the intelligent control system: verify the security of the intelligent control system.

[0086] Three embodiments have been carried out using the present invention.

[0087] Example 1: The virtual scene constructed is a mountainous desert terrain. There is an enemy military base in the scene, and the surrounding space is deployed with a navigation and positioning deception denial environment. When a drone flies into this area, the navigation and positioning information will be affected by the denial environment and deviate, and the drone will not be able to fly normally and will eventually be forced to land.

[0088] Example 2: The virtual scene constructed is an ocean island terrain. There is an enemy military base in the scene, and the surrounding space is deployed with a sensor interference denial environment. When a drone flies into this area, the sensor data will be affected by the denial environment and produce huge noise. The drone will not be able to fly normally and will eventually fall.

[0089] Example 3: The constructed virtual scene is an urban terrain with a large number of buildings in the scene. A communication suppression denial environment is deployed within a certain spatial range. When a drone flies into the area, the communication link will be interrupted by the denial environment, and the drone will not be able to communicate normally with the ground station.

[0090] In a method and device for verifying the security of an intelligent control system provided by an embodiment of the present invention, the device for verifying the security of an intelligent control system includes: an intelligent control system modeling subsystem, a three-dimensional display subsystem, an environmental modeling subsystem, a physical interaction subsystem and an auxiliary test subsystem. The intelligent control system modeling subsystem is used to simulate a real intelligent control system to obtain an intelligent control system model; the three-dimensional display subsystem is used to display a three-dimensional virtual rejection environment, the position and posture of the intelligent control system model, etc., so as to facilitate a more intuitive observation of the system security test process; the environmental modeling subsystem is used to generate a UE simulation environment in the three-dimensional display subsystem, including the rejection environment therein; the physical interaction subsystem realizes a security test environment in the form of semi-physical simulation by connecting the three-dimensional display subsystem, the hardware equipment of the real intelligent control system and the auxiliary test subsystem; the auxiliary test subsystem is used to analyze the communication data in the operation of the control hardware of the intelligent control system, and obtain information data used to evaluate the system security during the system security test process. The workflow of the method and device for verifying the security of an intelligent control system is as follows: security test environment configuration: first, the environment modeling unit of the environment modeling subsystem is used to complete the deployment and configuration of the UE simulation environment of the three-dimensional display subsystem, and then, the UE simulation environment is used to simulate the rejection environment, and the setting items of the UE simulation environment are configured according to the security test requirements; then, the three-dimensional display subsystem, the real hardware equipment of the intelligent control system and the auxiliary test subsystem are connected through the physical interaction subsystem to realize a security test environment in the form of semi-physical simulation; security test process: the intelligent control system model enters the virtual rejection environment, and its task generation unit configures the rejection type of the UE simulation environment; After the control system model enters the denied environment, it triggers the trigger in the denied electromagnetic network set by the UE development engine in the UE simulation environment, and then calls the search function to enable the UE development engine to obtain the role identity information of the intelligent control system and the simulation interface in the role identity information. The UE development engine transmits the type information of the denied environment to the UE simulation environment through the simulation interface, and then the UE simulation environment responds according to the denied environment type information; then, the intelligent control system model responds with different strategies according to the denied type and executes corresponding tasks; the security capability of the intelligent control system is verified according to the task status of the denied environment of the intelligent control system and the test result information data output by the auxiliary test subsystem.

[0091] A method and device for verifying the security of an intelligent control system of the present invention can achieve the following functions:

[0092] 1. The UE simulation environment of the three-dimensional display subsystem is based on the UE development engine, which can realize the realistic simulation of the three-dimensional terrain, electromagnetic environment, and network environment of the real environment; the denial environment adopts the hardware-in-the-loop mode to ensure that the control software working on the hardware equipment of the intelligent control system receives the virtual environment data provided by the UE simulation environment according to the fixed timing logic, and processes it as a real external environment, calculates the corresponding system status data under the actual conditions, and realizes the real-time transmission of data between the control hardware and the UE simulation environment.

[0093] 2. Conducting preliminary security tests on intelligent control systems in a simulated environment first can effectively avoid risks in physical testing. By building a UE simulation environment in a targeted and purposeful manner, the motion state of the intelligent control system in a specific environment can be tested, breaking away from the spatial limitations of physical testing. In addition, according to specific security testing requirements, interference factors can be artificially added to build a denial environment, simulate confrontation scenarios, and test the security execution of intelligent control systems for specific tasks.

[0094] The hardware-in-the-loop simulation subsystem of the device of the present invention includes a physical intelligent control system and a virtual simulation platform. The virtual environment, intelligent control system model and actual hardware entities such as radio stations and sensors are connected into a whole to provide real physical support such as communication, navigation, and identification. The intelligent control system may include a control panel, an onboard computer, ground station equipment, a communication encryption module, and auxiliary sensors such as binocular cameras, laser radars, and satellite antenna arrays. The virtual simulation platform may include three main models: an intelligent control system model, an environmental model, and a rejection environment model. The purpose of constructing the intelligent control system model is to make the intelligent control system in the virtual simulation platform consistent with the physical intelligent control system function. The purpose of constructing the environmental model is to make the environment in which the intelligent control system in the virtual simulation platform is located consistent with the real environment. The purpose of constructing the rejection environment model is to verify the security of the intelligent control system system in the virtual simulation platform. The virtual simulation platform is mainly developed based on AirSim. AirSim is an open source platform for simulators such as drones and cars built on the UE4 engine, and can be simulated and controlled through the PX4 flight controller across platforms, with a physically and visually realistic simulation environment. Not only does it simulate the drone dynamics model, but it also makes a very good simulation of the weather effect lighting control. In addition, AirSim provides a large number of C++ and Python programming APIs, which greatly facilitates secondary development by different projects and developers.

[0095] The method and device for verifying the security of an intelligent control system described in this application can achieve the following effects:

[0096] (1) Without the need for physical operation, the system can quickly and effectively test and verify the intelligent control system controller under development, reduce the testing time required for algorithm improvement, and improve development efficiency.

[0097] (2) It supports in-the-loop testing after the control algorithm is deployed on hardware. Compared with full software simulation, it is closer to actual operation and the test results are more reliable.

[0098] (3) Preliminary testing of the functionality and safety of intelligent control system controllers in a simulation environment is prioritized to effectively avoid risks in actual operation.

[0099] (4) During the simulation process, developers can communicate with the intelligent control system hardware in real time, which makes data collection more convenient and is more conducive to real-time verification of the correctness and effectiveness of the model and algorithm.

[0100] (5) By constructing virtual scenes in a targeted and purposeful manner, the flight performance of intelligent control systems in specific environments can be tested, thus breaking away from the spatial limitations of physical operation.

[0101] According to needs, interference factors can be added artificially to build a denial environment, simulate physical confrontation scenarios, and test the execution of intelligent control systems for specific tasks.

[0102] It is obvious to those skilled in the art that the embodiments of the present invention are not limited to the details of the above exemplary embodiments, and that the embodiments of the present invention can be implemented in other specific forms without departing from the spirit or basic features of the embodiments of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive, and the scope of the embodiments of the present invention is limited by the attached claims rather than the above description, so it is intended to include all changes that fall within the meaning and scope of the equivalent elements of the claims in the embodiments of the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved. In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units, modules or devices stated in the system, device or terminal claims can also be implemented by the same unit, module or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0103] Finally, it should be noted that the above implementation modes are only used to illustrate the technical solutions of the embodiments of the present invention and are not intended to limit them. Although the embodiments of the present invention have been described in detail with reference to the above preferred implementation modes, those skilled in the art should understand that the technical solutions of the embodiments of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A device for verifying the security of an intelligent control system, characterized in that: include: Intelligent control system modeling subsystem, 3D display subsystem, environment modeling subsystem, physical interaction subsystem and auxiliary testing subsystem; The modeling subsystem of the intelligent control system includes: a task generation unit, control software and AirSim simulation software; The task generation unit generates a task and sends it to the control software. The control software generates a task instruction according to the task and sends it to the AirSim simulation software. The AirSim simulation software generates a system instruction according to the task instruction and sends it to the physical interaction subsystem, and receives the system status data returned by the physical interaction subsystem. The AirSim simulation software sends the model power data to the three-dimensional display subsystem and receives the virtual environment data sent by the three-dimensional display subsystem. The three-dimensional display subsystem includes a workstation and a UE simulation environment, and the workstation is equipped with and runs the UE simulation environment; The UE simulation environment sends the virtual environment data to the AirSim simulation software and the physical interaction subsystem respectively, and receives the model power data and the system status data respectively; The environment modeling subsystem includes a UE development engine and an environment modeling unit; the UE development engine sends an environment material package to the environment modeling unit, the environment modeling unit generates the UE simulation environment using the environment material package, and sends the rejection environment type information to the UE simulation environment; The physical interaction subsystem collects raw data, receives the virtual environment data sent by the UE simulation environment, receives the system instructions and the system status data, and analyzes the raw data to obtain parsed data. The physical interaction subsystem processes the parsed data into the system status data and communication data to be analyzed, and sends the system status data to the UE simulation environment and the AirSim simulation software, and sends the communication data to be analyzed to the auxiliary test subsystem; The auxiliary test subsystem receives the communication data to be analyzed sent by the control hardware, and analyzes and processes the communication data to be analyzed; The modeling subsystem of the intelligent control system is used to virtualize the intelligent control system into an intelligent control system model; The interaction process between the intelligent control system modeling subsystem and the UE simulation environment includes: When the intelligent control system model enters the virtual denied environment, the task generation unit configures the denied environment type, converts the denied environment type into the task, and sends it to the control software; The control software converts the task into the task instruction and sends it to the Airsim simulation software. After receiving the task instruction, the AirSim simulation software sends the model power data to the UE simulation environment, so that the UE simulation environment triggers the UE development engine through the simulation interface; The UE development engine calls a search function to obtain the model power data, and sends the denied environment type information of the virtual denied environment to the UE simulation environment through the simulation interface, and the UE simulation environment responds according to the denied environment type information; After the UE simulation environment responds according to the denied environment type information, the Airsim simulation software responds to different strategies according to the denied environment type information and executes the corresponding tasks; and verifies the security of the intelligent control system according to the task status of the intelligent control system model.

2. The device for verifying the security of an intelligent control system according to claim 1, characterized in that: The physical interaction subsystem includes a sensor, an onboard computer, control hardware and a communication encryption module; the sensor collects raw data and sends the raw data to the onboard computer, the onboard computer analyzes the raw data to obtain parsed data, the onboard computer sends the parsed data to the control hardware, the control hardware processes the parsed data into the system status data and the communication data to be analyzed, and sends the system status data to the communication encryption module, and sends the communication data to be analyzed to the auxiliary test subsystem; the control hardware receives the virtual environment data sent by the UE simulation environment; The communication encryption module receives the system instructions and the system status data, encrypts the system status data and the system instructions to obtain the system status data; and sends the system status data to the AirSim simulation software and the UE simulation environment respectively.

3. The device for verifying the security of an intelligent control system according to claim 2, characterized in that: The auxiliary test subsystem includes: an oscilloscope, a spectrum analyzer, a programmable power supply, a signal generator, a wireless signal analyzer and a vector network analyzer; the environmental modeling unit includes tools, which include: a resource deployment tool, a terrain modification tool and a physical environment configuration tool; the tools are used to complete the resource deployment of the virtual battlefield scene and the modification of the denial environment factors.

4. The device for verifying the security of an intelligent control system according to claim 3, characterized in that: The environment modeling unit generates the UE simulation environment by using the tool, and obtains the virtual environment data by simulation; and configures the UE simulation environment according to the security test requirements; The UE simulation environment modifies a configuration file of the AirSim simulation software by sending the virtual environment data to the AirSim simulation software; The configuration file includes: simulation vehicle, view mode and placement time.

5. The device for verifying the security of an intelligent control system according to claim 4, characterized in that: The environmental material package includes: the virtual battlefield scenes of ocean, mountain and city and the denied environmental factors; the communication data to be analyzed includes: pulse width modulation signal and controller area network bus signal.

6. The device for verifying the security of an intelligent control system according to claim 5, characterized in that: The virtual denied environment adopts a hardware-in-the-loop mode to ensure that the software working on the control hardware receives the virtual environment data sent by the UE simulation environment, processes the virtual environment data as real external environment data, and calculates the corresponding system status data of the real external environment.

7. A method for verifying the security of an intelligent control system, characterized in that: The method adopts the device for verifying the security of an intelligent control system according to any one of claims 1 to 6, and the method comprises the following steps: First, the deployment and configuration of the UE simulation system are completed using the environment modeling unit; Then, simulating a virtual environment using the UE simulation environment, and configuring setting items of the UE simulation environment; The intelligent control system model enters a virtual denied environment, and the UE development engine configures the denied environment type information of the UE simulation environment; After the intelligent control system model enters the virtual denied environment, the UE development engine is triggered, the UE development engine calls a search function to obtain the model power data, the UE development engine sends the denied environment type information to the UE simulation environment through a simulation interface, and the UE simulation environment responds according to the denied environment type information; The intelligent control system model responds with different strategies according to the denied environment type information and executes corresponding tasks; Finally, according to the task status of the intelligent control system: verify the security of the intelligent control system.

Images