Safety system of all-vector control automobile, detection method thereof and storage medium

Abstract

The application relates to the technical field of full-vector control automobiles, in particular to a safety system of a full-vector control automobile, a detection method thereof and a storage medium, wherein the safety system comprises an electrical layer, a hardware layer and a software layer; the electrical layer is used for realizing the independent power supply between various controllers, sensors and actuators of the full-vector control automobile; the hardware layer is used for shielding the differences between different hardware in the hardware layer; the software layer is used for shielding the differences between different hardware in the hardware layer, making the upper application layer universal, and realizing preset driving of the vehicle; the system layer is used for realizing task allocation and scheduling, and processing requests of access and communication of external signals through a real-time operating system; and the algorithm layer comprises an automatic driving algorithm, a chassis controller safety algorithm and a bottom actuator algorithm. Therefore, the problems that the chassis controller of the full-vector control automobile cannot meet the requirements of safety performance and computing performance in the prior art, the safety, real-time performance and expandability of the vehicle are poor, and the work efficiency is reduced are solved.

Description

Technical Field

[0001] This application relates to the field of full vector control automotive technology, and in particular to a safety system for full vector control vehicles, its detection method, and storage medium. Background Technology

[0002] Compared to traditional vehicles, fully vector-controlled cars offer significantly increased degrees of freedom. The fully vector-driven chassis features four-wheel distributed drive, four-wheel independent steering control, four-wheel independent suspension adjustment, and four-wheel independent brake-by-wire, providing the vehicle with greater controllability and resulting in higher maneuverability and control redundancy. However, to meet the requirements of fully vector control, more comprehensive observation of the vehicle's motion and the use of more complex algorithms for control are necessary. Therefore, fully vector-controlled cars require more powerful computing capabilities and higher safety performance.

[0003] In related technologies, the computing performance requirements of full vector control vehicles are met by improving the computing performance of the chassis controller. However, in the current autonomous driving architecture, the amount of data that vehicles need to process is growing explosively. For example, with the addition of multiple sensors such as radar and cameras, the amount of data that vehicles need to process has increased dramatically. Simply improving the computing performance of the chassis controller is far from enough, and it cannot improve safety or meet the needs of intelligence. Summary of the Invention

[0004] This application provides a safety system for a fully vector-controlled vehicle, its detection method, and storage medium, to solve the problems in related technologies where the fully vector-controlled vehicle chassis controller cannot meet the requirements of safety and computing performance, resulting in poor vehicle safety, real-time performance, and scalability, and reduced work efficiency.

[0005] The first aspect of this application provides a safety system for a fully vector-controlled vehicle, comprising: an electrical layer for achieving independent power supply between the chassis controller, driving controller, driving sensors, and actuators of the fully vector-controlled vehicle; and a hardware layer including a main control chip of the chassis controller, an extended acceleration chip of the autonomous driving controller, external hardware corresponding to the driving sensors and the actuators, wherein the driving sensors are connected to the autonomous driving controller via a preset high-speed serial port and / or Ethernet, one or more preset communication and security modules are mounted on the internal bus of the main control chip, and the main control chip is externally connected to a power supply chip and a communication interface for communication between the lower-level actuators and the upper-level collaborative acceleration, and the various actuators of the chassis are connected to the chassis controller. Mounted on the vehicle network, the autonomous driving controller communicates with the vehicle network; the software layer is used to shield the differences between different hardware in the hardware layer, making the upper application layer universal and implementing one or more preset drives for the vehicle; the system layer is used to implement task allocation and scheduling, as well as the processing of external signal access and communication requests through a real-time operating system, and to complete timing allocation through time and event-triggered interrupts to coordinate multiple tasks on the vehicle chassis controller in parallel; the algorithm layer includes an autonomous driving algorithm, a chassis controller safety algorithm, and a low-level actuator algorithm, which realize the full-vector control function execution of the vehicle chassis controller through the autonomous driving algorithm, the chassis controller safety algorithm, and / or the low-level actuator algorithm.

[0006] Optionally, the chassis controller is connected to each actuator of the chassis via a brake bus, a drive bus, an attitude bus, and an attitude redundancy bus. When the attitude bus fails, the attitude redundancy bus can control the vehicle's attitude. When the drive bus fails, the brake bus and the attitude bus assist the vehicle in safely stopping. When the brake bus fails, the drive bus controls the wheel hub motors to achieve emergency braking.

[0007] Optionally, the chassis controller collects signals from chassis sensors using IO (Input / Output) interfaces, AD interfaces, and / or CC interfaces.

[0008] Optionally, the chassis controller can perform collaborative calculations with the high-performance computing unit for autonomous driving through a preset high-speed interface to achieve hardware expansion and acceleration of the chassis controller.

[0009] Optionally, the high-performance computing unit for autonomous driving includes an FFT (fast Fourier transform) coprocessor designed using an FPGA (Field Programmable Gate Array).

[0010] Optionally, the software layer includes a software layer for the autonomous driving controller design, a software layer for the design of other controller chips, and complex drivers. The software layer for the autonomous driving controller includes a base layer and a service layer. Each function in the layer consists of a functional cluster. The service layer includes the allocation of storage modules, reusable interfaces for various signals and communications, and the implementation of corresponding underlying safety function modules. In the software design of the other chips, a hardware abstraction layer defines the general hardware pin function allocation and basic communication interfaces.

[0011] Optionally, the chassis controller includes multiple monitoring modules from top to bottom. At the application layer, functional monitoring is achieved through the chassis controller. By verifying signals and combining the state information of the driver, vehicle, and road surface obtained from model calculations, the system's operating status is monitored at the algorithm level. At the system layer, the operating system monitors the running time and memory of each program segment. At the software layer, lockstep verification of the lockstep core and memory verification are performed by calling the values ​​of safety-related registers. At the hardware layer, the chip's own security unit is used to achieve one or more of the following functions: operating temperature monitoring, lockstep verification, memory diagnostics, and software watchdog timer.

[0012] A second aspect of this application provides a fully vector control vehicle, including the safety system for a fully vector control vehicle as described in the above embodiments.

[0013] A third aspect of this application provides a detection method for a safety system of a fully vector-controlled vehicle. The method is applied to the safety system of the fully vector-controlled vehicle as described in the above embodiments. The method includes the following steps: acquiring a target to be detected in the fully vector-controlled vehicle; acquiring time-domain information and / or frequency-domain information of one or more target-related hardware in the safety system based on the target to be detected; identifying the time-domain information and / or the frequency-domain information to obtain target features of the target to be detected; and detecting the target to be detected based on the target features.

[0014] Optionally, when the target to be detected is a road surface, the step of identifying the target features of the target to be detected by recognizing the time-domain information and / or the frequency-domain information, and detecting the target to be detected based on the target features, includes: monitoring the wheel speed signal of the tire using the frequency-domain method; recognizing the wheel speed signal to obtain the road surface features; and when the road surface is identified as a preset uneven road surface based on the road surface features, matching the actual type of the preset uneven road surface based on the road surface features.

[0015] Optionally, when the target to be detected is tire pressure, the step of identifying the target features of the target to be detected by recognizing the time-domain information and / or the frequency-domain information, and detecting the target to be detected based on the target features, includes: monitoring the wheel speed signal of the tire using the frequency domain method; recognizing the wheel speed signal to obtain the actual frequency of the tire; and mapping the actual frequency to the tire pressure according to a preset relationship between the tire's natural frequency and tire pressure, thereby realizing the monitoring of the tire pressure.

[0016] Optionally, when the target to be detected is a motor, the step of identifying the target features of the target to be detected by recognizing the time-domain information and / or the frequency-domain information, and detecting the target to be detected based on the target features, includes: recognizing the time-domain information and frequency-domain information of the motor bearing to obtain the motor's rotational speed pulse signal and vibration signal; generating the rotational frequency curve of the motor bearing based on the rotational speed pulse signal and the vibration signal, and segmenting the vibration signal at a preset stable speed from the rotational frequency curve; determining health indicators and fault frequencies based on the segmented vibration signals, matching the fault level of the motor bearing with the health indicators, and locating the fault location of the motor bearing with the fault frequency.

[0017] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement a detection method for a safety system of a fully vector-controlled vehicle as described in any of the preceding claims.

[0018] Therefore, this application has at least the following beneficial effects:

[0019] This application proposes a hardware and software architecture and hardware acceleration method for a full-vector control vehicle chassis controller. At the hardware level, based on high-speed interfaces such as SPI (Serial Peripheral Interface) and PCIe (Peripheral Component Interconnect Express), a scheme is implemented that supports extended computing by FPGA and GPU (Graphics Processing Unit) acceleration units, thus balancing efficiency and cost and improving vehicle control capabilities while ensuring safety. At the safety software architecture level, based on the collaborative computing of the autonomous driving controller and the full-vector control vehicle chassis controller, chassis signal information fusion and deep processing can be achieved. This improves the safety and computing power of the full-vector control vehicle chassis controller while enhancing the vehicle's scalability, meeting the future intelligent development needs of full-vector control vehicles.

[0020] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0021] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0022] Figure 1 This is a block diagram of a full-vector control vehicle safety system according to an embodiment of this application;

[0023] Figure 2 This is a schematic diagram of the overall scheme of the full vector control vehicle safety system provided according to the embodiments of this application;

[0024] Figure 3 This is a diagram illustrating the overall hardware and software architecture of a fully vector-controlled vehicle according to an embodiment of this application.

[0025] Figure 4 This is a flowchart illustrating the detection method for a full-vector control vehicle safety system according to an embodiment of this application. Detailed Implementation

[0026] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0027] The following description, with reference to the accompanying drawings, illustrates a safety system for a fully vector-controlled vehicle, its detection method, and a storage medium. Addressing the issues mentioned in the background art where fully vector-controlled vehicle chassis controllers fail to meet safety and computational performance requirements, resulting in poor vehicle safety, real-time performance, and scalability, and reduced work efficiency, this application provides a safety system for a fully vector-controlled vehicle. In this system, the hardware and software of the fully vector-controlled vehicle chassis controller directly cooperate, significantly improving the computing power of the controller and enhancing vehicle scalability while ensuring vehicle safety and real-time performance. This solves the problems in related technologies where fully vector-controlled vehicle chassis controllers fail to meet safety and computational performance requirements, leading to poor vehicle safety, real-time performance, and scalability, and reduced work efficiency.

[0028] Specifically, Figure 1 This is a block diagram illustrating a full-vector control vehicle safety system provided in an embodiment of this application.

[0029] like Figure 1 As shown, the full vector control vehicle safety system 10 includes: an electrical layer 100, a hardware layer 200, a software layer 300, a system layer 400, and an algorithm layer 500.

[0030] The electrical layer 100 enables independent power supply between the chassis controller, driving controller, driving sensors, and actuators of the fully vector-controlled vehicle. The hardware layer 200 includes the main control chip of the chassis controller, the extended acceleration chip of the autonomous driving controller, and external hardware corresponding to the driving sensors and actuators. The driving sensors are connected to the autonomous driving controller via a preset high-speed serial port and / or Ethernet. One or more preset communication and security modules are mounted on the internal bus of the main control chip. The main control chip is externally connected to a power chip and a communication interface for communication between the lower-level actuators and the upper-level collaborative acceleration. The various actuators of the chassis are connected to the chassis controller via the vehicle network. The controller communicates with the vehicle network; the software layer 300 is used to shield the differences between different hardware in the hardware layer, so that the upper application layer has universality and implements one or more preset drives of the vehicle; the system layer 400 is used to implement task allocation and scheduling, as well as the processing of external signal access and communication requests through the real-time operating system, and completes timing allocation through time and event-triggered interrupts to coordinate multiple tasks on the vehicle chassis controller in parallel; the algorithm layer 500 includes autonomous driving algorithms, chassis controller safety algorithms and underlying actuator algorithms, and implements the full vector control function execution of the vehicle chassis controller through autonomous driving algorithms, chassis controller safety algorithms and / or underlying actuator algorithms.

[0031] The preset high-speed serial port can be a high-speed interface such as SPI or PCIe, and no specific limitation is made here.

[0032] Among them, the preset communication and security modules can be pre-defined by the user and are used to collect signals from chassis sensors, monitor temperature, perform lockstep verification, memory diagnostics, and software watchdog. They can be set or adjusted according to the actual situation, and no specific limitations are made here.

[0033] Among them, the preset driver can realize the development of functions of the internal system security control unit module of the chip, the driver program of the external power chip, and hardware acceleration expansion, etc., without specific limitations.

[0034] It is understood that the embodiments of this application realize the hardware and software framework of the full vector control vehicle through five layers: electrical layer, hardware layer, software layer, system layer, and algorithm layer. The electrical layer is used to ensure the independence of power supply for each part, so that the hardware does not interfere with each other in use, which facilitates the safety control of the whole vehicle. The hardware layer is used to control the communication and safety modules between various devices. The software layer is used to shield the differences between different hardware and improve the versatility of the upper application layer. The system layer is used to realize the vehicle's multi-task parallel coordination function. The algorithm layer is used to realize the model calculation of vehicle-road-driver, state observation, and precise control of actuators, and realizes multiple functions such as safety monitoring and hardware expansion acceleration in the chassis controller. Overall, it improves the safety and computing power of the full vector control vehicle chassis controller while improving the vehicle's scalability, meeting the needs of the future intelligent development of full vector vehicles.

[0035] In this embodiment, the chassis controller is connected to various actuators of the chassis via a brake bus, a drive bus, an attitude bus, and an attitude redundancy bus. When the attitude bus fails, the attitude redundancy bus enables control of the vehicle's attitude. When the drive bus fails, the brake bus and attitude bus assist the vehicle in safely stopping. When the brake bus fails, the drive bus controls the wheel hub motors to achieve emergency braking of the vehicle.

[0036] It is understood that the four buses of the chassis controller in this application embodiment can ensure that the vehicle achieves redundant safety and multi-degree-of-freedom control in terms of architecture. When the attitude bus fails, the vehicle body attitude can be controlled through the redundant attitude bus, which can solve the overall failure caused by the bus failure. When the drive bus fails, the vehicle can be safely stopped through the braking and attitude buses. When the braking bus fails, the vehicle can be braked in an emergency by controlling the motor through the drive bus. This solves the problem of the vehicle losing overall control due to the bus failure and effectively ensures the safety of the fully vector control vehicle.

[0037] Specifically, such as Figure 2As shown, the four CAN (Controller Area Network) channels of the full-vector control vehicle chassis controller are connected to the various actuators of the chassis. Functionally, they can be divided into braking, drive, attitude, and attitude redundancy CAN channels, achieving redundant safety and multi-degree-of-freedom control in the vehicle's architecture. The steering and suspension modules of the full-vector control vehicle are connected to the chassis controller via the attitude CAN. The redundant attitude CAN can resolve overall failures caused by CAN line failures: when the drive CAN fails, the normal operation of the steering and braking CAN channels can assist the vehicle in safely stopping; when the braking CAN fails, the full-vector control vehicle can control the wheel hub motors through the drive CAN, thereby ensuring emergency braking. The braking system includes multiple solutions such as iBosster (electronic brake assist system), ESC (electronic stability control system), and parking brake; the steering system includes multiple solutions such as front-wheel steering, rear-wheel steering, and safe redundant four-wheel steering, effectively solving the problem of loss of steering or braking ability due to the failure of any one solution.

[0038] In this embodiment, the chassis controller collects signals from chassis sensors using IO interfaces, AD interfaces, and / or CC interfaces.

[0039] It is understood that in the embodiments of this application, the chassis controller collects signals from the chassis sensors through the IO interface, AD interface and / or CC interface, which can provide accurate chassis data for the autonomous driving algorithm in real time.

[0040] In this embodiment, the chassis controller performs collaborative calculations with the high-performance computing unit for autonomous driving through a preset high-speed interface, thereby achieving hardware expansion and acceleration of the chassis controller.

[0041] The high-performance computing unit for autonomous driving includes an FFT coprocessor designed with FPGA. This unit comprises two parts, as follows: Figure 2 The Zynq UltraScale+ section shows the heterogeneous acceleration unit and the high-performance CPU, respectively. The heterogeneous acceleration unit uses FPGA to design FFT for hardware expansion acceleration.

[0042] It should be noted that, since FPGAs are highly programmable, different acceleration units can be designed according to requirements to accelerate different algorithms. In this application, FFT can be used for acceleration, so FFT is used as an example.

[0043] FPGA is a product of further development based on programmable devices such as PAL (Programmable Array Logic) and GAL (General Purpose Array Logic). It is designed as a semi-custom circuit in the field of Application-Specific Integrated Circuits (ASIC), which solves the shortcomings of custom circuits and overcomes the limitation of the limited number of gate circuits in the original programmable devices.

[0044] Among them, FFT is a general term for efficient and fast calculation methods for the Discrete Fourier Transform (DFT) using computers. This algorithm can greatly reduce the number of multiplications required for the computer to calculate the Discrete Fourier Transform. In particular, the more sampling points N that are being transformed, the more significant the saving of computational load of the FFT algorithm becomes.

[0045] It is understood that in the embodiments of this application, the full vector control vehicle chassis controller can perform collaborative calculations with the high-performance computing unit for autonomous driving through high-speed interfaces such as SPI and PCIe to achieve hardware expansion and acceleration of the chassis controller; and an FPGA is used to design an FFT coprocessor to assist the full vector control vehicle chassis controller in accelerating frequency domain correlation calculations, thereby improving real-time performance and computational efficiency.

[0046] Specifically, in practical applications of signal filtering and cross-correlation spectrum analysis, STFT (Short Time Fourier Transform) can be converted into FFT for calculation after the signal is multiplied by a window function, and the inverse Fourier transform can also be converted into an algorithm similar to FFT for calculation. Therefore, this module has a certain degree of versatility and can effectively accelerate the information processing and calculation on the chassis domain controller.

[0047] In this embodiment, the software layer includes a software layer for the autonomous driving controller design, a software layer for the remaining controller chip designs, and complex drivers. The software layer for the autonomous driving controller software design includes a base layer and a service layer. Each function in the layer consists of a functional cluster. The service layer includes the allocation of storage modules, reusable interfaces for various signals and communications, and the implementation of corresponding underlying safety function modules. In the software design of the remaining chips, a hardware abstraction layer defines the general hardware pin function allocation and basic communication interfaces.

[0048] It is understood that the software layer in this application embodiment can shield the differences between different hardware, improve the versatility of the upper application layer, and realize the modularization of various basic functions.

[0049] In this embodiment, the chassis controller includes multiple monitoring modules from top to bottom. At the application layer, functional monitoring is achieved through the chassis controller. By verifying signals and combining the state information of the driver, vehicle, and road surface obtained from model calculations, the system's operating status is monitored at the algorithm level. At the system layer, the operating system monitors the runtime and memory of each program segment. At the software layer, lockstep verification and memory verification of the lockstep core are performed by calling the values ​​of safety-related registers. At the hardware layer, the chip's own security unit is used to implement one or more functions, including operating temperature monitoring, lockstep verification, memory diagnostics, and software watchdog timer.

[0050] It is understood that the embodiments of this application use multiple monitoring modules to monitor the operational status of each layer at the application layer, system layer, software layer, and hardware layer, thereby improving diagnostic coverage, enhancing system reliability, and further ensuring system security.

[0051] The safety system for a fully vector-controlled vehicle proposed in this application implements a hardware and software framework across five layers: electrical, hardware, software, system, and algorithm. This overall system enhances the safety and computing power of the fully vector-controlled vehicle chassis controller while ensuring vehicle safety and real-time performance, improving vehicle scalability, and meeting the future intelligent development needs of fully vector-controlled vehicles. Therefore, it solves the problems in related technologies where fully vector-controlled vehicle chassis controllers cannot meet the requirements for safety and computing performance, resulting in poor vehicle safety, real-time performance, and scalability, and reduced work efficiency.

[0052] The following will combine Figure 3 The safety system for a fully vector-controlled vehicle proposed in this application is described in detail. This system comprises five layers: electrical layer, hardware layer, software layer, system layer, and algorithm layer, as detailed below:

[0053] The lowest layer is the vehicle's low-voltage electrical layer. To ensure the electrical safety of the entire vehicle, the power supply of the four main components—chassis controller, autonomous driving controller, autonomous driving sensors, and traditional actuators—is independent and does not interfere with each other in terms of hardware. This effectively reduces sensor errors caused by voltage fluctuations and ensures the normal operation of redundant functions, facilitating the safe control of the entire vehicle. In order to prevent damage to the system from short circuits, reverse connections, and other operations, corresponding electrical safety measures such as emergency stop and air switches are also designed.

[0054] The next layer is the system's hardware layer, which includes: the chassis controller main control chip, an autonomous driving controller chip that can be used for expansion and acceleration, various sensors, and traditional actuators. The LiDAR is connected to the autonomous driving controller via Ethernet, while signals from surround-view cameras, IMU (Inertial Measurement Unit), and GPS (Global Positioning System) enter the autonomous driving controller via high-speed serial ports. The chassis controller chip contains various communication and security modules mounted on its internal bus. Externally, it connects to a high-security power supply chip and a communication interface for coordinating acceleration between the lower-level actuators and the upper-level layers. All chassis actuators and the chassis controller are directly connected to the vehicle's CAN bus. The autonomous driving controller communicates with the vehicle's CAN network via a CAN card installed on the PCIe interface.

[0055] The foundational software layer is the bottom layer of the software. It shields the differences between various hardware components, improving the versatility of the upper application layers. In the software design of an autonomous driving controller, the software can be divided into a foundational layer and a service layer. Each function in the layer consists of a functional cluster, conforming to the Adaptive AUTOSAR specification. In the software design of other chips, a hardware abstraction layer defines the general hardware pin function allocation and basic communication interfaces. The service layer is a modular implementation of various basic functions, including the allocation of storage modules, reusable interfaces for various signals and communications, and the implementation of corresponding low-level safety function modules, conforming to the AUTOSAR (Automotive Open System Architecture) specification. In addition, complex security-related drivers are implemented at this layer, including the development of functions of the internal system security control unit module, the driver program for the external power chip, and hardware acceleration expansion. Since a highly reliable power chip is selected in the hardware layer, special driver logic needs to be designed at the software level to implement relatively complex timing control and information transmission, and to complete the logic of controlling the external power chip and SPI communication. The external power chip monitors the internal software running status through a question-and-answer watchdog mechanism and can reset the system to ensure the safe and reliable operation of the upper-layer program.

[0056] At the system level, the ECU (Electronic Control Unit) at the bottom level mainly uses an RTOS (Real-Time Operating System) and interrupt mechanisms to acquire signals in real time, and then sends them to the CAN bus at fixed time intervals. High-performance sensors for autonomous driving are connected to the autonomous driving controller via different buses. Leveraging the powerful computing capabilities of the autonomous driving controller, multiple signals can be stored and recorded in real time. The message subscription-publishing mechanism pre-installed in Ubuntu is used to timestamp data from multiple different sources, thus solving the data time synchronization problem and enabling real-time vehicle control. The main task of this layer is to solve the problem of multi-task parallel coordination on the full-vector control vehicle chassis controller. For a high-safety controller like the chassis controller, the core of this layer is to ensure the strong real-time performance required for the underlying vehicle operation. This is mainly achieved through a real-time operating system to allocate and schedule tasks for the high-performance core, realize the access of external signals and real-time processing of communication requests, and complete timing allocation through time and event-triggered interrupts. The real-time system can ensure the rapid real-time response to external needs, while also completing the allocation of operating system computing tasks and hardware resources. Running tasks in the real-time operating system can improve the system's time determinism, reduce the impact of additional factors such as interrupts and computing fluctuations, and ensure the priority allocation of each task.

[0057] The topmost algorithm layer is the functional execution layer of the full-vector control vehicle chassis controller. Chassis-related safety and control execution algorithms, as well as autonomous driving-related planning and observation algorithms, run as software functions on this layer. Autonomous driving algorithms primarily fuse data from various vehicle sensors to obtain the real-time state of the vehicle and its surroundings, and use this result to further achieve prediction and planning functions. Chassis-related algorithms mainly implement vehicle-road-driver model solving, state observation, and precise control of actuators, and implement various functions such as safety monitoring and hardware expansion acceleration within the chassis controller. Furthermore, to ensure system safety and real-time performance, the correlation between tasks executed across multiple cores should be low in the task allocation of various algorithms within the chassis controller, minimizing cross-core task scheduling and data sharing to improve system efficiency.

[0058] Since improving diagnostic coverage can enhance system reliability, four monitoring modules were designed for the chassis controller in the overall hardware and software architecture, from top to bottom, located at the application layer, system layer, basic software layer, and hardware layer, respectively:

[0059] At the application layer, functional monitoring is mainly achieved through the chassis controller CPU (Central Processing Unit / Processor). By verifying the signals and combining the state information of the driver, vehicle, and road surface obtained from the model calculation, the system's operating status is monitored at the algorithm level.

[0060] At the system level, the operating system monitors the runtime and memory usage of each program segment to prevent errors such as program deadlocks and timeouts.

[0061] At the basic software layer, the values ​​of security-related registers within the microcontroller are called to ensure that the program and memory are not affected by external factors, including lockstep verification of the lockstep core and memory verification.

[0062] At the hardware layer, the chip's own security unit is used to implement functions such as operating temperature monitoring, lockstep verification, memory diagnostics, and software watchdog. Through the lowest-level hardware monitoring, the system's diagnosis and state switching are realized, ensuring the reliability of hardware computing.

[0063] In summary, at the hardware level, this application's embodiments, based on the chassis controller's high-speed interfaces such as SPI and PCIe, implement a solution that supports extended computing by FPGA and GPU acceleration units, thereby improving work efficiency. At the safety software architecture level, based on the collaborative computing of the autonomous driving controller and the full-vector control vehicle chassis controller, information fusion and deep processing of chassis signals are achieved, which can significantly improve the computing power of the full-vector control vehicle chassis controller and enhance the vehicle's scalability while ensuring vehicle safety and real-time performance.

[0064] This application also provides a full vector control vehicle, including a safety system for a full vector control vehicle as described in any of the above embodiments.

[0065] Next, referring to the accompanying drawings, a detection method for a full-vector control vehicle safety system proposed according to an embodiment of this application is described.

[0066] Figure 4 This is a flowchart of a detection method for a full-vector control vehicle safety system according to an embodiment of this application.

[0067] like Figure 4 As shown, the detection method for the full vector control vehicle safety system includes the following steps:

[0068] In step S101, the target to be detected for the full vector control vehicle is obtained.

[0069] The targets to be detected may include the road surface, tire pressure, and motor, without specific limitations.

[0070] It is understood that the embodiments of this application acquire the target to be detected in the fully vector controlled vehicle so as to subsequently obtain the time-domain and / or frequency-domain information of the relevant hardware in the safety system and analyze and process it.

[0071] In step S102, time-domain information and / or frequency-domain information of one or more target-related hardware in the security system are obtained based on the target to be detected.

[0072] It is understood that the embodiments of this application obtain the time-domain information and / or frequency-domain information of the relevant hardware in the security system from the target to be detected, so as to facilitate the subsequent identification of the time-domain information and / or frequency-domain information to obtain the target features of the target to be detected and to perform detection.

[0073] In step S103, the target features of the target to be detected are obtained by identifying time-domain information and / or frequency-domain information, and the target to be detected is detected based on the target features.

[0074] It is understood that the embodiments of this application obtain the target features of the target to be detected by identifying time-domain information and / or frequency-domain information, and detect the target based on this, which can improve the safety and comfort of full-vector control vehicle driving.

[0075] In this embodiment of the application, when the target to be detected is a road surface, the target features of the target to be detected are obtained by identifying time-domain information and / or frequency-domain information, and the target to be detected is detected based on the target features, including: monitoring the wheel speed signal of the tire using the frequency domain method; identifying the road surface features of the road surface by identifying the wheel speed signal; and matching the actual type of the preset uneven road surface according to the road surface features when the road surface is identified as a preset uneven road surface.

[0076] The preset uneven road surface can be a steep slope, a mountain road, a gravel road, etc., which can be determined according to the actual situation, and no specific limitation is made here.

[0077] It is understood that the embodiments of this application use the frequency domain method to monitor the wheel speed signal of the tires and detect the characteristics of the road surface. When the road surface is identified as an uneven road surface based on the road surface characteristics, the uneven road surface is classified according to the actual type of the uneven road surface, thereby improving vehicle safety and facilitating the control of vehicle comfort.

[0078] Specifically, assuming that road surface unevenness is a random vibration source, changes in road surface unevenness directly cause changes in longitudinal and vertical forces on the road surface. Wheel acceleration is selected as an indicator to measure road surface unevenness. Cross-correlation, spectral analysis and other methods are used to monitor wheel speed signals. By mining the hidden frequency domain information, the road surface unevenness can be detected.

[0079] It should be noted that the full vector control vehicle hardware and software architecture and hardware expansion acceleration method proposed in this application embodiment can perform secondary acquisition of wheel speed signals and accelerate the frequency domain analysis algorithm through the FFT coprocessor provided by the FPGA hardware accelerator, thereby realizing real-time processing of the frequency domain information of wheel speed signals. Furthermore, the accurate and real-time road surface roughness analysis achieved by leveraging the interface and computing power provided by the full vector control vehicle chassis controller can further improve the safety and comfort of the full vector control vehicle.

[0080] In this embodiment of the application, when the target to be detected is tire pressure, the target features of the target to be detected are obtained by identifying time-domain information and / or frequency-domain information, and the target to be detected is detected based on the target features, including: monitoring the wheel speed signal of the tire using the frequency domain method; identifying the wheel speed signal to obtain the actual frequency of the tire; and mapping the actual frequency to the tire pressure according to the preset relationship between the tire's natural frequency and the tire pressure, thereby realizing the monitoring of tire pressure.

[0081] The preset relationship could be that the tire's natural frequency increases with the increase of tire pressure. Since the change in tire pressure changes the tire stiffness and belt tension, it affects the tire's natural frequency.

[0082] It is understood that the embodiments of this application utilize the frequency domain method to monitor the wheel speed signal of the tire, identify the wheel speed signal to obtain the actual frequency of the tire, and map the actual frequency to the tire pressure according to the relationship between the tire's natural frequency and the tire pressure, thereby realizing the monitoring of tire pressure. By accurately monitoring the tire pressure, the service life of the tire can be determined, thus improving vehicle safety.

[0083] Specifically, wheel speed sensor signals are collected and preprocessed to establish a mathematical model of the tire's natural frequency. The natural frequency is then extracted by solving the mathematical model, and finally mapped to the tire pressure state.

[0084] It should be noted that the full vector control vehicle hardware and software architecture and hardware expansion acceleration method proposed in this application embodiment can collect wheel speed sensor signals and preprocess them, and accelerate the frequency domain analysis algorithm through the FFT coprocessor provided by the FPGA hardware accelerator, thereby realizing real-time processing of wheel speed signals. With the help of the interface and computing power provided by the full vector control vehicle chassis controller, accurate monitoring of tire pressure can be achieved, tire life can be determined, and the fuel economy and safety of the vehicle can be improved.

[0085] In this embodiment, when the target to be detected is a motor, the target features of the target to be detected are obtained by identifying time-domain information and / or frequency-domain information. The target to be detected is then detected based on the target features, including: identifying the time-domain information and frequency-domain information of the motor bearing to obtain the motor's rotational speed pulse signal and vibration signal; generating the rotational frequency curve of the motor bearing based on the rotational speed pulse signal and vibration signal, and segmenting the vibration signal at a preset stable speed from the rotational frequency curve; determining health indicators and fault frequencies based on the segmented vibration signals; matching the fault level of the motor bearing with the health indicators; and locating the fault location of the motor bearing using the fault frequency.

[0086] The vibration signal segment under the preset stable speed can be the vibration signal segment when the vehicle motor speed is in a stable state, and no specific limitation is made here.

[0087] It is understood that the embodiments of this application utilize time-frequency domain methods to detect and analyze the motor's speed and vibration signals, thereby determining the overall fault level and location of the motor bearings and improving the safety of the fully vector-controlled vehicle.

[0088] Specifically, the motor's speed pulse signal and vibration signal are collected simultaneously, and the rotational frequency curve is obtained. Based on the rotational frequency curve, the vibration signal at a stable speed is segmented to obtain health indicators and fault frequencies. Thus, the fault level and fault location of the motor bearing are determined based on the statistical results.

[0089] It should be noted that the full vector control vehicle hardware and software architecture and hardware expansion acceleration method proposed in this application embodiment can collect the motor speed pulse signal and vibration signal, and accelerate the frequency domain analysis algorithm through the FFT coprocessor provided by the FPGA hardware accelerator, thereby realizing real-time processing of speed pulse signal and vibration signal. With the help of the interface and computing power provided by the full vector control vehicle chassis controller, the overall fault degree and fault location of the motor bearing can be determined, thereby improving the safety of the full vector control vehicle.

[0090] The detection method for a safety system of a fully vector-controlled vehicle proposed in this application obtains the target to be detected from the fully vector-controlled vehicle and extracts the time-domain and / or frequency-domain information of the relevant hardware in the safety system. It identifies the relevant features of the target to be detected and examines the target based on these features. By utilizing the software and hardware architecture and hardware expansion acceleration method of the fully vector-controlled vehicle, it realizes the analysis and processing of the frequency domain part of the signal, significantly improves the computing power of the chassis controller of the fully vector-controlled vehicle, and improves the scalability of the vehicle while ensuring vehicle safety and real-time performance.

[0091] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the detection method of the full vector control vehicle safety system described above.

[0092] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0093] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0094] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0095] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (FPGAs), field-programmable gate arrays (FPGAs), etc.

[0096] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

Claims

1. A full-vector control safety system for automobiles, characterized in that, include: The electrical layer is used to achieve independent power supply between the chassis controller, autonomous driving controller, driving sensors, and actuators of a fully vector-controlled vehicle; The hardware layer includes the main control chip of the chassis controller, the extended acceleration chip of the autonomous driving controller, the external hardware corresponding to the driving sensors and the actuators. The driving sensors are connected to the autonomous driving controller through a preset high-speed serial port and / or Ethernet. One or more preset communication and security modules are mounted on the internal bus of the main control chip. The main control chip is externally connected to a power chip and a communication interface for the lower-level actuators and upper-level collaborative acceleration. The various actuators of the chassis and the chassis controller are connected to the vehicle network. The autonomous driving controller communicates with the vehicle network. The software layer is used to shield the differences between different hardware in the hardware layer, so that the upper application layer has universality and implements one or more preset drivers for the vehicle. The system layer is used to implement task allocation and scheduling, as well as the processing of external signal access and communication requests through the real-time operating system. It completes timing allocation through time and event-triggered interrupts to coordinate multiple tasks on the vehicle chassis controller in parallel. The algorithm layer includes an autonomous driving algorithm, a chassis controller safety algorithm, and a low-level actuator algorithm. The autonomous driving algorithm, the chassis controller safety algorithm, and / or the low-level actuator algorithm are used to realize the model calculation, state observation, and precise control of the actuators of the vehicle-road-driver system. The chassis controller is used to realize the functions of safety monitoring and hardware expansion acceleration, thereby realizing the full vector control of the vehicle chassis controller. The chassis controller is connected to various actuators of the chassis via a brake bus, a drive bus, an attitude bus, and an attitude redundancy bus. When the attitude bus fails, the attitude redundancy bus can control the vehicle's attitude. When the drive bus fails, the brake bus and the attitude bus assist the vehicle in safely stopping. When the brake bus fails, the drive bus controls the wheel hub motors to achieve emergency braking.

2. The system according to claim 1, characterized in that, The chassis controller collects signals from chassis sensors using IO, AD, and / or CC interfaces.

3. The system according to claim 1, characterized in that, The chassis controller collaborates with the high-performance computing unit for autonomous driving via a preset high-speed interface to achieve hardware expansion and acceleration of the chassis controller.

4. The system according to claim 3, characterized in that, The high-performance computing unit for autonomous driving includes an FFT coprocessor designed with an FPGA.

5. The system according to claim 1, characterized in that, The software layer includes the software layer for the autonomous driving controller design, the software layer for the other controller chips design, and complex drivers. The software layer for the autonomous driving controller design includes a base layer and a service layer. Each function in the layer consists of a functional cluster. The service layer includes the allocation of storage modules, reusable interfaces for various signals and communications, and the implementation of corresponding low-level safety function modules. In the software design of the other controller chips, a hardware abstraction layer defines the general hardware pin function allocation and basic communication interfaces.

6. The system according to claim 1, characterized in that, The chassis controller includes multiple monitoring modules from top to bottom, wherein, At the application layer, functional monitoring is achieved through the chassis controller. By verifying the signals and combining the status information of the driver, vehicle, and road surface obtained from the model calculation, the system's operating status is monitored at the algorithm level. At the system level, the operating system monitors the runtime and memory usage of each program segment. At the software layer, lockstep verification and memory verification of the lockstep core are performed by calling the values ​​of security-related registers; At the hardware level, the chip's own security unit is used to implement one or more functions such as operating temperature monitoring, lockstep verification, memory diagnostics, and software watchdog.

7. A fully vector-controlled vehicle, characterized in that, Including the full vector control vehicle safety system as described in any one of claims 1-6.

8. A detection method for a full-vector control vehicle safety system, characterized in that, The method is applied to the safety system of a fully vector-controlled vehicle as described in any one of claims 1-6, wherein the method includes the following steps: Obtain the target to be detected for the fully vector-controlled vehicle; Based on the target to be detected, obtain time-domain and / or frequency-domain information of one or more target-related hardware in the security system; The target features of the target to be detected are obtained by identifying the time-domain information and / or the frequency-domain information, and the target to be detected is detected based on the target features; By implementing autonomous driving algorithms, chassis controller safety algorithms, and / or underlying actuator algorithms, the system achieves model solving, state observation, and precise control of the actuators for the vehicle-road-driver system. It also implements safety monitoring and hardware expansion acceleration functions in the chassis controller, thereby realizing the full vector control function execution of the vehicle chassis controller. The chassis controller is connected to various actuators of the chassis via a brake bus, a drive bus, an attitude bus, and an attitude redundancy bus. When the attitude bus fails, the attitude redundancy bus can control the vehicle's attitude. When the drive bus fails, the brake bus and the attitude bus assist the vehicle in safely stopping. When the brake bus fails, the drive bus controls the wheel hub motors to achieve emergency braking.

9. The method according to claim 8, characterized in that, When the target to be detected is a road surface, the step of identifying the target features of the target by recognizing the time-domain information and / or the frequency-domain information, and detecting the target based on the target features, includes: The wheel speed signal of the tire is monitored using the frequency domain method; The road surface features are obtained by identifying the wheel speed signal; When the road surface is identified as a preset uneven road surface based on the road surface features, the actual type of the preset uneven road surface is matched based on the road surface features.

10. The method according to claim 8, characterized in that, When the target to be detected is tire pressure, the step of identifying the target features of the target by recognizing the time-domain information and / or the frequency-domain information, and detecting the target based on the target features, includes: The wheel speed signal of the tire is monitored using the frequency domain method; The actual frequency of the tire is obtained by identifying the wheel speed signal; Based on the preset relationship between the tire's natural frequency and tire pressure, the actual frequency is mapped to the tire pressure, thereby enabling the monitoring of the tire pressure.

11. The method according to claim 8, characterized in that, When the target to be detected is a motor, the step of identifying the target features of the target by recognizing the time-domain information and / or the frequency-domain information, and detecting the target based on the target features, includes: The time-domain and frequency-domain information of the motor bearings is identified to obtain the motor's speed pulse signal and vibration signal; The rotational frequency curve of the motor bearing is generated based on the rotational speed pulse signal and the vibration signal, and the vibration signal at a preset stable speed is segmented from the rotational frequency curve. Based on the vibration signal segments, health indicators and fault frequencies are determined. The health indicators are used to match the fault level of the motor bearing, and the fault frequency is used to locate the fault location of the motor bearing.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the detection method for the safety system of a fully vector-controlled vehicle as described in any one of claims 8-11.

Images