A method for controlling a vehicle and the vehicle

By analyzing the characteristic parameters of historical hardware protection events of electric drive chips, potential fault risks in electric drive systems can be identified, enabling proactive prevention and control, improving vehicle operation safety and fault identification accuracy, and reducing operation and maintenance costs.

CN123104400APending Publication Date: 2026-11-17GREAT WALL MOTOR CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611624139.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-20
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing technologies cannot proactively identify hardware faults caused by defects in the ground plane of printed circuit boards in electric drive systems, resulting in delayed fault identification and an inability to achieve proactive prevention and control.

Method used

By acquiring the operating condition characteristic parameters of multiple historical hardware protection events of the electric drive chip, clustering and quantile statistics are used to determine the concentrated distribution area and operating condition boundary range of the hardware protection events. The probability of failure risk is calculated by weighted fusion and maintenance suggestions are output to achieve proactive prevention and control.

Benefits of technology

Accurately identify the risk of hardware circuit failures in the electric drive system, reduce false alarms, improve vehicle operation safety and proactive prevention capabilities, and reduce maintenance costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123104400A_ABST
    Figure CN123104400A_ABST
Patent Text Reader

Abstract

This application provides a method for controlling a vehicle and a vehicle in the field of vehicle control technology. The method can identify the concentrated distribution area of ​​hardware protection events in a multi-dimensional operating condition space by identifying the region where the event occurs, reflecting the typical operating condition pattern of the event. Simultaneously, it can identify the risk interval of a single-dimensional operating condition characteristic parameter by the operating condition boundary range, avoiding the problem of missed fault detection due to preset fixed thresholds. The number of historical hardware protection events that meet the preset conditions corresponding to the concentrated distribution area and risk interval is determined. This allows for cross-validation from both spatial clustering and parameter boundary dimensions, effectively filtering out sporadic interference events and retaining events highly correlated with external hardware circuit faults, making the above count more realistically reflect fault trends, thereby accurately determining the risk probability. Therefore, the above scheme can improve the proactive fault prevention and control capability based on risk probability.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, and more specifically, to a method for controlling a vehicle and a vehicle within the field of vehicle control technology. Background Technology

[0002] Currently, vehicles rely on their electric drive systems to convert electrical energy into mechanical energy to achieve the overall vehicle drive control process. The operating status of the electric drive system directly affects the vehicle's power output, driving safety, and ride reliability. It is the core power system of the vehicle, and identifying and managing its operational risks is a key aspect of ensuring the reliable operation of the entire vehicle.

[0003] In electric drive systems, the external hardware circuitry of the electric drive chip is prone to hidden hardware failures such as printed circuit board (PCB) grounding defects. PCB grounding defects refer to design or manufacturing flaws in the grounding copper foil laid for the electric drive chip on the PCB. These defects directly affect the electric drive chip; for example, they can increase the chip's grounding impedance, cause abnormal signal return paths, and under specific operating conditions such as high current and high load, disrupt the chip's power supply and cause electromagnetic interference levels on the PCB to exceed permissible limits.

[0004] In related technologies, this type of hardware failure can only be addressed passively after a user complaint is triggered. There is no identification of the risk probability of hardware failure in external hardware circuits to accurately predict this type of hardware failure and thus improve the proactive prevention and control capabilities for this type of hardware failure. Summary of the Invention

[0005] This application provides a method for controlling a vehicle and a vehicle, which can accurately determine the risk probability of hardware failure in external hardware circuits and improve the proactive prevention and control capability against failures based on the risk probability.

[0006] In a first aspect, a method for controlling a vehicle is provided. The vehicle includes an electric drive system, which includes an electric drive chip. The method includes: acquiring operating condition characteristic parameters corresponding to multiple historical hardware protection events of the electric drive chip, the operating condition characteristic parameters including sub-characteristic parameters of multiple dimensions; determining, based on the multiple operating condition characteristic parameters, the number of historical hardware protection events that meet preset conditions among the multiple historical hardware protection events, the preset conditions indicating that the corresponding historical hardware protection event is in an event region and at least one corresponding operating condition characteristic parameter is in a corresponding dimension's operating condition boundary range, the event region including hardware protection events where the difference between the operating condition characteristic parameters is less than a preset difference; determining the risk probability of a failure in the external hardware circuit of the electric drive chip based on the number of historical hardware protection events, and controlling the vehicle based on the risk probability.

[0007] In the above technical solution, operating condition characteristic parameters of multiple historical hardware protection events of the electric drive chip are obtained. Based on multiple operating condition characteristic parameters, the number of historical hardware protection events that meet preset conditions is determined. Since the preset conditions indicate that the corresponding historical hardware protection event is located in the event's region and at least one corresponding operating condition characteristic parameter is within the corresponding dimension's operating condition boundary range, and the event's region includes hardware protection events where the difference between operating condition characteristic parameters is less than a preset difference, this can identify the concentrated distribution area of ​​hardware protection events in a multi-dimensional operating condition space, reflecting the typical operating condition pattern of the event occurrence. Furthermore, this can also identify the risk range of a single-dimensional operating condition characteristic parameter based on data distribution characteristics, avoiding the problem of misjudgment of faults caused by preset fixed thresholds. That is, after determining the number of historical hardware protection events, the above solution can perform cross-validation from the two dimensions of spatial clustering and parameter boundaries, effectively filtering out occasional interference events and retaining events highly correlated with external hardware circuit faults, making the above number more realistically reflect the fault trend, thereby accurately determining the risk probability. Therefore, the above solution can improve the proactive fault prevention capability based on risk probability, enhancing the vehicle's operational safety.

[0008] In conjunction with the first aspect, in some possible implementations, before determining the number of historical hardware protection events that meet the preset conditions among the multiple historical hardware protection events based on multiple operating condition feature parameters, the method further includes: clustering multiple operating condition feature parameters to obtain the region where the event is located; and performing quantile statistics on multiple sub-feature parameters of each single dimension to obtain the operating condition boundary range of the corresponding dimension.

[0009] The aforementioned technical solution can automatically generate multi-dimensional operating condition clustering regions (i.e., the regions where events occur) and corresponding operating condition boundary ranges for each dimension based on sample data of historical hardware protection events. This eliminates the need to rely entirely on calibrated fixed region and boundary thresholds, adapting to the differences in operating condition samples from different electric drive chips and individual vehicles. Clustering processing can uncover the clustering distribution patterns of historical hardware protection events in the multi-dimensional operating condition space, while quantile statistics can objectively output the risk boundaries of sub-feature parameters for each dimension. This provides a data-driven judgment benchmark for subsequent screening of historical hardware protection events that meet preset conditions, reducing subjective biases caused by parameter settings and ensuring the objectivity and reliability of the input basis for subsequent statistical analysis of the aforementioned frequency and determination of the risk probability of external hardware circuit failures. This further improves the adaptability and accuracy of risk identification, supporting vehicles in achieving more reliable proactive fault prevention.

[0010] Clustering multiple operating condition characteristic parameters yields the region where the event occurs. This identifies the concentrated distribution area of ​​hardware protection events in a multi-dimensional operating condition space, reflecting the typical operating condition patterns in which the event occurs. Quantile statistics are performed on the sub-feature parameters of each single dimension to obtain the corresponding operating condition boundary range. This allows for the identification of the risk interval of a single-dimensional operating condition characteristic parameter based on data distribution characteristics, avoiding misjudgments of faults caused by preset fixed thresholds.

[0011] Combining the first aspect and the above implementation methods, in some possible implementation methods, the sub-feature parameters of the multiple dimensions include the current peak value of the motor controller. Quantile statistics are performed on the multiple sub-feature parameters of each single dimension to obtain the operating condition boundary range of the corresponding dimension. This includes: sorting the multiple current peak values ​​to obtain the sorted multiple current peak values; determining the low quantile value and high quantile value of the sorted multiple current peak values ​​based on the quantile probability value corresponding to the current peak value; determining the low quantile value as the lower limit of the operating condition boundary range corresponding to the current peak value, and determining the high quantile value as the upper limit of the operating condition boundary range corresponding to the current peak value.

[0012] In the above technical solution, quantile statistics are performed on the current peak values ​​of the motor controller, and multiple current peak values ​​are sorted. This enables the orderly organization of multiple current peak values, providing a data foundation for quantile value extraction. Subsequently, based on the quantile probability values ​​corresponding to the current peak values, the lower and higher quantile values ​​of the sorted current peak values ​​are determined. This eliminates the need to forcibly specify a fixed current threshold, and can adaptively generate the corresponding operating condition boundary range based on the current peak value dimension and the distribution of current peak values, filtering out interference from a small number of discrete current samples. The above solution can adapt to sporadic faults triggered only in specific current ranges due to external hardware circuit failures (i.e., latent hardware faults), obtaining objective and accurate current operating condition boundaries, and reducing the risk of fault misjudgment caused by manually setting thresholds.

[0013] In combination with the first aspect and the above implementation methods, in some possible implementation methods, the risk probability of the external hardware circuit of the electric drive chip failing is determined based on the number of historical hardware protection events, including: comparing the number of historical hardware protection events with multiple preset number ranges, determining a target preset number range that matches the number of historical hardware protection events from the multiple preset number ranges, where each preset number range corresponds to a sample risk probability; and determining the sample risk probability corresponding to the target preset number range as the risk probability.

[0014] In the above technical solution, the number of historical hardware protection events is compared with multiple preset count ranges, and the target preset count range is determined by interval matching. This eliminates the need for complex floating-point calculations. Subsequently, the counts are directly converted into the probability of external hardware circuit failure through interval mapping. This simplifies the calculation logic of the risk probability and reduces computational overhead. In other words, the above solution uses a hierarchical interval mapping method, which can mitigate the probability jitter caused by fluctuations in single counts and determine accurate risk probabilities. Furthermore, the logic of this method is simple and easy to implement, enabling rapid output of risk probabilities. Therefore, the above solution can balance the calculation efficiency and accuracy of risk probabilities.

[0015] In conjunction with the first aspect and the above implementation methods, in some possible implementation methods, the sub-feature parameters of the multiple dimensions also include the recovery time of historical hardware protection events. The method further includes: determining the maximum recovery time among the multiple operating condition feature parameters; determining a first weight based on the number of triggers of historical hardware protection events, and determining a second weight based on the maximum recovery time, wherein the first weight is used to indicate the contribution of the number of triggers in determining the risk probability, and the second weight is used to indicate the contribution of the maximum recovery time in determining the risk probability; and determining the risk probability of the external hardware circuit of the electric drive chip failing based on the number of historical hardware protection events, including: weighting and fusing the number of triggers, the maximum recovery time, and the number of historical hardware protection events based on the first weight, the second weight, and the third weight to obtain the risk probability, wherein the third weight is used to indicate the contribution of the number of triggers in determining the risk probability.

[0016] In the above technical solution, the recovery time of historical hardware protection events is incorporated into the sub-feature parameters to obtain the maximum recovery time among the operating condition feature parameters. This dimension indicator can reflect the severity of hardware protection events. Subsequently, based on the number of triggers and the maximum recovery time, a first weight and a second weight are determined respectively. Based on the first weight, the second weight, and the third weight, the number of triggers, the maximum recovery time, and the number of events are weighted and fused to obtain the risk probability. This overcomes the limitation of relying solely on a single event count to assess risk probability, taking into account the number of hardware protection events triggered, their severity, and the statistical count of valid samples (the number of valid hardware protection events, i.e., the statistical count of valid events). Furthermore, the weights allow for flexible adjustment of the influence ratio of each dimension indicator, adapting to the performance characteristics of different faults and suppressing assessment bias caused by fluctuations in a single indicator. Therefore, the above solution can improve the accuracy and comprehensiveness of assessing the risk probability of external hardware circuit failures, providing a more comprehensive basis for subsequent vehicle management.

[0017] Combining the first aspect and the above implementation methods, in some possible implementation methods, the risk probability is obtained by weighted fusion of the number of triggers, the maximum recovery time, and the number of historical hardware protection events based on the first weight, the second weight, and the third weight. This includes: determining the ratio between the number of triggers and the first preset number as a first amplitude; determining the ratio between the maximum recovery time and the first preset time as a second amplitude; determining a third amplitude based on whether the number of historical hardware protection events is greater than the second preset number, wherein the second preset number is less than the first preset number; and weighted fusion of the first amplitude, the second amplitude, and the third amplitude based on the first weight, the second weight, and the third weight to obtain the risk probability.

[0018] In the above technical solution, the ratio between the number of triggers and a first preset number of triggers is determined as the first amplitude, and the ratio between the maximum recovery time and a first preset time is determined as the second amplitude. This eliminates the dimensional differences between different physical quantities through normalization, making indicators of different dimensions comparable. A third amplitude is determined based on whether the number of triggers exceeds the second preset number of triggers. This standardizes the valid events using the second preset number of triggers. Subsequently, the first, second, and third amplitudes are weighted and fused based on the first, second, and third weights to obtain the risk probability. This avoids the problem of weighting result distortion caused by differences in the numerical magnitudes of the original parameters, taking into account the number of triggers, severity, and the statistical number of valid events. The contribution of each indicator is flexibly adjusted by the weights, which reduces the evaluation bias caused by scale differences between different parameters and improves the stability and rationality of determining the risk probability.

[0019] In conjunction with the first aspect and the above implementation methods, in some possible implementation methods, a third amplitude is determined based on whether the number of historical hardware protection events is greater than a second preset number, including: if the number of historical hardware protection events is greater than the second preset number, the third amplitude is determined as a first preset amplitude, which is used to indicate that the probability of an external hardware circuit failure is greater than a first probability; if the number of historical hardware protection events is less than or equal to the second preset number, the third amplitude is determined as a second preset amplitude, which is less than the first preset amplitude.

[0020] In the above technical solution, a second preset number of times is used as the dividing line to determine the magnitude of the statistical count of valid events. When the statistical count is greater than the second preset number of times, the third amplitude is assigned the first preset amplitude. This is used to characterize that the external hardware circuit has a high probability of failure in such cases. When the statistical count is less than or equal to the second preset number of times, it is assigned the second preset amplitude, which is smaller in value. This enables discretized amplitude mapping of the statistical count. The above solution does not require complex continuous calculations on the statistical count, thus reducing computational overhead. In addition, directly determining the third amplitude by the preset number range in which the statistical count falls also allows for rapid acquisition of the third amplitude, improving the speed of determining the risk probability.

[0021] Combining the first aspect and the above implementation methods, in some possible implementation methods, the vehicle is controlled based on the risk probability, including: determining whether the risk probability is greater than a second probability, where the second probability is greater than the first probability; and if the risk probability is greater than the second probability, outputting a maintenance list to suggest that the target object repair the external hardware circuit based on the maintenance list.

[0022] In the above technical solution, the risk probability is compared with a second probability, which has a larger value, to distinguish the level of fault risk. When the risk probability is greater than the second probability, a repair list is output to suggest that the target device should repair the external hardware circuitry. The above judgment logic is based on quantified risk probability as the triggering basis, rather than solely on fault codes to trigger repair prompts. This allows for the identification of potential hidden dangers in the external hardware circuitry of the electric drive chip before a fault occurs. Setting a second probability greater than the first probability as the trigger threshold can filter out low-risk noise samples and avoid frequent false repair suggestions. In addition, outputting a repair list allows maintenance personnel to quickly locate the hardware points to be checked, reducing the possibility of fault deterioration leading to abnormal vehicle operation, improving the vehicle's proactive prevention capabilities, and also reducing the maintenance costs associated with reactive handling after a fault occurs.

[0023] In conjunction with the first aspect and the above implementation methods, in some possible implementation methods, the method further includes: after repairing the external hardware circuit, obtaining the first number of times the hardware protection event is triggered within a unit mileage under preset operating conditions, wherein the preset operating conditions are conditions that are prone to triggering hardware protection events; determining a fourth amplitude based on the first and second trigger counts, wherein the second trigger count is the trigger count corresponding to the time before repair, and the fourth amplitude is used to indicate the degree of reduction in the number of triggers; and determining the repair result based on the fourth amplitude, wherein the repair result is the result of repairing the external hardware circuit.

[0024] In the above technical solution, after the external hardware circuit has been inspected, a preset operating condition that is prone to triggering hardware protection events is selected, and the first trigger count per unit mileage is obtained. Using this type of operating condition can fully expose potential fault causes and avoid obtaining false inspection results under mild operating conditions. Based on the first trigger count after inspection and the second trigger count before inspection, a fourth amplitude is determined, which allows for a quantitative expression of the repair effect. This solution eliminates the process of judging repair effectiveness based on subjective experience and unifies the evaluation criteria for repair effectiveness. This can identify hidden situations where faults are not completely eliminated, providing objective evidence for subsequent re-inspection and repair upgrades, and improving the closed-loop process for fault risk management.

[0025] In combination with the first aspect and the above implementation methods, in some possible implementation methods, the maintenance result is determined based on the fourth amplitude, including: if the first trigger count is less than the second trigger count and the fourth amplitude is greater than or equal to the third preset amplitude, the maintenance result is determined to be valid; if the fourth amplitude is greater than the fourth preset amplitude and less than the third preset amplitude, the maintenance result is determined to be partially valid, and a re-inspection list is output, which is used to indicate the execution of a preset verification process; if the fourth amplitude is less than or equal to the fourth preset amplitude, the maintenance result is determined to be invalid, and a reminder message is output, which is used to remind the vehicle to be sent to a higher-level repair shop for in-depth maintenance.

[0026] In the above technical solution, when the first trigger count is less than the second trigger count and the fourth amplitude is greater than or equal to the third preset amplitude, the repair is deemed effective. This takes into account both the downward trend of the trigger count and the degree of improvement. When the fourth amplitude is between the fourth preset amplitude and the third preset amplitude, the repair is deemed partially effective, and a re-inspection list is output. This guides the target to execute the preset verification process. When the fourth amplitude is less than or equal to the fourth preset amplitude, the repair is deemed ineffective, and a reminder message is output. This reminds the target to send the vehicle to a higher-level repair facility for in-depth repair. In other words, the above solution divides the repair effectiveness into three categories of results through graded thresholds. This avoids the problem of coarse judgment caused by a single threshold and allows for differentiated handling strategies to be matched for different degrees of improvement. This can recognize repair results that are completely repaired, as well as cases of insufficient improvement and complete failure, forming a graded closed-loop handling logic, improving the precision of repair effect evaluation, and perfecting the whole-process risk management of the external hardware circuit of the electric drive chip.

[0027] Secondly, a device for controlling a vehicle is provided. The vehicle includes an electric drive system, which includes an electric drive chip. The device includes: an acquisition module, configured to acquire operating condition characteristic parameters corresponding to multiple historical hardware protection events of the electric drive chip, the operating condition characteristic parameters including sub-characteristic parameters of multiple dimensions; a determination module, configured to: determine, based on the multiple operating condition characteristic parameters, the number of historical hardware protection events that meet preset conditions among the multiple historical hardware protection events, the preset conditions indicating that the corresponding historical hardware protection event is in the event region and at least one corresponding operating condition characteristic parameter is in the operating condition boundary range of the corresponding dimension, the event region including hardware protection events where the difference between the operating condition characteristic parameters is less than a preset difference; and a control module, configured to determine the risk probability of a failure in the external hardware circuit of the electric drive chip based on the number of historical hardware protection events, and control the vehicle based on the risk probability.

[0028] In conjunction with the second aspect, in some possible implementations, before determining the number of historical hardware protection events that meet the preset conditions among the multiple historical hardware protection events based on multiple operating condition feature parameters, the determining module is further used to cluster the multiple operating condition feature parameters to obtain the region where the event is located; and the determining module is further used to perform quantile statistics on multiple sub-feature parameters of each single dimension to obtain the operating condition boundary range of the corresponding dimension.

[0029] Combining the second aspect and the above implementation methods, in some possible implementation methods, the sub-feature parameters of the multiple dimensions include the current peak value of the motor controller. The determining module is specifically used for: sorting the multiple current peak values ​​to obtain sorted multiple current peak values; determining the low quantile value and high quantile value of the sorted multiple current peak values ​​based on the quantile probability value corresponding to the current peak value; determining the low quantile value as the lower limit of the operating condition boundary range corresponding to the current peak value, and determining the high quantile value as the upper limit of the operating condition boundary range corresponding to the current peak value.

[0030] In conjunction with the second aspect and the above implementation methods, in some possible implementation methods, the determining module is further specifically used to: compare the number of the historical hardware protection events with multiple preset number ranges, determine a target preset number range that matches the number of the historical hardware protection events from the multiple preset number ranges, where each preset number range corresponds to a sample risk probability; and determine the sample risk probability corresponding to the target preset number range as the risk probability.

[0031] In conjunction with the second aspect and the above implementation methods, in some possible implementation methods, the sub-feature parameters of the multiple dimensions also include the recovery time of historical hardware protection events. The determining module is further configured to: determine the maximum recovery time among the multiple operating condition feature parameters; determine a first weight based on the number of triggers of historical hardware protection events, and determine a second weight based on the maximum recovery time. The first weight is used to indicate the contribution of the number of triggers in determining the risk probability, and the second weight is used to indicate the contribution of the maximum recovery time in determining the risk probability. Specifically, the determining module is further configured to perform weighted fusion of the number of triggers, the maximum recovery time, and the number of historical hardware protection events based on the first weight, the second weight, and the third weight to obtain the risk probability. The third weight is used to indicate the contribution of the number of events in determining the risk probability.

[0032] In conjunction with the second aspect and the above implementation methods, in some possible implementation methods, the determining module is specifically used to: determine the ratio between the number of triggers and the first preset number as a first amplitude; determine the ratio between the maximum recovery time and the first preset time as a second amplitude; determine a third amplitude based on whether the number of historical hardware protection events is greater than the second preset number, wherein the second preset number is less than the first preset number; and perform weighted fusion of the first amplitude, the second amplitude, and the third amplitude based on the first weight, the second weight, and the third weight to obtain the risk probability.

[0033] In conjunction with the second aspect and the above implementation methods, in some possible implementation methods, the determining module is further configured to: determine the third amplitude as the first preset amplitude when the number of historical hardware protection events is greater than the second preset number, wherein the first preset amplitude is used to indicate that the probability of an external hardware circuit failure is greater than the first probability; and determine the third amplitude as the second preset amplitude when the number of historical hardware protection events is less than or equal to the second preset number, wherein the second preset amplitude is less than the first preset amplitude.

[0034] In combination with the second aspect and the above implementation methods, in some possible implementation methods, the determining module is specifically used to determine whether the risk probability is greater than the second probability, and the second probability is greater than the first probability; the control module is specifically used to output a maintenance list when the risk probability is greater than the second probability, so as to suggest that the target object repair the external hardware circuit based on the maintenance list.

[0035] In conjunction with the second aspect and the above implementation methods, in some possible implementation methods, the acquisition module is further configured to acquire, after repairing the external hardware circuit, the first trigger count of the hardware protection event within a unit mileage under preset operating conditions, wherein the preset operating conditions are operating conditions that are prone to triggering hardware protection events; the determination module is further configured to: determine a fourth amplitude based on the first trigger count and the second trigger count, wherein the second trigger count is the trigger count corresponding to the time before repair, and the fourth amplitude is used to indicate the degree of reduction in the trigger count; and determine the repair result based on the fourth amplitude, wherein the repair result is the result of repairing the external hardware circuit.

[0036] In conjunction with the second aspect and the above implementation methods, in some possible implementation methods, the determining module is further configured to: determine the inspection result as valid when the first trigger count is less than the second trigger count and the fourth amplitude is greater than or equal to the third preset amplitude; determine the inspection result as partially valid when the fourth amplitude is greater than the fourth preset amplitude and less than the third preset amplitude, and output a re-inspection list, which is used to instruct the execution of a preset verification process; determine the inspection result as invalid when the fourth amplitude is less than or equal to the fourth preset amplitude, and output a reminder message, which is used to remind the vehicle to be sent to a higher-level repair facility for in-depth inspection.

[0037] Thirdly, a vehicle is provided, including a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, causing the vehicle to perform the methods described in the first aspect or any possible implementation thereof. Attached Figure Description

[0038] Figure 1 This is an internal structural diagram of an electric drive system provided in an embodiment of this application; Figure 2 This is a schematic flowchart illustrating a method for controlling a vehicle according to an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a vehicle control device provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a controller provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of a vehicle provided in an embodiment of this application. Detailed Implementation

[0039] The technical solutions in this application will be clearly and thoroughly described below with reference to the accompanying drawings. In the description of the embodiments of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. "And / or" in the text is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, in the description of the embodiments of this application, "multiple" refers to two or more than two.

[0040] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0041] For example, such as Figure 1 As shown, the electric drive system includes a motor controller, whose hardware carrier is a printed circuit board (PCB), serving as the "physical substrate" of the motor controller. The PCB also houses various electronic components required by the motor controller, including electric drive chips and communication modules. These electronic components transmit data through conductive lines (such as copper foil) on the PCB. The design quality of the PCB directly affects the operational stability of core components such as the electric drive chip within the motor controller. The electric drive chip (such as an Insulated Gate Bipolar Transistor (IGBT) driver chip) receives logic commands from the motor controller and converts them into high-power electrical signals (a type of drive signal) to drive the motor windings, directly controlling the energizing state of the motor.

[0042] The printed circuit board (PCB) of the electric drive chip within the motor controller is prone to hidden hardware failures (specifically, PCB grounding defects). PCB grounding defects refer to design or manufacturing flaws in the grounding copper foil for the electric drive chip, such as ground plane breaks or ground islands. Ground plane breaks occur when a complete grounding copper foil is cut into multiple pieces by traces or vias. Under high current flow, the return current is forced to take a longer route, increasing the grounding impedance of the electric drive chip. Ground islands refer to a small piece of grounding copper foil completely isolated from the surrounding area, connected only to the main ground plane (a large copper-clad grounding copper foil area on the PCB) through a very thin line. The potential of this "grounding copper foil" is unstable, and power noise can easily accumulate, interfering with the signal pins of surrounding electric drive chips. These PCB grounding defects directly affect the electric drive chip; for example, increased grounding impedance, abnormal signal return paths, and under specific operating conditions such as high current and high load, disturbances to the power supply of the electric drive chip and electromagnetic interference levels on the PCB exceeding permissible limits.

[0043] In related technologies, this type of hardware failure can only be addressed passively after a user complaint is triggered. There is no identification of the risk probability of hardware failure in external hardware circuits to accurately predict this type of hardware failure and thus improve the proactive prevention and control capabilities for this type of hardware failure.

[0044] To address the aforementioned issues, this application proposes a method for controlling a vehicle to accurately determine the probability of hardware failures in external hardware circuits, and based on this probability, improve the proactive fault prevention capabilities. Specific implementation steps can be found in [reference needed]. Figure 2 .

[0045] Figure 2 This is a schematic flowchart of a method for controlling a vehicle provided in an embodiment of this application.

[0046] It should be understood that the method for controlling a vehicle provided in this application embodiment can be applied to, for example... Figure 1 The vehicle shown includes an electric drive system, which includes an electric drive chip.

[0047] For example, such as Figure 2 As shown, the method 200 includes the following steps 201 to 203.

[0048] Step 201: Obtain the operating condition characteristic parameters corresponding to multiple historical hardware protection events of the electric drive chip. The operating condition characteristic parameters include sub-characteristic parameters of multiple dimensions.

[0049] It should be understood that in step 201 above, the hardware protection event of the electric drive chip is the hardware protection mechanism of the electric drive chip. When the electric drive chip detects abnormal electrical environments such as excessive disturbances in the power supply or excessive electromagnetic interference, it automatically executes protection actions (such as blocking the drive signal output) to prevent the electric drive chip itself from being permanently damaged due to the harsh electrical environment; and to prevent incorrect drive signal output due to malfunction or damage of the electric drive chip, thereby avoiding safety accidents such as runaway drive motor and interruption of vehicle power.

[0050] Furthermore, each hardware protection event is recorded, forming corresponding operating condition characteristic parameters, providing data for subsequent analysis of whether there are hidden hardware faults in the external hardware circuit (printed circuit board). Each hardware protection event corresponds to one operating condition characteristic parameter (specifically, a set of characteristic parameters). Each operating condition characteristic parameter includes multiple sub-characteristic parameters, referring to the operating condition characteristic parameters related to the electric drive chip in multiple dimensions when the hardware protection event is triggered. Optionally, the multiple sub-characteristic parameters include the peak current of the motor controller (peak bus current), bus voltage, actual speed of the drive motor, actual torque, actual temperature (first temperature), actual temperature of the motor controller (second temperature), and the recovery time of the hardware protection event (obtained by subtracting the second time (trigger time) from the first time (i.e., the recovery time)). The peak current, bus voltage, actual speed, actual torque, first temperature, and second temperature included in the multiple sub-characteristic parameters of the same operating condition characteristic parameter are all collected at the same time, which is the second time. The recovery time of a hardware protection event refers to the time interval between when the electric drive chip enters the protection state after the hardware protection event is triggered and when it exits the protection state (i.e., the duration of the hardware protection mechanism). After a hardware protection event is triggered, the electric drive system is in an unavailable state. This first moment is the moment when the electric drive system re-enters the available state.

[0051] It is important to note that, to avoid inconsistencies in the determination of recovery time for different vehicles due to inconsistent methods of obtaining the recovery time, the following limitations need to be imposed on the method of obtaining the recovery time. Considering that the trigger time is obtained directly from the edge of the electric drive chip's status register or fault pin by the motor controller, it is a chip-level event. The recovery time, by definition, is a system-level event, and the two have different source levels. Method 200 specifies that the recovery time is the moment when the electric drive system re-enters a usable state. This moment is specifically the moment when the motor controller outputs the high-voltage ready signal or the output torque recovery signal, thus eliminating the influence of the time difference between the instantaneous exit of the electric drive chip's protection state and the release of the high-voltage path protection lockout (i.e., the motor controller outputs the high-voltage ready signal or the output torque recovery signal) on the recovery time. Furthermore, after triggering the hardware protection event, the electric drive chip enters a protection state. This protection state is latched, meaning that although the fault condition has been eliminated, the status bit of the electric drive chip in the protection state remains set. The motor controller needs to actively send a status clear command to de-set the status bit. Therefore, the moment when the electric drive system returns to an available state is the recovery moment, in order to avoid including the delay caused by the execution of the control state clearing command in the recovery time, thus preventing the recovery time from being mixed with software strategy factors unrelated to hardware failures.

[0052] It's also important to clarify that the existence of a hardware protection event for the electric drive chip does not necessarily mean that the external hardware circuitry of the electric drive chip has failed. The relationship between the two is as follows: 1. The triggering sources of hardware protection events are diverse. In addition to faults such as PCB ground plane defects, non-hardware fault factors such as transient overcurrent under extreme operating conditions and sudden changes in the external electromagnetic environment can also trigger hardware protection events for the electric drive chip. Therefore, a single hardware protection event is merely an "abnormal signal" and cannot be directly equated to a failure in the external hardware circuitry. 2. The relationship between external hardware circuit failure and hardware protection events is a "conditional causal relationship." PCB ground plane defects, as a type of latent hardware fault, may not exhibit obvious abnormalities under normal operating conditions. Only under specific operating conditions such as high current and high load will this PCB ground plane defect trigger the hardware protection event of the electric drive chip through paths such as increased grounding impedance and abnormal signal return. In other words, external hardware circuit failure is one of the potential causes of hardware protection events, but the triggering conditions for hardware protection events still need to be activated by specific operating conditions.

[0053] Step 202: Based on multiple operating condition characteristic parameters, determine the number of historical hardware protection events that meet preset conditions among the multiple historical hardware protection events. The preset conditions are used to indicate that the corresponding historical hardware protection event is in the event area and at least one corresponding operating condition characteristic parameter is in the operating condition boundary range of the corresponding dimension. The event area includes hardware protection events where the difference between the operating condition characteristic parameters is less than a preset difference.

[0054] It should be understood that in step 202 above, the event location area can be a range of operating conditions formed by clustering hardware protection events with similar operating condition characteristics after clustering multiple operating condition feature parameters. In simple terms, imagine each hardware protection event as a "point" in a multi-dimensional space, with each dimension corresponding to a sub-feature parameter. Clustering groups "closely related" points together; this "group" is the event location area. Only when the differences between the sub-feature parameters of two hardware protection events in each dimension are sufficiently small are they considered to belong to the same area. The significance of the event location area is that if a large number of hardware protection events are concentrated in a specific operating condition area (such as high current peaks, high actual speed of the drive motor, and high actual temperature), it indicates that this operating condition area is a high-incidence area for hardware protection events, which may be closely related to hidden hardware faults such as printed circuit board grounding defects.

[0055] It should also be understood that in step 202 above, the operating condition boundary range can be the corresponding numerical interval obtained after performing quantile statistics on multiple sub-characteristic parameters of any single dimension. In simpler terms, assuming the sub-characteristic parameter is the current peak value, arranging the current peak values ​​at all hardware protection events from smallest to largest, and taking certain quantile probability values ​​(such as 90% and 99%), a numerical interval is obtained, such as... .Should The operating condition boundary range for the current peak value dimension represents the range within which the value of a hardware protection event is most likely to fall. The significance of this operating condition boundary range lies in the fact that, through quantile statistics, the dangerous operating condition intervals for each dimension can be objectively determined from the data, rather than relying on subjective judgment based on experience. If the operating condition boundary range for a certain dimension is very narrow and concentrated in the extreme value region, it indicates that this dimension has a strong correlation with the triggering of hardware protection events. Furthermore, the operating condition boundary ranges corresponding to sub-feature parameters of different dimensions are different.

[0056] It's also important to note that the choice of different quantile combinations depends on the physical direction and fault mechanism of the sub-feature parameters triggering hardware protection events in the corresponding dimension. In other words, different dimensions require different quantile combinations. The 90% and 99% quantiles are used to define the high-value boundary, meaning the high-value boundary applies to dimensions where larger values ​​are more likely to trigger hardware protection events, such as peak current. Specifically, the 90% quantile indicates that 90% of hardware protection events in this dimension have values ​​below this value, representing the common upper limit for that dimension; the 99% quantile indicates that 99% of hardware protection events in this dimension have values ​​below this value, representing the extreme upper limit, or range. This defines the high-value side operating condition boundary range where hardware protection events tend to occur most frequently in this dimension. 1% and 10% are used to define the low-value side boundary; that is, the low-value side boundary applies to dimensions where smaller values ​​are more likely to trigger hardware protection events, such as the bus voltage dimension. Specifically, the 1% quantile indicates that only 1% of hardware protection events in this dimension have values ​​below this value, i.e., the extreme lower limit; the 10% quantile indicates that 10% of hardware protection events in this dimension have values ​​below this value, i.e., the common lower limit, or range. This allows for the definition of the low-value side operating condition boundary range where hardware protection events tend to occur concentratedly on this dimension. The essence of the aforementioned quantile statistics lies in arranging the values ​​(multiple sub-feature parameters) of multiple hardware protection events on a certain dimension from smallest to largest. By selecting specific quantiles, the numerical range of the "typical fault condition" on this dimension is determined. In other words, the operating condition boundary range refers to the numerical range corresponding to this typical fault condition, characterizing the range within which hardware protection events typically concentrate on occurring on this dimension. Furthermore, the aforementioned... This indicates the 1st percentile (the value that is located at approximately 1%).

[0057] For example, 20 historical hardware protection events of a certain electric drive chip are given. Each event corresponds to a current peak value in amperes (A), specifically including 135, 120, 140, 148, 152, 160, 165, 170, 175, 185, 180, 190, 195, 200, 220, 210, 230, 240, 255 and 280. The 20 current peak values ​​are sorted to obtain the following 20 sorted current peak values: 120, 135, 140, 148, 152, 160, 165, 170, 175, 180, 185, 190, 195, 200, 210, 220, 230, 240, 255, and 280. Based on the product of the quantile probability value and the number of historical hardware protection events, the quantile position (i.e., the position number) corresponding to the quantile probability value is determined, which is 20. 90% = 18 (first position number) 240A at position 18, which is 20 99% = 19.8 (second position number) The current peak at position 19.8 is given. Since position 19.8 does not exist, linear interpolation of its two corresponding current peaks (i.e., the current peak at position 19 and the current peak at position 20) yields 275A. Therefore, .

[0058] It should be understood that in step 202 above, when determining the number (statistical count) of historical hardware protection events that meet the preset conditions, the corresponding historical hardware protection events need to meet two conditions. Condition one: The corresponding historical hardware protection event is located in the event's region. This ensures that the corresponding historical hardware protection event exhibits clustering with other historical hardware protection events in the multi-dimensional joint operating condition space. That is, multiple operating condition characteristic parameters of the corresponding historical hardware protection event combine together with a group of similar historical hardware protection events to form a high-incidence operating condition cluster. This can exclude isolated and sporadic historical hardware protection events. Condition two: At least one operating condition characteristic parameter (sub-characteristic parameter) in the corresponding historical hardware protection event is within the operating condition boundary range of the corresponding dimension. This ensures that the value of the corresponding historical hardware protection event, in at least one key dimension, is within the numerical range corresponding to the typical fault operating condition of the corresponding dimension. This can exclude hardware protection events that, although clustered, have sub-characteristic parameters within the normal range in each dimension. In other words, simultaneously meeting both conditions is to use a dual screening mechanism of multi-dimensional joint clustering and single-dimensional extremism to lock in hardware protection events strongly associated with latent hardware-type faults.

[0059] Step 203: Based on the number of historical hardware protection events, determine the probability of failure of the external hardware circuit of the electric drive chip, and control the vehicle based on the probability of failure.

[0060] It should be understood that in step 203 above, the risk probability is positively correlated with the number of historical hardware protection events. The higher the number, the more frequently the external hardware circuit triggers hardware protection events under specific operating conditions, and the higher the risk probability; the lower the number, the lower the risk probability. Furthermore, from the perspective of cumulative probability, even if the probability of a single hardware protection event being caused by a latent hardware fault is very small, when hardware protection events that meet specific operating conditions occur repeatedly, these hardware protection events all point to the same root cause (such as a printed circuit board grounding defect). The more times the event occurs, the more likely it is to rule out the possibility of sporadic occurrences and the more certain it is that the fault is caused by an external hardware circuit.

[0061] The embodiments corresponding to step 202 above will be described in detail below.

[0062] In one possible implementation, before determining the number of historical hardware protection events that meet the preset conditions among the multiple historical hardware protection events based on multiple operating condition feature parameters in step 202, the method 200 further includes: clustering multiple operating condition feature parameters to obtain the region where the event is located; and performing quantile statistics on multiple sub-feature parameters of each single dimension to obtain the operating condition boundary range of the corresponding dimension.

[0063] The aforementioned technical solution can automatically generate multi-dimensional operating condition clustering regions (i.e., the regions where events occur) and corresponding operating condition boundary ranges for each dimension based on sample data of historical hardware protection events. This eliminates the need to rely entirely on calibrated fixed region and boundary thresholds, adapting to the differences in operating condition samples from different electric drive chips and individual vehicles. Clustering processing can uncover the clustering distribution patterns of historical hardware protection events in the multi-dimensional operating condition space, while quantile statistics can objectively output the risk boundaries of sub-feature parameters for each dimension. This provides a data-driven judgment benchmark for subsequent screening of historical hardware protection events that meet preset conditions, reducing subjective biases caused by parameter settings and ensuring the objectivity and reliability of the input basis for subsequent statistical analysis of the aforementioned frequency and determination of the risk probability of external hardware circuit failures. This further improves the adaptability and accuracy of risk identification, supporting vehicles in achieving more reliable proactive fault prevention.

[0064] In one possible implementation, the sub-feature parameters of the multiple dimensions include the peak current of the motor controller. Quantile statistics are performed on the multiple sub-feature parameters of each single dimension to obtain the operating condition boundary range of the corresponding dimension. This includes: sorting the multiple current peaks to obtain sorted multiple current peaks; determining the low quantile value and high quantile value of the sorted multiple current peaks based on the quantile probability value corresponding to the current peak; determining the low quantile value as the lower limit of the operating condition boundary range corresponding to the current peak, and determining the high quantile value as the upper limit of the operating condition boundary range corresponding to the current peak.

[0065] It should be understood that in the above scheme, different quantile probability values ​​are selected for different dimensions. 90% and 99% (adjustable adaptively) are used to define the high-value side boundary, that is, larger quantile probability values ​​are suitable for dimensions where larger values ​​are more likely to trigger hardware protection events, such as the current peak dimension. 1% and 10% (adjustable adaptively) are used to define the low-value side boundary, that is, smaller quantile probability values ​​are suitable for dimensions where smaller values ​​are more likely to trigger hardware protection events, such as the bus voltage dimension.

[0066] It should be noted that for dimensions such as peak current, where larger values ​​increase the likelihood of triggering hardware protection events, in practical applications, there may be instances where the peak current of a historical hardware protection event was greater than [a certain value]. These extreme value events, although numerically exceeding the corresponding boundary condition range, have the same physical causes as hardware protection events within the boundary condition range. If they are not included in the statistics, hardware protection events strongly correlated with latent hardware faults will be missed, thus underestimating the risk probability. Therefore, it is stipulated that "values ​​above the upper limit are still classified as high-risk operating condition events." This ensures that extreme value events are not truncated by the boundary condition range and are not lost, allowing the final statistical count to fully cover all hardware protection events strongly correlated with latent hardware faults, guaranteeing the accuracy and conservatism of the risk probability. Similarly, for dimensions such as bus voltage, where smaller values ​​are more likely to trigger hardware protection events, it is stipulated that "values ​​below the upper limit are still classified as high-risk operating condition events."

[0067] In the above technical solution, quantile statistics are performed on the current peak values ​​of the motor controller, and multiple current peak values ​​are sorted. This enables the orderly organization of multiple current peak values, providing a data foundation for quantile value extraction. Subsequently, based on the quantile probability values ​​corresponding to the current peak values, the lower and higher quantile values ​​of the sorted current peak values ​​are determined. This eliminates the need to forcibly specify a fixed current threshold, and can adaptively generate the corresponding operating condition boundary range based on the current peak value dimension and the distribution of current peak values, filtering out interference from a small number of discrete current samples. The above solution can adapt to sporadic faults triggered only in specific current ranges due to external hardware circuit failures (i.e., latent hardware faults), obtaining objective and accurate current operating condition boundaries, and reducing the risk of fault misjudgment caused by manually setting thresholds.

[0068] In some embodiments, the sorted multiple current peaks are obtained by sorting multiple current peaks from smallest to largest. The quantile probability value includes a first probability value and a second probability value, where the second probability value is greater than the first probability value. Based on the quantile probability value corresponding to the current peak, the lower quantile value and the higher quantile value of the sorted multiple current peaks are determined, including: determining the product between the first probability value and the number of current peaks to obtain a first quantile; determining the product between the second probability value and the number of current peaks to obtain a second quantile; determining the first current peak corresponding to the first quantile from the sorted multiple current peaks and determining the first current peak as the lower quantile value; and determining the second current peak corresponding to the second quantile from the sorted multiple current peaks and determining the second current peak as the higher quantile value.

[0069] It should be understood that the above describes the process for determining the low-order and high-order percentile values ​​corresponding to the current peak value. For dimensions like the current peak value, where larger values ​​are more likely to trigger hardware protection events, the principle for determining the low-order and high-order percentile values ​​is the same. Therefore, the process for determining the low-order and high-order percentile values ​​for dimensions other than the current peak value, where larger values ​​are more likely to trigger hardware protection events, will not be repeated here. Furthermore, for dimensions like the bus voltage value, where smaller values ​​are more likely to trigger hardware protection events, the process for determining the low-order and high-order percentile values ​​can also refer to the current peak value, and will not be repeated here.

[0070] In some embodiments, before clustering multiple operating condition feature parameters to obtain the region where the event is located, the method 200 further includes: normalizing each operating condition feature parameter to obtain normalized feature parameters; and clustering multiple operating condition feature parameters to obtain the region where the event is located, including: clustering multiple normalized feature parameters to obtain the region where the event is located.

[0071] The following describes in detail several embodiments of determining the risk probability in step 203.

[0072] The first method is to determine the risk probability based on the number of occurrences (i.e., the statistical number of occurrences mentioned above).

[0073] In one possible implementation, step 203, which determines the risk probability of the external hardware circuit of the electric drive chip failing based on the number of historical hardware protection events, includes: comparing the number of historical hardware protection events with multiple preset number ranges; determining a target preset number range that matches the number of historical hardware protection events from the multiple preset number ranges, where each preset number range corresponds to a sample risk probability; and determining the sample risk probability corresponding to the target preset number range as the risk probability.

[0074] It should be understood that a larger preset number of times within the preset range corresponds to a higher probability of sample risk.

[0075] In the above technical solution, the number of historical hardware protection events is compared with multiple preset count ranges, and the target preset count range is determined by interval matching. This eliminates the need for complex floating-point calculations. Subsequently, the counts are directly converted into the probability of external hardware circuit failure through interval mapping. This simplifies the calculation logic of the risk probability and reduces computational overhead. In other words, the above solution uses a hierarchical interval mapping method, which can mitigate the probability jitter caused by fluctuations in single counts and determine accurate risk probabilities. Furthermore, the logic of this method is simple and easy to implement, enabling rapid output of risk probabilities. Therefore, the above solution can balance the calculation efficiency and accuracy of risk probabilities.

[0076] For example, the preset number of times range includes , , , and The corresponding sample risk probabilities are 5%, 20%, 50%, 80%, and 95%, which can be interpreted as low risk, low-medium risk, medium-high risk, high risk, and extremely high risk, respectively.

[0077] In some embodiments, before determining the risk probability of a failure in the external hardware circuit of the electric drive chip based on the number of historical hardware protection events in step 203, the method 200 further includes: determining whether the number of historical hardware protection events is greater than or equal to a third preset number, wherein the third preset number is less than the second preset number; determining the risk probability of a failure in the external hardware circuit of the electric drive chip based on the number of historical hardware protection events in step 203 includes: determining the risk probability if the number of historical hardware protection events is greater than or equal to the third preset number.

[0078] In some embodiments, the method 200 further includes: if the number of historical hardware protection events is less than the third preset number, continuously monitoring the hardware protection events of the electric drive chip and prohibiting the determination of risk probability.

[0079] The second approach also incorporates other influencing factors (the number of times historical hardware protection events are triggered and the maximum recovery time) to determine the probability of risk.

[0080] In one possible implementation, the sub-feature parameters of the multiple dimensions also include the recovery time of historical hardware protection events. The method 200 further includes: determining the maximum recovery time among the multiple operating condition feature parameters; determining a first weight based on the number of triggers of historical hardware protection events, and determining a second weight based on the maximum recovery time. The first weight is used to indicate the contribution of the number of triggers in determining the risk probability, and the second weight is used to indicate the contribution of the maximum recovery time in determining the risk probability. In step 203, determining the risk probability of the external hardware circuit of the electric drive chip failing based on the number of historical hardware protection events includes: weighting and fusing the number of triggers, the maximum recovery time, and the number of historical hardware protection events based on the first weight, the second weight, and the third weight to obtain the risk probability. The third weight is used to indicate the contribution of the number of triggers in determining the risk probability.

[0081] It should be understood that in the above scheme, each historical hardware protection event corresponds to a recovery time, and the maximum recovery time is the maximum value among the multiple recovery times corresponding to multiple historical hardware protection events. The trigger count refers to the total number of historical hardware protection events.

[0082] In addition, the sum of the first weight, the second weight, and the third weight is 1, and the first weight and the second weight have corresponding weight caps, which can both be 0.45.

[0083] In the above technical solution, the recovery time of historical hardware protection events is incorporated into the sub-feature parameters to obtain the maximum recovery time among the operating condition feature parameters. This dimension indicator can reflect the severity of hardware protection events. Subsequently, based on the number of triggers and the maximum recovery time, a first weight and a second weight are determined respectively. Based on the first weight, the second weight, and the third weight, the number of triggers, the maximum recovery time, and the number of events are weighted and fused to obtain the risk probability. This overcomes the limitation of relying solely on a single event count to assess risk probability, taking into account the number of hardware protection events triggered, their severity, and the statistical count of valid samples (the number of valid hardware protection events, i.e., the statistical count of valid events). Furthermore, the weights allow for flexible adjustment of the influence ratio of each dimension indicator, adapting to the performance characteristics of different faults and suppressing assessment bias caused by fluctuations in a single indicator. Therefore, the above solution can improve the accuracy and comprehensiveness of assessing the risk probability of external hardware circuit failures, providing a more comprehensive basis for subsequent vehicle management.

[0084] In some embodiments, before weighting and fusing the number of triggers, the maximum recovery time, and the number of historical hardware protection events based on the first weight, the second weight, and the third weight to obtain the risk probability, the method 200 further includes: determining whether the number of historical hardware protection events is greater than or equal to a third preset number; weighting and fusing the number of triggers, the maximum recovery time, and the number of historical hardware protection events based on the first weight, the second weight, and the third weight to obtain the risk probability, including: if the number of historical hardware protection events is greater than or equal to the third preset number, weighting and fusing the number of triggers, the maximum recovery time, and the number of historical hardware protection events based on the first weight, the second weight, and the third weight to obtain the risk probability.

[0085] In some embodiments, determining a first weight based on the number of times a historical hardware protection event is triggered includes: obtaining a first trigger threshold, wherein when the number of triggers reaches the first trigger threshold, the first weight reaches a weight upper limit; determining the ratio between the number of triggers and the first trigger threshold to obtain a first coefficient; if the first coefficient is less than a preset coefficient, determining the product between the first coefficient and the weight upper limit as the first weight; and if the first coefficient is greater than or equal to the preset coefficient, determining the first weight as the weight upper limit.

[0086] It should be understood that the above-mentioned preset coefficient is 1.

[0087] In some embodiments, determining a second weight based on the maximum recovery time includes: obtaining a first duration threshold, wherein when the maximum recovery time reaches the first duration threshold, the second weight reaches a weight upper limit; determining the ratio between the maximum recovery time and the first duration threshold to obtain a second coefficient; if the second coefficient is less than a preset coefficient, determining the product between the second coefficient and the weight upper limit as the second weight; and if the second coefficient is greater than or equal to the preset coefficient, determining the second weight as the weight upper limit.

[0088] In one possible implementation, the risk probability is obtained by weighted fusion of the number of triggers, the maximum recovery time, and the number of historical hardware protection events based on the first weight, the second weight, and the third weight. This includes: determining a first amplitude as the ratio between the number of triggers and a first preset number; determining a second amplitude as the ratio between the maximum recovery time and the first preset time; determining a third amplitude based on whether the number of historical hardware protection events is greater than the second preset number, wherein the second preset number is less than the first preset number; and weighted fusion of the first amplitude, the second amplitude, and the third amplitude based on the first weight, the second weight, and the third weight to obtain the risk probability.

[0089] It should be understood that in the above scheme, the first preset number of times is the upper limit of the allowed number of triggers, and the first preset duration is the upper limit of the allowed recovery duration.

[0090] In the above technical solution, the ratio between the number of triggers and a first preset number of triggers is determined as the first amplitude, and the ratio between the maximum recovery time and a first preset time is determined as the second amplitude. This eliminates the dimensional differences between different physical quantities through normalization, making indicators of different dimensions comparable. A third amplitude is determined based on whether the number of triggers exceeds the second preset number of triggers. This standardizes the valid events using the second preset number of triggers. Subsequently, the first, second, and third amplitudes are weighted and fused based on the first, second, and third weights to obtain the risk probability. This avoids the problem of weighting result distortion caused by differences in the numerical magnitudes of the original parameters, taking into account the number of triggers, severity, and the statistical number of valid events. The contribution of each indicator is flexibly adjusted by the weights, which reduces the evaluation bias caused by scale differences between different parameters and improves the stability and rationality of determining the risk probability.

[0091] In some embodiments, the risk probability is obtained by weighted fusion of the first amplitude, the second amplitude, and the third amplitude based on the first weight, the second weight, and the third weight, including: determining the product between the first weight and the first amplitude to obtain a third probability; determining the product between the second weight and the second amplitude to obtain a fourth probability; determining the product between the third weight and the third amplitude to obtain a fifth probability; and determining the sum of the third probability, the fourth probability, and the fifth probability as the risk probability.

[0092] In some embodiments, the risk probability is obtained by weighted fusion of the first amplitude, the second amplitude, and the third amplitude based on the first weight, the second weight, and the third weight, including: determining the risk probability based on the following formula (1); (1) in, For this risk probability, As the first weight, As the second weight, As the third weight, The first value, The second value, This is the third value.

[0093] In one possible implementation, determining a third amplitude based on whether the number of historical hardware protection events is greater than a second preset number includes: if the number of historical hardware protection events is greater than the second preset number, determining the third amplitude as a first preset amplitude, the first preset amplitude being used to indicate that the probability of an external hardware circuit failure is greater than a first probability; if the number of historical hardware protection events is less than or equal to the second preset number, determining the third amplitude as a second preset amplitude, the second preset amplitude being less than the first preset amplitude.

[0094] It should be understood that in the above scheme, the first preset amplitude can be 1, indicating that the external hardware circuit is in a high-risk fault state in the dimension corresponding to the number of occurrences. The second preset amplitude is 0.

[0095] In the above technical solution, a second preset number of times is used as the dividing line to determine the magnitude of the statistical count of valid events. When the statistical count is greater than the second preset number of times, the third amplitude is assigned the first preset amplitude. This is used to characterize that the external hardware circuit has a high probability of failure in such cases. When the statistical count is less than or equal to the second preset number of times, it is assigned the second preset amplitude, which is smaller in value. This enables discretized amplitude mapping of the statistical count. The above solution does not require complex continuous calculations on the statistical count, thus reducing computational overhead. In addition, directly determining the third amplitude by the preset number range in which the statistical count falls also allows for rapid acquisition of the third amplitude, improving the speed of determining the risk probability.

[0096] In one possible implementation, the vehicle is controlled based on the risk probability, including: determining whether the risk probability is greater than a second probability, which is greater than a first probability; and if the risk probability is greater than the second probability, outputting a maintenance list to suggest that the target object repair the external hardware circuit based on the maintenance list.

[0097] It should be understood that the aforementioned second probability can be considered as the probability value corresponding to a higher risk level. At this point, the risk of hardware failure has exceeded the safety threshold, and the external hardware circuit is highly likely to have suffered substantial damage or face catastrophic failure at any time (such as short circuit and fire). In other words, the current hardware circuit failure risk has reached the point where immediate physical repair is necessary, and a repair list should be directly generated to guide maintenance personnel to intervene and eliminate safety hazards.

[0098] In the above technical solution, the risk probability is compared with a second probability, which has a larger value, to distinguish the level of fault risk. When the risk probability is greater than the second probability, a repair list is output to suggest that the target device should repair the external hardware circuitry. The above judgment logic is based on quantified risk probability as the triggering basis, rather than solely on fault codes to trigger repair prompts. This allows for the identification of potential hidden dangers in the external hardware circuitry of the electric drive chip before a fault occurs. Setting a second probability greater than the first probability as the trigger threshold can filter out low-risk noise samples and avoid frequent false repair suggestions. In addition, outputting a repair list allows maintenance personnel to quickly locate the hardware points to be checked, reducing the possibility of fault deterioration leading to abnormal vehicle operation, improving the vehicle's proactive prevention capabilities, and also reducing the maintenance costs associated with reactive handling after a fault occurs.

[0099] In some embodiments, before outputting the repair list, the method 200 further includes: identifying the vehicle as an affected vehicle and transmitting the vehicle's identification code and information referred to as the affected vehicle to the cloud.

[0100] In some embodiments, the number of vehicles marked as affected within the same model and production batch, and the total number of vehicles operating online within the same model and production batch, are obtained from the cloud; the ratio between the number of affected vehicles and the total number of vehicles is determined to obtain a batch risk coefficient, which is used to indicate the overall risk level of failure of the external hardware circuit of the electric drive chip within the same model and production batch; if the batch risk coefficient is greater than or equal to a risk coefficient threshold, it is determined that there is a batch-specific design defect in the external hardware circuit; if the batch risk coefficient is less than the risk coefficient threshold, it is determined that the failure of the external hardware circuit of the electric drive chip in the vehicle is an isolated event occurring in a single vehicle.

[0101] Optionally, if a batch design defect is found in the external hardware circuit, multiple vehicles corresponding to the same model can be recalled to redesign the external hardware circuit of the electric drive chip.

[0102] In one possible implementation, the method 200 further includes: after repairing the external hardware circuit, obtaining the first number of times the hardware protection event is triggered within a unit mileage under preset operating conditions, wherein the preset operating conditions are conditions that are prone to triggering hardware protection events; determining a fourth amplitude based on the first and second trigger counts, wherein the second trigger count is the trigger count corresponding to the time before repair, and the fourth amplitude is used to indicate the degree of reduction in the number of triggers; and determining the repair result based on the fourth amplitude, wherein the repair result is the result of repairing the external hardware circuit.

[0103] It should be understood that in the above scheme, the preset operating conditions can be specific conditions such as high current and high load.

[0104] In the above technical solution, after the external hardware circuit has been inspected, a preset operating condition that is prone to triggering hardware protection events is selected, and the first trigger count per unit mileage is obtained. Using this type of operating condition can fully expose potential fault causes and avoid obtaining false inspection results under mild operating conditions. Based on the first trigger count after inspection and the second trigger count before inspection, a fourth amplitude is determined, which allows for a quantitative expression of the repair effect. This solution eliminates the process of judging repair effectiveness based on subjective experience and unifies the evaluation criteria for repair effectiveness. This can identify hidden situations where faults are not completely eliminated, providing objective evidence for subsequent re-inspection and repair upgrades, and improving the closed-loop process for fault risk management.

[0105] In some embodiments, determining a fourth amplitude based on the first trigger count and the second trigger count includes: determining the ratio between the first trigger count and the second trigger count to obtain a fifth amplitude; and determining the difference between the preset coefficient and the fifth amplitude as the fourth amplitude.

[0106] In one possible implementation, determining the repair result based on the fourth amplitude includes: if the first trigger count is less than the second trigger count and the fourth amplitude is greater than or equal to the third preset amplitude, determining the repair result as valid; if the fourth amplitude is greater than the fourth preset amplitude and less than the third preset amplitude, determining the repair result as partially valid and outputting a re-inspection list, which is used to instruct the execution of a preset verification process; if the fourth amplitude is less than or equal to the fourth preset amplitude, determining the repair result as invalid and outputting a reminder message, which is used to remind the vehicle to be sent to a higher-level repair facility for in-depth repair.

[0107] It should be understood that, in the above solutions, "effective repair" means that the target (repair personnel) completely resolves the fault after repairing the external hardware circuit of the electric drive chip. "Partially effective repair" means that the target has been able to alleviate the fault symptoms or reduce the probability of the fault occurring, but has not completely eradicated the problem. "Ineffective repair" means that the repair measures taken by the target have failed to resolve the fault problem of the external hardware circuit.

[0108] In the above technical solution, when the first trigger count is less than the second trigger count and the fourth amplitude is greater than or equal to the third preset amplitude, the repair is deemed effective. This takes into account both the downward trend of the trigger count and the degree of improvement. When the fourth amplitude is between the fourth preset amplitude and the third preset amplitude, the repair is deemed partially effective, and a re-inspection list is output. This guides the target to execute the preset verification process. When the fourth amplitude is less than or equal to the fourth preset amplitude, the repair is deemed ineffective, and a reminder message is output. This reminds the target to send the vehicle to a higher-level repair facility for in-depth repair. In other words, the above solution divides the repair effectiveness into three categories of results through graded thresholds. This avoids the problem of coarse judgment caused by a single threshold and allows for differentiated handling strategies to be matched for different degrees of improvement. This can recognize repair results that are completely repaired, as well as cases of insufficient improvement and complete failure, forming a graded closed-loop handling logic, improving the precision of repair effect evaluation, and perfecting the whole-process risk management of the external hardware circuit of the electric drive chip.

[0109] Figure 3 This is a schematic diagram of a device for controlling a vehicle provided in an embodiment of this application.

[0110] For example, the vehicle includes an electric drive system, which includes an electric drive core, such as... Figure 3 As shown, the device 300 includes: The acquisition module 301 is used to acquire the operating condition characteristic parameters corresponding to multiple historical hardware protection events of the electric drive chip. The operating condition characteristic parameters include sub-characteristic parameters of multiple dimensions. The determination module 302 is used to determine the number of historical hardware protection events that meet preset conditions among the multiple historical hardware protection events based on multiple operating condition characteristic parameters. The preset conditions are used to indicate that the corresponding historical hardware protection event is in the event area and at least one corresponding operating condition characteristic parameter is in the operating condition boundary range of the corresponding dimension. The event area includes hardware protection events where the difference between the operating condition characteristic parameters is less than a preset difference. The control module 303 is used to determine the probability of failure of the external hardware circuit of the electric drive chip based on the number of historical hardware protection events, and to control the vehicle based on the probability of failure.

[0111] Optionally, before determining the number of historical hardware protection events that meet the preset conditions among the multiple historical hardware protection events based on multiple operating condition feature parameters, the determining module 302 is further configured to: cluster the multiple operating condition feature parameters to obtain the region where the event is located; and perform quantile statistics on multiple sub-feature parameters of each single dimension to obtain the operating condition boundary range of the corresponding dimension.

[0112] Optionally, the sub-feature parameters of the multiple dimensions include the peak current of the motor controller. The determining module 302 is specifically used to: sort the multiple peak currents to obtain sorted multiple peak currents; determine the low quantile value and high quantile value of the sorted multiple peak currents based on the quantile probability value corresponding to the peak current; determine the low quantile value as the lower limit of the operating condition boundary range corresponding to the peak current, and determine the high quantile value as the upper limit of the operating condition boundary range corresponding to the peak current.

[0113] Optionally, the determining module 302 is further configured to: compare the number of the historical hardware protection events with multiple preset number ranges, determine a target preset number range that matches the number of the historical hardware protection events from the multiple preset number ranges, wherein each preset number range corresponds to a sample risk probability; and determine the sample risk probability corresponding to the target preset number range as the risk probability.

[0114] Optionally, the sub-feature parameters of the multiple dimensions also include the recovery time of historical hardware protection events. The determining module 302 is further configured to: determine the maximum recovery time among the multiple operating condition feature parameters; determine a first weight based on the number of triggers of historical hardware protection events, and determine a second weight based on the maximum recovery time, wherein the first weight is used to indicate the contribution of the number of triggers in determining the risk probability, and the second weight is used to indicate the contribution of the maximum recovery time in determining the risk probability; and the determining module 302 is further configured to perform weighted fusion of the number of triggers, the maximum recovery time, and the number of historical hardware protection events based on the first weight, the second weight, and the third weight to obtain the risk probability, wherein the third weight is used to indicate the contribution of the number of events in determining the risk probability.

[0115] Optionally, the determining module 302 is specifically used to: determine the ratio between the number of triggers and the first preset number as a first amplitude; determine the ratio between the maximum recovery time and the first preset time as a second amplitude; determine a third amplitude based on whether the number of historical hardware protection events is greater than the second preset number, wherein the second preset number is less than the first preset number; and perform weighted fusion of the first amplitude, the second amplitude, and the third amplitude based on the first weight, the second weight, and the third weight to obtain the risk probability.

[0116] Optionally, the determining module 302 is further configured to: determine the third amplitude as the first preset amplitude when the number of historical hardware protection events is greater than the second preset number, wherein the first preset amplitude is used to indicate that the probability of an external hardware circuit failure is greater than the first probability; and determine the third amplitude as the second preset amplitude when the number of historical hardware protection events is less than or equal to the second preset number, wherein the second preset amplitude is less than the first preset amplitude.

[0117] Optionally, the determining module 302 is specifically used to determine whether the risk probability is greater than the second probability, and the second probability is greater than the first probability; the control module 303 is specifically used to output a maintenance list when the risk probability is greater than the second probability, so as to suggest that the target object repair the external hardware circuit based on the maintenance list.

[0118] Optionally, the acquisition module 301 is further configured to acquire, after repairing the external hardware circuit, the first trigger count of the hardware protection event within a unit mileage under preset operating conditions, wherein the preset operating conditions are operating conditions that are prone to triggering hardware protection events; the determination module 302 is further configured to: determine a fourth amplitude based on the first trigger count and the second trigger count, wherein the second trigger count is the trigger count corresponding to the time before repair, and the fourth amplitude is used to indicate the degree of reduction in the trigger count; and determine the repair result based on the fourth amplitude, wherein the repair result is the result of repairing the external hardware circuit.

[0119] Optionally, the determining module 302 is further configured to: determine the repair result as valid when the first trigger count is less than the second trigger count and the fourth amplitude is greater than or equal to the third preset amplitude; determine the repair result as partially valid when the fourth amplitude is greater than the fourth preset amplitude and less than the third preset amplitude, and output a re-inspection list, which is used to instruct the execution of a preset verification process; and determine the repair result as invalid when the fourth amplitude is less than or equal to the fourth preset amplitude, and output a reminder message, which is used to remind the vehicle to be sent to a higher-level repair facility for in-depth repair.

[0120] Figure 4 This is a schematic diagram of the structure of a controller provided in an embodiment of this application.

[0121] For example, such as Figure 4 As shown, the controller 400 includes a storage module 401 and a processing module 402. The storage module 401 stores executable program code 403, and the processing module 402 is used to call and execute the executable program code 403 to perform a method for controlling a vehicle.

[0122] Figure 5 This is a schematic diagram of the structure of a vehicle provided in an embodiment of this application.

[0123] For example, such as Figure 5 As shown, the vehicle 500 includes a memory 501 and a processor 502, wherein the memory 501 stores executable program code 403, and the processor 502 is used to call and execute the executable program code 403 to perform a method for controlling the vehicle.

[0124] Furthermore, embodiments of this application also protect an apparatus that may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to perform a method for controlling a vehicle provided in embodiments of this application.

[0125] This embodiment can divide the device into functional modules based on the above method example. For example, each module can correspond to a separate function, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0126] When the functional modules are divided according to their respective functions, the device may also include an acquisition module, a determination module, and a control module. It should be noted that all relevant content in the above method embodiments can be referenced in the functional descriptions of the corresponding functional modules, and will not be repeated here.

[0127] It should be understood that the device provided in this embodiment is used to execute the above-described method for controlling a vehicle, and therefore can achieve the same effect as the above-described implementation method.

[0128] When using an integrated unit, the device may include a processing module and a storage module. When the device is applied to a vehicle, the processing module can be used to control and manage the vehicle's movements. The storage module can be used to support the vehicle in executing relevant executable program code.

[0129] The processing module may be a processor or a controller, which can implement or execute various exemplary logic blocks, modules, and circuits shown in conjunction with the disclosure of this application. The processor may also be a combination of functions that implement computing capabilities, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc., and the storage module may be a memory.

[0130] In addition, the device provided in the embodiments of this application may specifically be a chip, component or module. The chip may include a connected processor and a memory. The memory is used to store instructions. When the processor calls and executes the instructions, the chip can execute a method for controlling a vehicle provided in the above embodiments.

[0131] This embodiment also provides a computer-readable storage medium storing executable program code. When the executable program code is run on a computer, the computer performs the aforementioned method steps to implement the method for controlling a vehicle provided in the above embodiment.

[0132] This embodiment also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned steps to implement a method for controlling a vehicle provided in the above embodiment.

[0133] In this embodiment, the device, computer-readable storage medium, computer program product, or chip are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0134] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0135] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0136] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for controlling a vehicle, characterized in that, The vehicle includes an electric drive system, the electric drive system includes an electric drive chip, and the method includes: Obtain the operating condition feature parameters corresponding to multiple historical hardware protection events of the electric drive chip, wherein the operating condition feature parameters include sub-feature parameters of multiple dimensions. Based on multiple operating condition feature parameters, the number of historical hardware protection events that meet preset conditions among the multiple historical hardware protection events is determined. The preset conditions are used to indicate that the corresponding historical hardware protection event is in the event region and at least one corresponding operating condition feature parameter is in the operating condition boundary range of the corresponding dimension. The event region includes hardware protection events where the difference between the operating condition feature parameters is less than a preset difference. Based on the number of historical hardware protection events, the probability of failure of the external hardware circuit of the electric drive chip is determined, and the vehicle is controlled based on the probability of failure.

2. The method according to claim 1, characterized in that, Before determining the number of historical hardware protection events that meet preset conditions among the multiple historical hardware protection events based on multiple operating condition characteristic parameters, the method further includes: Clustering is performed on multiple operating condition characteristic parameters to obtain the region where the event occurs; and, Quantile statistics are performed on multiple sub-feature parameters of each single dimension to obtain the working condition boundary range of the corresponding dimension.

3. The method according to claim 2, characterized in that, The multiple sub-feature parameters include the peak current of the motor controller. The quantile statistics of the multiple sub-feature parameters for each single dimension to obtain the corresponding operating condition boundary range includes: The multiple current peak values ​​are sorted to obtain the sorted current peak values; Based on the quantile probability value corresponding to the current peak value, the low quantile value and high quantile value of the sorted multiple current peak values ​​are determined. The low quantile value is determined as the lower limit of the operating condition boundary range corresponding to the current peak value, and the high quantile value is determined as the upper limit of the operating condition boundary range corresponding to the current peak value.

4. The method according to claim 1, characterized in that, The determination of the probability of failure of the external hardware circuit of the electric drive chip based on the number of historical hardware protection events includes: The number of historical hardware protection events is compared with multiple preset number ranges, and a target preset number range that matches the number of historical hardware protection events is determined from the multiple preset number ranges. Each preset number range corresponds to a sample risk probability. The sample risk probability corresponding to the target preset number range is determined as the risk probability.

5. The method according to claim 1, characterized in that, The sub-feature parameters of the multiple dimensions also include the recovery time of historical hardware protection events, and the method further includes: Determine the maximum recovery time among the multiple operating condition characteristic parameters; A first weight is determined based on the number of times historical hardware protection events are triggered, and a second weight is determined based on the maximum recovery time. The first weight is used to indicate the contribution of the number of triggers in determining the risk probability, and the second weight is used to indicate the contribution of the maximum recovery time in determining the risk probability. And, determining the probability of failure of the external hardware circuit of the electric drive chip based on the number of historical hardware protection events includes: Based on the first weight, the second weight, and the third weight, the number of triggers, the maximum recovery time, and the number of historical hardware protection events are weighted and fused to obtain the risk probability. The third weight is used to indicate the degree of contribution of the number of events in determining the risk probability.

6. The method according to claim 5, characterized in that, The risk probability is obtained by weighting and fusing the number of triggers, the maximum recovery time, and the number of historical hardware protection events based on the first weight, the second weight, and the third weight, including: The ratio between the number of triggers and the first preset number is determined as the first amplitude. The ratio between the maximum recovery time and the first preset time is determined as the second amplitude. A third value is determined based on whether the number of historical hardware protection events is greater than a second preset number, wherein the second preset number is less than the first preset number; Based on the first weight, the second weight, and the third weight, the first amplitude, the second amplitude, and the third amplitude are weighted and fused to obtain the risk probability.

7. The method according to claim 6, characterized in that, The determination of the third value based on whether the number of historical hardware protection events is greater than the second preset number includes: If the number of historical hardware protection events is greater than the second preset number, the third amplitude is determined as the first preset amplitude, and the first preset amplitude is used to indicate that the probability of external hardware circuit failure is greater than the first probability. If the number of historical hardware protection events is less than or equal to the second preset number, the third amplitude is determined as the second preset amplitude, which is less than the first preset amplitude.

8. The method according to claim 1, characterized in that, The control of the vehicle based on the risk probability includes: Determine whether the risk probability is greater than the second probability, where the second probability is greater than the first probability; If the probability of risk is greater than the second probability, a maintenance list is output to recommend that the target object repair the external hardware circuit based on the maintenance list.

9. The method according to claim 8, characterized in that, The method further includes: After inspecting the external hardware circuit, the first number of times the hardware protection event is triggered within a unit mileage of the vehicle under preset operating conditions is obtained. The preset operating conditions are those that are prone to triggering hardware protection events. Based on the first and second trigger counts, a fourth value is determined, wherein the second trigger count is the trigger count before maintenance, and the fourth value is used to indicate the degree of reduction in the trigger count; Based on the fourth amplitude value, the repair result is determined, which is the result of repairing the external hardware circuit.

10. The method according to claim 9, characterized in that, The determination of the maintenance result based on the fourth amplitude includes: If the first number of triggers is less than the second number of triggers and the fourth amplitude is greater than or equal to the third preset amplitude, the inspection result is determined to be valid. If the fourth amplitude is greater than the fourth preset amplitude and less than the third preset amplitude, the inspection result is determined to be valid and a re-inspection list is output. The re-inspection list is used to indicate the execution of a preset verification process. If the fourth amplitude is less than or equal to the fourth preset amplitude, the inspection result is determined to be invalid, and a reminder message is output. The reminder message is used to remind the vehicle to be sent to a higher-level repair institution for in-depth inspection.

11. A vehicle, characterized in that, The vehicles include: Memory, used to store executable program code; A processor for calling and running the executable program code from the memory, causing the vehicle to perform the method as described in any one of claims 1 to 10.