A threshold optimization method, device and electronic equipment for a vehicle anti-pinch scene

Abstract

The application relates to a threshold optimization method, device and electronic equipment for a vehicle anti-pinch scene, and is applied to a cloud server. The method comprises the following steps: determining current anti-pinch category information of a vehicle end based on obtained anti-pinch scene classification scores; the anti-pinch scene classification scores are determined by analyzing fault phenomena and fault causes of running state data of a motor in the vehicle end; the running state data of the motor is obtained by collecting data of the running state of the motor when an anti-pinch event is triggered at a target pinch position; in the case that the anti-pinch category information is false anti-pinch, a target anti-pinch judgment threshold corresponding to the target pinch position is determined based on the running state data of the motor sent by the vehicle end; and the target anti-pinch judgment threshold is sent to the vehicle end; wherein the vehicle end is used for performing vehicle anti-pinch judgment based on the target anti-pinch judgment threshold, which can improve the accuracy of subsequent anti-pinch judgment of the vehicle end to a certain extent and effectively reduce the false anti-pinch rate.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and specifically to a threshold optimization method, apparatus, and electronic device for vehicle anti-pinch scenarios. Background Technology

[0002] Current vehicles commonly employ designs such as power windows, power hood switches, and power trunk lid switches. Due to vehicle safety regulations, these power switches must be equipped with corresponding anti-pinch measures to ensure user safety. Traditional solutions monitor changes in the drive motor's current. For example, with power windows, when the glass encounters an obstacle, the resistance increases sharply, requiring the motor to output greater torque to overcome the resistance. This is directly reflected in a significant and rapid increase in operating current. Using this characteristic, the system can effectively detect a pinching event through abnormal current surges. However, current anti-pinch algorithms often experience false alarms, triggering the anti-pinch function when no object is being held, causing windows to descent abnormally and the hood and trunk lid to fail to close properly. The current anti-pinch algorithms have a high false alarm rate, resulting in a poor user experience. Summary of the Invention

[0003] This application provides a threshold optimization method, apparatus, electronic device, computer-readable storage medium, and computer program product for vehicle anti-pinch scenarios.

[0004] In a first aspect, embodiments of this application provide a threshold optimization method for vehicle anti-pinch scenarios, applied to a cloud server; the method includes: Based on the acquired anti-pinch scenario classification score, the current anti-pinch category information of the vehicle is determined. The anti-pinch scenario classification score is determined by analyzing the fault phenomena and causes of the motor's operating status data in the vehicle. The motor's operating status data is obtained by collecting data on the motor's operating status when the anti-pinch event is triggered at the target clamping position. The fault phenomena include at least one of the following: voltage-related, current-related, anti-pinch position-related, signal quality-related, and user operation abnormality-related. The causes of the faults include at least one of the following: mechanical error-related, signal error-related, and sensor error-related. The anti-pinch category information includes correct anti-pinch and false anti-pinch. When the anti-pinch category information is false anti-pinch, the target anti-pinch judgment threshold corresponding to the target clamping position is determined based on the motor operating status data sent by the vehicle. Send the target anti-pinch determination threshold to the vehicle terminal; wherein, the vehicle terminal is used to perform vehicle anti-pinch determination based on the target anti-pinch determination threshold.

[0005] Based on the aforementioned technical means, the cloud server can determine whether the most recent anti-pinch event on the vehicle is a false pinch or a normal pinch based on the anti-pinch scenario classification score generated after an anti-pinch event occurs on the vehicle. When the most recent anti-pinch event on the vehicle is identified as a false pinch, the anti-pinch judgment threshold is adjusted based on the motor operating status data generated by the event. The adjusted target anti-pinch judgment threshold is then sent to the vehicle. The vehicle can use the target anti-pinch judgment threshold for subsequent anti-pinch judgments. Through the anti-pinch scenario classification score, the vehicle's anti-pinch events can be quantified, thereby clarifying the boundary between correct and false pinches and improving the accuracy of correct and false pinch judgments. Since false pinches usually occur when the anti-pinch judgment threshold is too low, sending the target anti-pinch judgment threshold through the cloud server can improve the accuracy of subsequent anti-pinch judgments on the vehicle to a certain extent and effectively reduce the false pinch rate.

[0006] Furthermore, based on the motor's operating status data, the target anti-pinch judgment threshold corresponding to the target clamping position is determined, including: Determine the cause of the anti-pinch incident; In cases where the cause of the event includes mechanical error and / or signal error, determine the first-order and second-order difference values ​​of the motor's operating status data, and determine the first-order and second-order difference thresholds corresponding to the target clamping position. If the first-order difference value is greater than the first-order difference threshold and the second-order difference value is greater than the second-order difference threshold, the anti-pinch judgment threshold corresponding to the target clamping position is increased according to the preset rules to obtain the target anti-pinch judgment threshold.

[0007] Based on the above technical means, the first-order difference value can be used to evaluate the rate of change of the motor's operating status data, and the second-order difference value can be used to evaluate the local abrupt change of the motor's operating status data. When both the first-order difference value and the second-order difference value meet the conditions, it indicates that the motor's operating status data changes rapidly and abruptly occurs. At this time, the anti-pinch judgment threshold corresponding to the target clamping position can be adjusted upward according to the preset rules, thereby improving the accuracy and reliability of the adjusted target anti-pinch judgment threshold to a certain extent.

[0008] Furthermore, the motor's operating status data includes at least drive current sequence data; the drive current sequence data includes the correspondence between multiple drive positions and drive currents; determining the first-order and second-order difference values ​​of the motor's operating status data includes: Based on multiple correspondences, the first-order and second-order difference values ​​of the driving current relative to the driving position are determined respectively.

[0009] Based on the above technical means, the correspondence between multiple driving positions and driving currents can form a curve of driving current changing with position. Since the accuracy of driving current measurement is higher, the first-order difference value and second-order difference value obtained through driving current sequence data can have higher accuracy. Therefore, the above method can improve the accuracy and reliability of the first-order difference value and second-order difference value to a certain extent.

[0010] Furthermore, the first-order and second-order difference thresholds corresponding to the target clamping position are determined, including: Historical drive current sequence data to determine the target clamping position; Determine the first-order difference mean, first-order difference standard deviation, second-order difference mean, and second-order difference standard deviation of the historical drive current sequence data. Based on the mean and standard deviation of the first-order current difference, the first-order difference threshold corresponding to the target clamping position is determined; and based on the mean and standard deviation of the second-order current difference, the second-order difference threshold corresponding to the target clamping position is determined.

[0011] Based on the aforementioned technical means, the cloud server can calculate the mean and standard deviation of the first-order difference of the current based on historical driving current sequence data. Then, it can calculate the corresponding first-order and second-order difference thresholds based on these values. Since the calculation ratios of the mean and standard deviation of the first-order difference are directly determined by the preset confidence level in actual calculations, this method can improve the accuracy of the first-order and second-order difference thresholds to a certain extent. This enhances the effectiveness of comparisons between first-order difference values ​​and first-order difference thresholds, as well as between second-order difference values ​​and second-order difference thresholds, thereby fully ensuring the necessity and effectiveness of the final adjustment of the anti-pinch judgment threshold.

[0012] Furthermore, the anti-pinch judgment threshold corresponding to the target clamping position is increased according to preset rules to obtain the target anti-pinch judgment threshold, including: The anti-pinch judgment threshold corresponding to the target clamping position is calculated by multiplying it with the preset adjustment coefficient to obtain the target anti-pinch judgment threshold; the preset adjustment coefficient is greater than 1.

[0013] By using the above-mentioned technical means, the anti-pinch judgment threshold corresponding to the target clamping position can be increased according to a preset multiplier, which can effectively control the increase of the anti-pinch judgment threshold and improve the accuracy of the target anti-pinch judgment threshold to a certain extent.

[0014] Furthermore, based on the anti-pinch scenario classification score sent by the vehicle, the current anti-pinch category information of the vehicle is determined, including at least one of the following: If the anti-pinch scenario classification score sent by the vehicle is greater than or equal to the preset score threshold, the current anti-pinch category information of the vehicle is determined to be false anti-pinch. If the anti-pinch scenario classification score is less than the preset score threshold, the anti-pinch category information is determined to be the correct anti-pinch.

[0015] Based on the above technical means, the relationship between the preset scoring threshold and the classification scores of the anti-pinch scenario can be used to determine whether the current anti-pinch event is a false alarm or a correct anti-pinch. Since the preset scoring threshold is obtained through measurement experiments, it can improve the accuracy of the anti-pinch category information to a certain extent.

[0016] Secondly, embodiments of this application provide a threshold optimization method for vehicle anti-pinch scenarios, applied to the vehicle end; the method includes: When an anti-pinch event is triggered at the target clamping position on the vehicle side, the operating status of the drive motor that triggered the anti-pinch event is collected to obtain the motor's operating status data. Analyze the fault phenomena and causes of the motor's operating status data to determine the scenario classification score of the current anti-pinch scenario on the vehicle side; the fault phenomenon types include at least one of voltage-related, current-related, anti-pinch position-related, signal quality-related, and user operation abnormality-related; the fault cause types include at least one of mechanical error-related, signal error-related, and sensor error-related. Send scenario classification scores to the cloud server; Based on the target anti-pinch judgment threshold sent by the cloud server, the vehicle anti-pinch judgment is performed; the cloud server is used to determine the target anti-pinch judgment threshold based on the anti-pinch scenario classification score and the motor's operating status data.

[0017] Based on the aforementioned technical means, the vehicle can generate the operating status data of the motor at the target clamping position when an anti-pinch event is triggered. It can further determine the scenario classification score of the current anti-pinch scenario, and then use the computing resources of the cloud server to derive the corresponding target anti-pinch judgment threshold to execute subsequent vehicle anti-pinch judgments. Through the anti-pinch scenario classification score, the vehicle's anti-pinch events can be quantified, thereby clarifying the boundary between correct and false anti-pinch, and improving the accuracy of the judgment between correct and false anti-pinch. Since false anti-pinch occurs when the anti-pinch judgment threshold is too low, the distribution of the target anti-pinch judgment threshold through the cloud server can, to a certain extent, improve the accuracy of subsequent anti-pinch judgments on the vehicle side and effectively reduce the false anti-pinch rate.

[0018] Furthermore, the motor's operating status data includes at least drive current sequence data, motor speed, and ambient temperature; the fault phenomena and causes of the motor's operating status data are analyzed to determine the scenario classification score of the current anti-pinch scenario on the vehicle side, including: Based on drive current sequence data, motor speed and ambient temperature, the target fault phenomenon and target fault cause at the vehicle end are determined. The weighted score of the mapping formed by the fault phenomena and causes of each target is obtained by looking up the table. The weighted scores are summed to obtain the scene classification score for the current anti-pinch scenario.

[0019] Based on the aforementioned technical means, the vehicle end can determine the weight score corresponding to each fault phenomenon through drive current sequence data, motor speed, and ambient temperature. Then, the corresponding scenario classification score can be calculated based on the weight score, which can improve the data richness of the motor's operating status data. Since the weight score can effectively reflect the probability of occurrence of each fault phenomenon, the scenario classification score can effectively characterize the actual situation of the current anti-pinch scenario and improve the accuracy of the scenario classification score to a certain extent.

[0020] Furthermore, the method also includes: Obtain anti-pinch feedback information input by the user; If the anti-pinch feedback information indicates that the anti-pinch category information is normal anti-pinch, the judgment threshold used in the vehicle anti-pinch determination will be reverted from the target anti-pinch determination threshold to the most recent anti-pinch determination threshold.

[0021] Based on the above technical means, the vehicle can also accept user feedback information, and when the user's anti-pinch feedback information indicates that the anti-pinch category information is normal anti-pinch, the target anti-pinch judgment threshold issued by the cloud server is rolled back. Through the user's active feedback, the problem of abnormal increase in the anti-pinch judgment threshold caused by the cloud server's misjudgment can be reduced to a certain extent, which can improve the accuracy and reliability of the vehicle's anti-pinch judgment to a certain extent.

[0022] Thirdly, embodiments of this application provide a threshold optimization device for vehicle anti-pinch scenarios, applied to a cloud server. The device includes: The category determination module is used to determine the current anti-pinch category information of the vehicle based on the acquired anti-pinch scenario classification score. The anti-pinch scenario classification score is determined by analyzing the fault phenomena and causes of the fault phenomena in the motor's operating status data in the vehicle. The motor's operating status data is obtained by collecting data on the motor's operating status when the anti-pinch event is triggered at the target clamping position. The fault phenomenon types include at least one of voltage-related, current-related, anti-pinch position-related, signal quality-related, and user operation abnormality-related. The fault cause types include at least one of mechanical error-related, signal error-related, and sensor error-related. The anti-pinch category information includes correct anti-pinch and false anti-pinch. The threshold determination module is used to determine the target anti-pinch judgment threshold corresponding to the target clamping position based on the motor operating status data sent by the vehicle when the anti-pinch category information is false anti-pinch. The threshold sending module is used to send the target anti-pinch judgment threshold to the vehicle terminal; the vehicle terminal is used to perform vehicle anti-pinch judgment based on the target anti-pinch judgment threshold.

[0023] Fourthly, embodiments of this application provide a threshold optimization device for vehicle anti-pinch scenarios, applied to the vehicle end, the device comprising: The data acquisition module is used to collect data on the operating status of the drive motor that triggered the anti-pinch event when the anti-pinch event is triggered at the target clamping position on the vehicle side, and obtain the operating status data of the motor. The analysis and determination module is used to analyze the fault phenomena and causes of the motor's operating status data in the vehicle, and determine the scenario classification score of the current anti-pinch scenario in the vehicle. The fault phenomenon type includes at least one of voltage-related, current-related, anti-pinch position-related, signal quality-related, and user operation abnormality-related. The cause type includes at least one of mechanical error-related, signal error-related, and sensor error-related. The scoring sending module is used to send scenario classification scores to the cloud server; The execution module is used to perform vehicle anti-pinch judgment based on the target anti-pinch judgment threshold sent by the cloud server; the cloud server is used to determine the target anti-pinch judgment threshold based on the anti-pinch scenario classification score and the motor's operating status data.

[0024] Fifthly, embodiments of this application provide an electronic device, including a processor and a memory, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the threshold optimization method for vehicle anti-pinch scenarios as described above.

[0025] Sixthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a vehicle controller, implements the aforementioned threshold optimization method for vehicle anti-pinch scenarios.

[0026] In a seventh aspect, embodiments of this application provide a computer program product, including a computer program or instructions, which, when executed by a vehicle controller in a vehicle, implement the aforementioned threshold optimization method for vehicle anti-pinch scenarios.

[0027] The beneficial effects of this application are: In the embodiments of this application, the current anti-pinch category information of the vehicle is determined based on the acquired anti-pinch scenario classification score; the anti-pinch scenario classification score is determined by analyzing the fault phenomena and causes of the motor's operating status data in the vehicle; the motor's operating status data is obtained by collecting data on the motor's operating status when the anti-pinch event is triggered at the target clamping position; the fault phenomenon types include at least one of voltage type, current type, anti-pinch position type, signal quality type, and user operation abnormality type; the cause types include at least one of mechanical error type, signal error type, and sensor error type; the anti-pinch category information includes correct anti-pinch and false anti-pinch; when the anti-pinch category information is false anti-pinch, the target anti-pinch judgment threshold corresponding to the target clamping position is determined based on the motor's operating status data sent by the vehicle; the target anti-pinch judgment threshold is sent to the vehicle; wherein, the vehicle is used to determine the target anti-pinch judgment threshold based on the target anti-pinch judgment. By setting a threshold and performing vehicle anti-pinch judgment, the cloud server can determine whether the most recent anti-pinch event on the vehicle is a false alarm or a normal anti-pinch event based on the anti-pinch scenario classification score generated after the anti-pinch event occurs on the vehicle. When the most recent anti-pinch event on the vehicle is identified as a false alarm, the anti-pinch judgment threshold is adjusted based on the motor operating status data generated by the event. The adjusted target anti-pinch judgment threshold is then sent to the vehicle. The vehicle can use the target anti-pinch judgment threshold for subsequent anti-pinch judgments. Through anti-pinch scenario classification scoring, the anti-pinch events of the vehicle can be quantified, thereby clarifying the boundary between correct and false anti-pinch, and improving the accuracy of correct and false anti-pinch judgments. Since false anti-pinch events usually occur when the anti-pinch judgment threshold is too low, sending the target anti-pinch judgment threshold through the cloud server can improve the accuracy of subsequent anti-pinch judgments on the vehicle to a certain extent, effectively reducing the false anti-pinch rate. Attached Figure Description

[0028] Figure 1 A flowchart illustrating a threshold optimization method for vehicle anti-pinch scenarios provided in this application embodiment; Figure 2 A flowchart illustrating another threshold optimization method for vehicle anti-pinch scenarios provided in this application embodiment; Figure 3 A system architecture diagram of a threshold optimization method for vehicle anti-pinch scenarios provided in this application embodiment; Figure 4 A schematic diagram of a collaborative closed-loop process between the cloud and the vehicle is provided for an embodiment of this application; Figure 5 A data upload flow diagram provided in an embodiment of this application; Figure 6 This application provides a first-order and second-order difference curve diagram of a current sequence. Figure 7 A window lifting resistance curve provided in an embodiment of this application; Figure 8 A resistance threshold curve provided for embodiments of this application; Figure 9 A logic block diagram of a threshold optimization device for vehicle anti-pinch scenarios applied to a cloud server, provided in an embodiment of this application; Figure 10 This application provides a threshold optimization device for vehicle anti-pinch scenarios, applied to the vehicle end. Figure 11 This is a schematic diagram of the hardware entity of an electronic device provided in an embodiment of this application. Detailed Implementation

[0029] The embodiments of this application will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be understood that the preferred embodiments are only for illustrating this application and are not intended to limit the scope of protection of this application.

[0030] Current vehicles commonly employ designs such as power windows, power hood switches, and power trunk lid switches. Due to vehicle safety regulations, these power switches must be equipped with corresponding anti-pinch measures to ensure user safety. Traditional solutions monitor changes in the drive motor's current. For example, with power windows, when the glass encounters an obstacle, the resistance increases sharply, requiring the motor to output greater torque to overcome the resistance. This is directly reflected in a significant and rapid increase in operating current. Using this characteristic, the system can effectively detect a pinching event through abnormal current surges. However, current anti-pinch algorithms often experience false alarms, triggering the anti-pinch function when no object is being held, causing windows to descent abnormally and the hood and trunk lid to fail to close properly. The current anti-pinch algorithms have a high false alarm rate, resulting in a poor user experience.

[0031] To address the aforementioned issues, this application provides a threshold optimization method for vehicle anti-pinch scenarios, applied to a cloud server. Figure 1 This application provides a flowchart illustrating a threshold optimization method for vehicle anti-pinch scenarios, as illustrated in the embodiments of this application. Figure 1 As shown, the method includes the following steps S101 to S103: Step S101: Based on the acquired anti-pinch scenario classification score, determine the current anti-pinch category information of the vehicle. The anti-pinch scenario classification score is determined by analyzing the fault phenomena and causes of the motor's operating status data in the vehicle. The motor's operating status data is obtained by collecting data on the motor's operating status when the anti-pinch event is triggered at the target clamping position. The fault phenomena include at least one of the following: voltage, current, anti-pinch position, signal quality, and user operation anomaly. The causes of the faults include at least one of the following: mechanical error, signal error, and sensor error. The anti-pinch category information includes correct anti-pinch and false anti-pinch.

[0032] In the embodiments of this application, the anti-pinch determination method on the vehicle side can be: when the drive current of the drive motor at a certain clamping position is greater than the current threshold corresponding to the clamping position, or the motor speed is less than the motor speed corresponding to the clamping position, it is determined that anti-pinch needs to be performed, and the vehicle side can immediately input a reverse drive current to drive the drive motor in reverse.

[0033] In the embodiments of this application, an anti-pinch scenario classification score can be generated each time an anti-pinch event is triggered on the vehicle side. The anti-pinch scenario classification score can characterize the abnormality of the anti-pinch event on the vehicle side. For example, when the vehicle side experiences faults such as position calculation errors, mechanical failures, foreign object adhesion, or temperature changes, the probability of the vehicle erroneously triggering an anti-pinch event increases. When an anti-pinch event occurs on the vehicle side, the vehicle side can identify and statistically analyze these fault causes, thereby generating a corresponding anti-pinch scenario classification score.

[0034] The specific steps for generating anti-pinch scene classification scores may include: 1) By analyzing the operating status data, determine the fault phenomenon and cause of the fault.

[0035] The various fault phenomena and their causes are shown in Table 1 below: Table 1

[0036] In Table 1, "mechanical error" and "signal error" are fault causes that can be automatically eliminated by the vehicle, while "sensor error" is a fault cause that cannot be automatically eliminated by the vehicle and requires intervention from maintenance personnel.

[0037] Among the fault symptoms, "current-related" can include abnormal current slope and prolonged operation; "voltage-related" can include abnormal voltage; "anti-pinch position-related" can include anti-pinch being close to the top, large differences in total stroke, repeated positions, and instantaneous startup; "signal quality-related" can include poor current ripple, signal glitches, and sudden changes in resistance; and "user operation abnormality-related" can include user annotations, assembly defects, and jogging operations.

[0038] Among them, "large difference in total stroke" is obtained by comparing the motor rotation count with the total stroke; "repeated position" is obtained by comparing the current sensor signal with the sensor signal of the historical anti-pinch trigger point; "abnormal current slope", "current ripple difference" and "signal glitch" are obtained by performing frequency domain analysis on the current signal and / or sensor signal; "starting moment" refers to the moment when the window (or hood, trunk lid) is opened and closed, triggering the stop; "long-term operation" refers to the anti-pinch triggering after the motor has been running continuously for a period of time (motor thermal attenuation), which is obtained by detecting the frequency of the window control signal (for example, if the frequency attenuation exceeds the set value, it is considered to trigger long-term operation).

[0039] 2) By analyzing the fault phenomena and causes, and based on their correspondence in Table 1 above, the corresponding abnormal scores are queried and accumulated to obtain the anti-pinch scenario classification score.

[0040] For example, if the fault phenomena include abnormal current slope and poor current ripple, and the fault causes include mechanical faults and dynamic load interference, then the anti-pinch scenario classification score is 1+2+1+1=5.

[0041] An anti-pinch event refers to an event triggered by the vehicle. For example, if a window stops rising and falls a short distance without any user input during the window's upward movement, this constitutes an anti-pinch event.

[0042] Because vehicle-side computing resources are limited, after generating an anti-pinch scenario classification score, the vehicle-side can send this score to a cloud server. The cloud server then uses this score to classify the current anti-pinch event on the vehicle-side, obtaining the anti-pinch category information. The cloud server can be a cloud computing center or a cloud server, used to receive data from the vehicle-side and perform the corresponding calculations.

[0043] In addition, the vehicle can directly upload the motor's operating status data to the cloud server, which will then calculate the anti-pinch scenario classification score. This will further save the vehicle's computing resources and ensure that the vehicle's computing resources can be used for control tasks in other projects (such as intelligent driving, multimedia control, etc.).

[0044] In some embodiments, the target clamping position refers to the position where the anti-pinch event is triggered when the window is raised and the hood or trunk lid is closed. In this embodiment, the position where the window is at its lowest point and the hood and trunk lid are fully open is defined as 0%, while the position where the window is fully raised and the hood and trunk lid are fully closed is defined as 100%. This allows the target clamping position to be expressed as a percentage. For example, when the target clamping position of the window is 70%, it indicates that the window has risen 70%.

[0045] In some embodiments, the motor's operating status data may include at least the drive current data of the driving motor, and may also include data such as motor speed, target clamping position, ambient temperature, and motor output torque. Since different target clamping positions may result in different drive currents, the drive current data can be represented in the form of a drive current sequence. This drive current sequence can represent the relationship between the drive current and the target clamping position.

[0046] Data acquisition refers to the process by which the vehicle stores data collected by multiple sensors before and after the anti-pinch event is triggered into a designated cache location on the vehicle, so that the vehicle can quickly read or transmit this data.

[0047] It's easy to understand that a correct anti-pinch event means that the vehicle actually needs to prevent pinching and triggers the anti-pinch event. For example, when the window is rolled up, if a passenger extends their arm out of the window, this is a correct anti-pinch event. A false anti-pinch event means that the vehicle does not need to prevent pinching and triggers the anti-pinch event incorrectly. For example, when the hood is closed, it may inexplicably retract to a fully open state.

[0048] Step S102: If the anti-pinch category information is false anti-pinch, determine the target anti-pinch judgment threshold corresponding to the target clamping position based on the motor operating status data sent by the vehicle.

[0049] In the embodiments of this application, when the anti-pinch category information is false anti-pinch, it indicates that the anti-pinch judgment threshold used in the anti-pinch judgment process of the vehicle under the current operating condition is too low. At this time, the target anti-pinch judgment threshold corresponding to the target clamping position can be determined according to the motor's operating status data and the corresponding rules. For example, if the motor's operating status data changes rapidly and locally manifests as extremely rapid changes similar to pulses, the anti-pinch judgment threshold in the current anti-pinch judgment process can be appropriately increased; the faster the change (i.e., the higher the rate of change), the greater the increase can be; in addition, to avoid the possibility that the vehicle cannot correctly trigger anti-pinch under the same conditions, a fixed increase can also be used to ensure its safety; after the increase, the target anti-pinch judgment threshold corresponding to the target clamping position can be obtained.

[0050] It's easy to understand that blindly increasing the anti-pinch detection threshold would greatly increase the likelihood of the vehicle failing to correctly trigger the anti-pinch mechanism. Therefore, a certain false pinch rate needs to be maintained in the vehicle's anti-pinch detection. Thus, when the motor's operating status data changes slowly or without sharp changes, the currently used anti-pinch detection threshold can be used as the target anti-pinch detection threshold.

[0051] Step S103: Send the target anti-pinch determination threshold to the vehicle terminal; wherein, the vehicle terminal is used to perform vehicle anti-pinch determination based on the target anti-pinch determination threshold.

[0052] In the embodiments of this application, the cloud server can send the determined target anti-pinch judgment threshold to the vehicle terminal regardless of whether the anti-pinch judgment threshold is adjusted. After receiving the target anti-pinch judgment threshold, the vehicle terminal can directly use the target anti-pinch judgment threshold in the subsequent anti-pinch judgment process, or it can first determine whether the target anti-pinch judgment threshold is the same as the currently used anti-pinch judgment threshold. If they are the same, the currently used anti-pinch judgment threshold will not be changed (in the execution code, this is specifically reflected in not assigning a value), and the vehicle anti-pinch judgment will continue to be performed using the same anti-pinch judgment threshold as the target anti-pinch judgment threshold.

[0053] Based on the embodiments disclosed in this application, the cloud server can determine whether the most recent anti-pinch event on the vehicle is a false anti-pinch or a normal anti-pinch event by generating an anti-pinch scene classification score after an anti-pinch event occurs on the vehicle. When the most recent anti-pinch event on the vehicle is identified as a false anti-pinch, the anti-pinch judgment threshold is adjusted according to the motor operating status data generated by the anti-pinch event, and then the adjusted target anti-pinch judgment threshold is sent to the vehicle. The vehicle can use the target anti-pinch judgment threshold for subsequent anti-pinch judgments. Through the anti-pinch scene classification score, the anti-pinch events of the vehicle can be quantified, thereby clarifying the boundary between correct and false anti-pinch, and improving the accuracy of the judgment between correct and false anti-pinch. Since false anti-pinch occurs when the anti-pinch judgment threshold is low, sending the target anti-pinch judgment threshold through the cloud server can improve the accuracy of subsequent anti-pinch judgments on the vehicle to a certain extent and effectively reduce the false anti-pinch rate.

[0054] In some embodiments, step S102 can be implemented by steps S201 and S202: Step S201: Determine the cause of the anti-pinch event.

[0055] In the embodiments of this application, the cause of the anti-pinch event refers to the fault cause that caused the false anti-pinch event, obtained through analysis of the operating status data.

[0056] The cause of the event can be analyzed by the vehicle and sent to the cloud server, or it can be determined by the cloud server based on the received operating status data.

[0057] Step S202: If the cause of the event includes faults of mechanical error type and / or signal error type, determine the first-order difference value and second-order difference value of the motor's operating status data, and determine the first-order difference threshold and second-order difference threshold corresponding to the target clamping position; wherein, faults of mechanical error type and / or signal error type represent faults that can be automatically excluded.

[0058] In the embodiments of this application, mechanical error and / or signal error fault causes represent fault causes that can be automatically eliminated, i.e., fault causes that are "compatiblely solvable" in Table 1 above. If the event cause includes mechanical error and / or signal error fault causes, it indicates that the cause of the false anti-pinch at the vehicle end can be resolved to a certain extent at the vehicle end. Therefore, a target anti-pinch determination threshold can be further determined for anti-pinch determination to eliminate or temporarily eliminate the event cause at the vehicle end.

[0059] In one possible implementation, if the cause of the event also includes a fault that requires "human intervention", an alarm message can be generated to prompt the user to perform vehicle repairs as soon as possible.

[0060] To quantify the rate of change of the motor's operating status data, a differential method can be used to determine the first and second difference values ​​of the motor's operating status data. In the specific calculation process, since the motor's operating status data is not continuous but discrete data determined by the sampling frequency, a discrete data calculation method can be used, as shown in Formula 1 below: (Formula 1); in, , and The data includes the operating status of the motors from three consecutive data collection points. , and The sampling positions (i.e., clamping positions) of the three consecutive sampling points mentioned above. The first difference value, It is the second-order difference value.

[0061] Alternatively, first-order and second-order difference thresholds can be calculated using historical operating status data or historical difference thresholds of the motor at the target clamping position. For example, if using historical operating status data of the motor, the average value of the historical operating status data of the motor at each clamping position can be calculated first, and the first-order difference value calculated based on this average value can be used as the first-order difference threshold. If using historical difference thresholds, the average value of historical first-order difference thresholds can be directly used as the first-order difference threshold. The calculation of the second-order difference threshold is similar.

[0062] In step S203, if the first-order difference value is greater than the first-order difference threshold and the second-order difference value is greater than the second-order difference threshold, the anti-pinch judgment threshold corresponding to the target clamping position is increased according to the preset rule to obtain the target anti-pinch judgment threshold.

[0063] In the embodiments of this application, if the first-order difference value is greater than the first-order difference threshold and the second-order difference value is greater than the second-order difference threshold, it indicates that the operating status data of the motor changes rapidly and drastic changes occur in a local position. At this time, the cloud server can consider that the vehicle terminal needs to adjust the anti-pinch judgment threshold corresponding to the target clamping position under this working condition, so that the anti-pinch judgment threshold corresponding to the target clamping position can be increased according to the preset rules to obtain the target anti-pinch judgment threshold.

[0064] The preset rule can be one of the following: 1) Increase the anti-pinch judgment threshold by a preset percentage; 2) Increase the anti-pinch judgment threshold by a preset value; 3) Determine the increase in the anti-pinch judgment threshold based on the ratio of the first-order difference value to the first-order difference threshold and the ratio of the second-order difference value to the second-order difference threshold, and increase the anti-pinch judgment threshold accordingly.

[0065] Based on the above embodiments disclosed in this application, the first-order difference value can be used to evaluate the rate of change of the motor's operating state data, and the second-order difference value can be used to evaluate the local abrupt change of the motor's operating state data. When both the first-order difference value and the second-order difference value meet the conditions, it indicates that the motor's operating state data changes rapidly and abruptly occurs. At this time, the anti-pinch judgment threshold corresponding to the target clamping position can be adjusted upward according to the preset rules, thereby improving the accuracy and reliability of the adjusted target anti-pinch judgment threshold to a certain extent.

[0066] In some embodiments, the motor's operating state data includes at least drive current sequence data; the drive current sequence data includes the correspondence between multiple drive positions and drive currents; the "determining the first-order difference value and the second-order difference value of the motor's operating state data" in step S202 can be achieved through step S211: Step S211: Based on multiple correspondences, determine the first-order difference value and the second-order difference value of the driving current relative to the driving position.

[0067] In the embodiments of this application, the correspondence between multiple driving positions and driving currents can be represented by points in a coordinate system. The following steps can then be performed: 1) Using curve fitting, connect multiple points corresponding to the above to obtain the relationship curve of the driving current relative to the driving position; 2) Using MATLAB, calculate the first and second derivatives of the above relationship curve at the target clamping position to obtain the first and second difference values.

[0068] Specifically, it is represented by Formula 2 as follows: (Formula 2); in, Represents the first-order difference value. Represents the second-order difference value. Indicates the drive current. Indicates the target clamping position.

[0069] Based on the above embodiments disclosed in this application, the correspondence between multiple driving positions and driving currents can form a curve of driving current changing with position. Since the accuracy of driving current measurement is higher, the first-order difference value and second-order difference value obtained through driving current sequence data can have higher accuracy. Therefore, the above method can improve the accuracy and reliability of the first-order difference value and second-order difference value to a certain extent.

[0070] In some embodiments, step S202, "determining the first-order difference threshold and the second-order difference threshold corresponding to the target clamping position," includes: Step S212: Determine the historical drive current sequence data of the target clamping position.

[0071] In the embodiments of this application, the cloud server can save the historical operating status data of the motor sent by the vehicle (including drive current sequence data); when it is determined that the anti-pinch event occurring on the vehicle is a false anti-pinch event, the cloud server can read the historical drive current sequence data of the target clamping position of the anti-pinch event.

[0072] The number of sequences included in the historical drive current sequence data can be preset, for example, to 7 or 10. Furthermore, considering that the reference value of drive current sequence data further removed from the current time decreases as vehicle usage time increases, the number of sequences included in the historical drive current sequence data can be reduced as the vehicle's usage year or month increases.

[0073] Step S213: Determine the mean of the first-order difference of the current, the standard deviation of the first-order difference of the current, the mean of the second-order difference of the current, and the standard deviation of the second-order difference of the current for the historical driving current sequence data.

[0074] In the embodiments of this application, the first-order difference mean of the current refers to the average of the first-order difference values ​​of the target clamping position and its nearby clamping points in the historical driving current sequence data; the first-order difference standard deviation of the current refers to the standard deviation of the first-order difference values ​​of the target clamping position and its nearby clamping points, as shown in Formula 3 below: (Formula 3); in, This represents the first-order difference value at position i. This represents the first-order difference mean of the current. This represents the first-order difference standard deviation of the current.

[0075] The calculation methods for the second-order difference mean and the second-order difference standard deviation of current are similar to those for the first-order difference mean and the first-order difference standard deviation of current. Please refer to the calculation process in Formula 3.

[0076] Step S214: Based on the mean of the first-order difference of the current and the standard deviation of the first-order difference of the current, determine the first-order difference threshold corresponding to the target clamping position; and based on the mean of the second-order difference of the current and the standard deviation of the second-order difference of the current, determine the second-order difference threshold corresponding to the target clamping position.

[0077] In the embodiments of this application, the first-order difference threshold corresponding to the target clamping position is calculated based on the mean of the first-order difference of the current and the standard deviation of the first-order difference of the current, as shown in Formula 4 below: (Formula 4); in, This represents the first-order difference threshold. It is an adjustable parameter. Since in this application's embodiments, there is typically only a case where the anti-pinch detection threshold is increased, therefore... It is usually set to a positive integer. This can be set using the confidence level theory of the normal distribution; for example, if set... If the confidence level is 95%, then If set The confidence level is 99.7%, then If the user sets a confidence level, then the determination can be based on the confidence level. The value is set to 0; otherwise, the default value of 3 is used.

[0078] Based on the embodiments disclosed in this application, the cloud server can calculate the mean and standard deviation of the first-order difference of the current based on historical driving current sequence data, and calculate the corresponding first-order difference threshold and second-order difference threshold based on the mean and standard deviation of the first-order difference of the current. Since the calculation multiples of the mean and standard deviation of the first-order difference of the current are directly determined by the preset confidence level in actual calculation, the above method can improve the accuracy of the first-order difference threshold and the second-order difference threshold to a certain extent, thereby improving the effectiveness of comparison between the first-order difference value and the first-order difference threshold, as well as between the second-order difference value and the second-order difference threshold, and thus fully ensuring the necessity and effectiveness of finally adjusting the anti-pinch judgment threshold.

[0079] In some embodiments, the step S202, "adjusting the anti-pinch judgment threshold corresponding to the target clamping position according to a preset rule to obtain the target anti-pinch judgment threshold", can be achieved through step S221: Step S221: Calculate the product of the anti-pinch judgment threshold corresponding to the target clamping position and the preset adjustment coefficient to obtain the target anti-pinch judgment threshold; the preset adjustment coefficient is greater than 1.

[0080] In the embodiments of this application, the preset adjustment coefficient needs to ensure that the adjusted target anti-pinch judgment threshold is greater than the anti-pinch judgment threshold before adjustment. Therefore, the preset adjustment coefficient needs to be greater than 1. In addition, the magnitude of the adjustment at one time should not be set too high to reduce the possibility that the vehicle cannot correctly trigger the anti-pinch event. Therefore, in addition to meeting the basic requirement of being greater than 1, the difference between the preset adjustment coefficient and 1 should usually be close to 0. For example, the preset adjustment coefficient can be set to values ​​such as 1.05, 1.08 or 1.10.

[0081] Based on the above embodiments disclosed in this application, the anti-pinch judgment threshold corresponding to the target clamping position can be increased by a preset multiplier, which can effectively control the increase of the anti-pinch judgment threshold and improve the accuracy of the target anti-pinch judgment threshold to a certain extent.

[0082] In some embodiments, step S101 includes at least one of the following: Step S301: If the anti-pinch scenario classification score sent by the vehicle terminal is greater than or equal to the preset score threshold, the current anti-pinch category information of the vehicle terminal is determined to be false anti-pinch.

[0083] In the embodiments of this application, as described in the above embodiments, the anti-pinch scenario classification score can characterize the degree of abnormality of the anti-pinch event on the vehicle side. Therefore, the higher the anti-pinch scenario classification score, the more likely the anti-pinch event is a false alarm. Therefore, when the anti-pinch scenario classification score sent by the vehicle side is greater than or equal to a preset score threshold, the anti-pinch event on the vehicle side can be considered a false alarm, and the current anti-pinch category information on the vehicle side is set as a false alarm.

[0084] The preset scoring threshold can be determined through measurement experiments. For different car models, since their windows, hoods and trunk lids usually have different physical parameters, and their drive structures may also be different, different car models can have different preset scoring thresholds.

[0085] Step S302: If the anti-pinch scenario classification score is less than the preset score threshold, determine the anti-pinch category information as correct anti-pinch.

[0086] In the embodiments of this application, if the anti-pinch scenario classification score is less than the preset score threshold, it can be considered that the fault phenomena contained in the current anti-pinch event are at a low level, and the cloud server can determine that the anti-pinch category information of the current anti-pinch event is correct anti-pinch.

[0087] Based on the above embodiments disclosed in this application, the relationship between the preset scoring threshold and the classification scores of the anti-pinch scenario can be used to determine whether the anti-pinch category information of the current anti-pinch event is a false anti-pinch or a correct anti-pinch. Since the preset scoring threshold is obtained through measurement experiments, it can improve the accuracy of the anti-pinch category information to a certain extent.

[0088] This application also provides another threshold optimization method for vehicle anti-pinch scenarios, applied to the vehicle side. Figure 2 This application provides a flowchart illustrating a threshold optimization method for vehicle anti-pinch scenarios, as illustrated in the embodiments of this application. Figure 2 As shown, the method includes the following steps S401 to S404: Step S401: When an anti-pinch event is triggered at the target clamping position on the vehicle side, data on the operating status of the drive motor that triggered the anti-pinch event is collected to obtain the motor's operating status data.

[0089] In the embodiments of this application, when the vehicle triggers an anti-pinch event, the drive current, motor speed and motor output torque of the corresponding drive motor will change. Therefore, the vehicle can collect these data through corresponding sensors and obtain the motor's operating status data such as drive current, motor speed and motor output torque collected by the sensors when the vehicle triggers an anti-pinch event.

[0090] Step S402: Analyze the fault phenomena and causes of the motor's operating status data, and determine the scenario classification score of the current anti-pinch scenario on the vehicle side; the fault phenomenon type includes at least one of voltage type, current type, anti-pinch position type, signal quality type, and user operation abnormality type; the cause type includes at least one of mechanical error type, signal error type, and sensor error type.

[0091] In embodiments of this application, the vehicle can also output a scene classification score for the current anti-pinch scenario based on the motor's operating status data using preset scoring rules or a lightweight artificial intelligence model. The preset scoring rules may include a scoring table, scoring items, and scoring formulas.

[0092] Step S403: Send the scene classification score to the cloud server.

[0093] In the embodiments of this application, after the vehicle receives the scene classification score, it can send the scene classification score to the cloud server so that the cloud server can perform subsequent classification and adjustment of the anti-pinch judgment threshold based on the scene classification score.

[0094] Step S404: Based on the target anti-pinch determination threshold sent by the cloud server, perform vehicle anti-pinch determination; The cloud server is used to determine the target anti-pinch judgment threshold based on the anti-pinch scenario classification score and the motor's operating status data.

[0095] In the embodiments of this application, after receiving the target anti-pinch determination threshold, the vehicle terminal can directly use the target anti-pinch determination threshold in the subsequent anti-pinch determination process, or it can first determine whether the target anti-pinch determination threshold is the same as the currently used anti-pinch determination threshold. If they are the same, the currently used anti-pinch determination threshold will not be changed (in the execution code, this is specifically reflected in not assigning a value), and the currently used anti-pinch determination threshold that is the same as the target anti-pinch determination threshold will continue to be used to perform vehicle anti-pinch determination.

[0096] Based on the embodiments disclosed in this application, the vehicle can generate operating status data of the motor at the target clamping position when an anti-pinch event is triggered. In some embodiments, the scenario classification score of the current anti-pinch scenario can be determined. Then, the computing resources of the cloud server are used to obtain the corresponding target anti-pinch judgment threshold to perform subsequent vehicle anti-pinch judgment. Through the anti-pinch scenario classification score, the vehicle's anti-pinch event can be quantified, thereby clarifying the boundary between correct anti-pinch and false anti-pinch, and improving the accuracy of the judgment between correct and false anti-pinch. Since false anti-pinch occurs when the anti-pinch judgment threshold is low, the target anti-pinch judgment threshold can be issued by the cloud server to a certain extent, which can improve the accuracy of the subsequent anti-pinch judgment of the vehicle and effectively reduce the false anti-pinch rate.

[0097] In some embodiments, the motor operating status data includes at least drive current sequence data, motor speed, and ambient temperature; step S402 includes: Step S501: Based on the drive current sequence data, motor speed and ambient temperature, determine the target fault phenomenon and the target fault cause at the vehicle end.

[0098] In the embodiments of this application, the target fault phenomenon may be one or more of the following: voltage type, current type, anti-pinch position type, signal quality type, and user operation abnormality type; the target fault cause may be one or more of the following: mechanical error type, signal error type, and sensor error type.

[0099] Step S502: Look up the table to obtain the weighted score of the mapping formed by each target fault phenomenon and each target fault cause; In the embodiments of this application, the scoring table used for "table lookup" is shown in Table 1 above. If a corresponding fault occurs in the current anti-pinch scenario, the score corresponding to that fault is added to the original score; the base score is 0. The fault phenomenon represents the correspondence between the fault phenomenon and its cause. The causes of fault include nine types: stroke calculation error, mechanical failure, foreign object adhesion, low-temperature false anti-pinch, power supply abnormality, sampling interference, sensor drift / noise, dynamic load interference, and poor assembly / maintenance (i.e., "Causes of Fault" in Table 1).

[0100] Step S503: Accumulate the weighted scores to obtain the scene classification score of the current anti-pinch scene.

[0101] In the embodiments of this application, the scene classification score of the current anti-pinch scenario can be obtained by directly accumulating the above weight scores.

[0102] In some possible embodiments, the scene classification score for the current anti-pinch scenario can also be calculated by weighted summation based on the weighted scores. Taking four fault causes—stroke calculation error, mechanical failure, foreign object adhesion, and low-temperature false anti-pinch—as examples, the calculation method is shown in Formula 5 below: (Formula 5); in, An important factor representing the cause of failure i; Weighted score for the mapping between fault cause i and fault phenomenon j; Classify and score the scenes.

[0103] The overall application process of scene classification and scoring is as follows: After calculating the scene classification score, the vehicle sends the score to the cloud server. The cloud server determines whether the current anti-pinch event is a normal or false alarm based on the relationship between the scene classification score and the anti-pinch judgment threshold. If it's a false alarm, the cloud server calculates the latest anti-pinch judgment threshold and sends it to the vehicle. The vehicle then applies this threshold to its anti-pinch judgment. Simultaneously, the cloud server can also receive the scene classification score calculation index (target fault phenomenon and target fault cause) from the vehicle. The cloud server generates a false alarm error report and sends it to the vehicle, or the vehicle can generate its own false alarm error report containing the target fault phenomenon and target fault cause. Furthermore, if the current anti-pinch event is a normal one, no further processing is performed; the current process ends, and the system waits for the next anti-pinch event to trigger.

[0104] Based on the embodiments disclosed in this application, the vehicle can determine the weight score corresponding to each fault phenomenon through drive current sequence data, motor speed and ambient temperature, and then calculate the corresponding scene classification score based on the weight score, which can improve the data richness of the motor's operating status data. Since the weight score can effectively reflect the probability of occurrence of each fault phenomenon, the scene classification score can effectively characterize the actual situation of the current anti-pinch scenario and improve the accuracy of the scene classification score to a certain extent.

[0105] In some embodiments, the threshold optimization method for vehicle anti-pinch scenarios described above further includes: Step S405: Obtain the anti-pinch feedback information input by the user.

[0106] In the embodiments of this application, the vehicle terminal can also generate and display an anti-pinch feedback control to the user, allowing the user to select whether the current anti-pinch event is a correct anti-pinch or a false anti-pinch. In this way, the vehicle terminal can obtain the anti-pinch feedback information input by the user.

[0107] Step S406: If the anti-pinch feedback information indicates that the anti-pinch category information is normal anti-pinch, the judgment threshold used in the vehicle anti-pinch determination is changed from the target anti-pinch determination threshold to the most recent anti-pinch determination threshold.

[0108] In the embodiments of this application, if the anti-pinch feedback information indicates that the anti-pinch category information is normal anti-pinch, it indicates that the classification result of the cloud server is incorrect. At this time, the judgment threshold used in the vehicle anti-pinch determination can be reverted from the target anti-pinch determination threshold to the most recent anti-pinch determination threshold on the vehicle side to avoid using an unreasonable target anti-pinch determination threshold.

[0109] Based on the above embodiments disclosed in this application, the vehicle terminal can also accept user feedback information, and when the user's anti-pinch feedback information indicates that the anti-pinch category information is normal anti-pinch, the target anti-pinch judgment threshold issued by the cloud server is rolled back. Through the user's active feedback, the problem of abnormal upward adjustment of the anti-pinch judgment threshold caused by the cloud server's misjudgment can be reduced to a certain extent, and the accuracy and reliability of the vehicle terminal's anti-pinch judgment can be improved to a certain extent.

[0110] The following describes the application of the threshold optimization method for vehicle anti-pinch scenarios provided in the embodiments of this application in real-world scenarios.

[0111] Traditional car window systems, especially power windows, while offering convenience, also pose potential safety risks. If a window encounters an unexpected obstacle (such as a passenger's limb, a child's finger, or other objects) while it is closing, and the system fails to detect and stop it in time, it could cause pinching injuries or damage to the object. Such safety issues have received widespread attention in automotive design, especially given increasingly stringent regulations concerning child occupant safety, and have become a critical technical problem that must be solved.

[0112] To address this challenge, anti-pinch windows were developed. Their core objective is to enhance the safety of the window system. Through technological means, they detect abnormal resistance in real time during the window's upward movement. Upon detecting a pinching event, they immediately stop the window's upward movement and execute an appropriate retraction action to release the pinch and prevent injury.

[0113] The key to achieving this safety function lies in the accurate perception and intelligent judgment of the operating status of the window drive system. The main technical approaches focus on monitoring the operating characteristics of the drive motor and sensing the contact pressure in the window sealing area. A commonly used technique is to monitor changes in the drive motor's current. During normal upward movement, the motor load (overcoming the weight of the glass, guide rail friction, etc.) is relatively stable, and the corresponding operating current exhibits specific characteristics. Once the glass encounters an obstacle during upward movement, the resistance increases sharply, and the motor needs to output greater torque to try to overcome the resistance, which is directly reflected in a significant and rapid increase in the operating current. By acquiring the motor current signal in real time using a high-sensitivity current sensor and comparing it with a preset normal current threshold or rate of change model, the system can effectively identify this abnormal current surge, thereby determining the occurrence of a clamping event.

[0114] However, the current technical solution has the following problems: 1. Static threshold defect: Traditional anti-pinch algorithms rely on fixed thresholds (current / torque), which cannot adapt to climate differences (such as increased resistance at low temperatures), mechanical wear, and changes in user habits.

[0115] The conflict between sensitivity and safety: a loose threshold leads to the risk of missed clipping, while a strict threshold causes false triggering (such as frequent window retraction).

[0116] 2. Lack of closed-loop optimization: It cannot distinguish between genuine clamping and accidental triggering, and there is no user feedback mechanism.

[0117] 3. Hardware failure latency: Mechanical defects (such as sudden changes in resistance at specific locations due to gear jamming in a reducer) cannot be identified in advance. Risks can only be identified through durability testing during the R&D process, and there is a certain probability of missing them. If hardware failures enter the market, there may be a recall risk.

[0118] 4. Calibration costs are huge. To achieve high precision and functional stability in the anti-pinch function, extensive high and low temperature durability and fault injection tests need to be conducted on the windows of each vehicle model in the early stages of the project to simulate the physical performance degradation and structural aging of the windows under long-term use. Furthermore, the calibration samples may not be representative.

[0119] To address the above issues, this application proposes a threshold optimization system for vehicle anti-pinch scenarios. In this system, the vehicle-side monitors the motor current, speed, and environmental parameters during window operation in real time. When a sudden change in resistance is detected, it is reported to the cloud. The cloud platform establishes a statistical and classification model based on large-scale vehicle operation data, and dynamically adjusts the anti-pinch threshold and compensation strategy by combining user feedback and active classification conditions. Compared with existing solutions, this invention emphasizes an active classification mechanism, which can accurately distinguish between genuine anti-pinch measures and false anti-pinch measures.

[0120] Its system architecture is as follows Figure 3 As shown. Figure 3 In this system, the vehicle-side (301), cloud-side (302), and OEM (Original Equipment Manufacturer) (303) collaborate with each other, using the user as the connecting link. The vehicle-side (301) includes an anti-pinch algorithm module (3011), a vehicle infotainment system (3012), multi-source sensors (3013), and a TBOX (3014). The cloud-side (302) includes an adaptive algorithm module (3021), an event logging module (3022), and a big data analysis module (3023). The vehicle-side (301) and cloud-side (302) interact through their respective modules to optimize the anti-pinch detection threshold. Simultaneously, the OEM (Original Equipment Manufacturer) (303) utilizes the anti-pinch event sentiment analysis module (3031), hardware performance evaluation module (3032), and after-sales warning module (3033) to perform post-installation maintenance of the anti-pinch detection threshold and issue anti-pinch warnings to users at appropriate times to remind them to perform vehicle maintenance.

[0121] Figure 4 This is a schematic diagram of the collaborative closed-loop process between the cloud and the vehicle provided in this application embodiment. The vehicle terminal 401 and the cloud terminal 402 jointly complete the closed-loop process from steps S41 to S49. The vehicle terminal 401 is mainly responsible for executing steps S41 to S45, including executing the anti-pinch algorithm, transmitting motor operating status data before and after the anti-pinch event to the cloud terminal 402, and receiving updated anti-pinch judgment threshold data from the OTA (Over-The-Air) system. The cloud terminal is mainly responsible for executing steps S46 to S49, including performing data analysis, obtaining updated anti-pinch judgment threshold data, and sending the updated anti-pinch judgment threshold data to the vehicle terminal 401 via OTA.

[0122] The threshold optimization process provided in this application mainly includes the following steps: The main steps are as follows: 1. Vehicle-side monitoring: During vehicle operation, the vehicle-side monitors motor current, speed, displacement, and environmental parameters in real time. If a deviation from the normal value is detected, the anti-pinch event judgment process is initiated.

[0123] 2. Anti-pinch event determination and execution: The vehicle-side algorithm makes a comprehensive judgment on the detection results. If an anti-pinch event is confirmed, the protective action (such as stopping or reversing) is executed immediately, and an event record is generated at the same time.

[0124] 3. Data Acquisition and Upload: The vehicle collects multi-dimensional data, including the current waveform at the time of the event, window position, ambient temperature and humidity, and user operation information, and uploads it to the cloud through the communication module.

[0125] 4. Anti-pinch event classification: The internal event classification model is used to distinguish between different types of accidental pinch prevention. 5. Cloud-based analysis and optimization: The cloud-based algorithm module receives operating status data of motors from a large number of vehicles, performs batch statistical analysis, identifies common features and environmental dependencies, and generates optimized parameter sets for different vehicle models and environmental conditions.

[0126] 6. Parameter distribution and update: The cloud sends the optimized parameters to the vehicle end via encrypted communication, and the vehicle end replaces or adjusts local anti-pinch threshold, sensitivity coefficient and other control parameters.

[0127] The detailed process is as follows: 0. Startup Phase When the vehicle enters the operating state, the vehicle-side system begins to perform anti-pinch monitoring and data acquisition tasks.

[0128] 1. Anomaly detection; The vehicle-side system monitors operating parameters such as motor, current, and speed in real time. If a deviation from normal operating conditions is detected, a potential anomaly is identified, and the anti-pinch event determination process is initiated.

[0129] 2. Anti-pinch event determination; The system determines whether an anti-pinch event has actually occurred: if not, it continues normal operation; if an anti-pinch event is confirmed, it executes protective actions and uploads relevant information.

[0130] 3. Data collection and uploading; After the anti-pinch event is triggered, the vehicle automatically collects multi-dimensional operating data (such as motor current curve, window displacement, ambient temperature, user operation, etc.) and uploads it to the cloud through the communication module.

[0131] 4. Classification of anti-pinch events; An internal event classification model is used to differentiate between various causes of accidental clamping, such as position calculation errors, mechanical malfunctions, foreign objects causing sticking points, and low-temperature accidental clamping. Weighting is applied according to Table 1. Source data is collected locally from the vehicle, and then the scores for each category are uploaded to the cloud.

[0132] The vehicle-side controller collects multi-source signals in real time, including motor current, Hall effect position, ambient temperature, and power supply voltage. The event classification model employs a weighted scoring strategy, mapping multi-dimensional features to a comprehensive category score for determining the type of anti-pinch event.

[0133] Set a fault determination threshold. When the overall category score is greater than the fault determination threshold, it is determined to be a mechanical fault or an environmental anomaly caused by false anti-pinch that requires special attention; otherwise, it is determined to be "real anti-pinch" or other negligible interference.

[0134] After this step is completed, relevant information on "abnormal anti-pinch" can be obtained. The data can then be packaged into CAN format, uploaded to the cloud via the vehicle gateway TBOX, and recorded. Later, the data can be replayed according to the protocol definition when viewing it. An example of the uploaded data is shown in Table 2 below: Table 2

[0135] Figure 5 This application provides a data upload flow diagram. After collecting the corresponding anti-pinch event data, the motor controller 501 can send the data to the TBOX 503 via the bus network management system 502. The TBOX 503 then transmits the anti-pinch event data to the cloud-based automotive remote service provider TSP 504 via a 4G network.

[0136] 5. Cloud-based batch data analysis; The cloud-based algorithm module receives anti-pinch related data from different vehicles and performs large-scale statistical and pattern analysis. The cloud can uncover common features, environmental influencing factors, and shortcomings in vehicle-side calibration.

[0137] 6. Parameter optimization and generation; Based on big data analysis results, the cloud generates a more adaptive set of optimized parameters, such as new anti-pinch current thresholds, dynamic sensitivity coefficients based on working conditions (e.g., 1.5 times greater force when the door is open, 2 times greater force when there is a mechanical structure failure, and reduced to 0.8 times in the most stable area), and compensation factors that take into account climate conditions.

[0138] 7. Parameter distribution and local updates; The cloud sends optimization parameters to the vehicle via a secure encrypted channel. The vehicle's algorithm module receives and replaces the local parameters, resulting in higher accuracy and robustness in subsequent operations.

[0139] 8. Algorithm iteration and closed-loop optimization; While collecting new data, the vehicle continuously collaborates with the cloud in a closed-loop process of "detection-upload-optimization-distribution" to continuously improve the intelligence level of anti-pinch detection.

[0140] 9. The process is complete; The anti-pinch process is now complete, and the vehicle is entering routine monitoring mode, awaiting the next event.

[0141] The core innovation of this application lies in proposing an intelligent anti-pinch system for vehicle windows. Its core lies in introducing a multi-dimensional analysis and feedback-driven active classification mechanism: the system comprehensively analyzes each anti-pinch event, including position repeatability detection (excluding mechanical jamming), environmental temperature correlation (low-temperature compensation to avoid false triggering), near-top position determination (identifying travel estimation errors), and current curve feature analysis (distinguishing between noise interference and actual clamping). Combined with user feedback on "actual anti-pinch / false anti-pinch," a closed-loop self-labeling mechanism is innovatively constructed, using user input as a supervisory signal to optimize the cloud model and parameters, while enhancing user peace of mind. The system adopts a cloud-edge collaborative architecture, with the vehicle side handling lightweight real-time detection and the cloud performing heavy-duty analysis and fault prediction. Furthermore, the system possesses algorithmic adaptive dynamic compensation capabilities, identifying and compensating for resistance mutation points, stagnation points, or travel calculation errors (such as ripple signal loss) caused by mechanical aging (e.g., guide rail wear, turbine damage) through historical data comparison. Finally, the system also provides scenario-based threshold configuration options, which can automatically increase anti-pinch sensitivity based on camera recognition of specific human features (e.g., children).

[0142] Specifically as follows: 1. Proactive Classification Mechanism: This mechanism comprehensively assesses the authenticity of anti-pinch events across multiple dimensions, including position repeatability, ambient temperature, top position, and current curve characteristics (slope, duration, curvature, and peak duration), significantly reducing false triggering rates. This mechanism integrates position repeatability (identifying mechanical sticking at the fixed point), ambient temperature correlation (compensating for low-temperature characteristic drift), end-of-stroke approximation judgment (correcting accumulated errors in stroke estimation), and current waveform time-domain feature analysis (distinguishing between noise and continuous clamping), among other scoring methods, to proactively identify and preliminarily classify anti-pinch events. This improves the confidence of uploaded data and reduces false triggering rates from the source. 2. Closed-loop self-labeling mechanism: Utilizes optional user feedback on the authenticity of anti-pinch events as a supervisory signal to solve the problem of high-quality training data labeling, continuously optimize cloud models and local parameters, and improve user experience (peace of mind).

[0143] 3. Cloud-Edge Collaborative Architecture: Rationally allocate computing load, with lightweight real-time detection on the vehicle side and complex analysis and long-term prediction in the cloud.

[0144] 4. Algorithm-adaptive dynamic compensation: Based on historical data, it intelligently identifies and compensates for the effects of mechanical aging (resistance mutation point, hysteresis point) or signal defects (stroke error caused by ripple loss), thereby improving the robustness of the system.

[0145] 5. Contextualized threshold configuration: Combined with environmental perception (such as camera recognition of children), the anti-pinch sensitivity is dynamically adjusted to achieve more accurate safety protection.

[0146] The technical effects of the technical solutions in this application are compared in Tables 3 and 4 below: Table 3

[0147] Table 4

[0148] User-side value: 1) Enhanced safety: After self-learning and adapting, the anti-pinch function can remain unchanged in some harsh working conditions; 2) Optimized experience: The number of lifting interruptions caused by accidental triggering is reduced.

[0149] Value for automakers: 1) After-sales costs: Reduce recalls due to mechanical failures by predicting hysteresis points; 2) Data monetization: Desensitized anti-pinch scenario data can be used to optimize subsystems such as steering column / electric tailgate. Establish an ISO safety certification framework for the vehicle-cloud collaborative anti-pinch system.

[0150] In the preceding process, the vehicle-mounted system collects the following signals in real time during the window raising and lowering process: the current position of the window; the motor current sequence; the motor speed; the hash (unique vehicle identifier); and the ambient temperature. After the above data is preprocessed locally, the vehicle-mounted system performs fusion / classification analysis on the motor's operating status data, calculates multi-dimensional factors and generates a comprehensive classification score, and then uploads the comprehensive classification score to the cloud.

[0151] The cloud receives a large amount of historical operating data from multiple vehicles and calculates the first and second differences of the current sequence for locations where there are sudden changes in resistance during window movement. Figure 6 As shown in the figure, the clamping position increases from 0 to 80%, with a sudden change occurring around 60%, and after 80%, the power to clamp the obstacle continuously increases.

[0152] The cloud-based system dynamically generates differential threshold tables or classification parameters based on statistical results and sends them to the vehicle. The vehicle then uses these updated parameters to perform rapid judgment and compensation control during subsequent operation.

[0153] The cloud platform can also determine the type of anti-pinch event based on the ratio of the distribution to the mean. If the standard deviation is small, it is likely a common problem for the vehicle model, while if the standard deviation is large, it is likely a problem for an individual vehicle. (In empirical terms, if the ratio of the mean to the standard deviation is less than 10%, the problem is very consistent, while if it is greater than 50%, it is unreliable.)

[0154] When the vehicle is actually running, if a matching scenario is detected, it will start interacting with the cloud to achieve dynamic threshold compensation.

[0155] Similarly, the window travel current parameter can be replaced with ambient temperature, number of times the window is raised or lowered, vehicle speed, thermal protection count, etc.

[0156] If a common problem is found, then threshold correction parameters will be generated by the cloud.

[0157] Figure 7 The window lifting resistance curve provided in the embodiments of this application is as follows: Figure 8 The resistance threshold curve provided for the embodiments of this application. Figure 7 The solid line in the middle shows a sudden change in resistance at the 60% position (simulating mechanical jamming), compared to the normal reference curve ( Figure 7 The line (dotted line) suddenly rises compared to the previous line. Figure 8 It is a first-order differential signal (equivalent to monitoring the rate of change of current). When a sudden change occurs, the differential signal increases sharply, exceeding the set threshold. Figure 8 When the value of the dashed line (mean + 3 standard deviations) is reached, a sudden change in resistance can be determined.

[0158] The following are examples of the application of the technical solution in this application: 1. Early warning of mechanical structure failures; Scenario: In a certain vehicle model project, multiple vehicles reported frequent anti-pinch issues at 70% of their positions; Execution process: Cloud clustering analysis → identification of common problems → automatic distribution of resistance compensation parameters + manual analysis of reducer gear jamming issues → push of after-sales testing suggestions.

[0159] Result: Mass recalls were avoided, and maintenance costs were reduced by approximately 40%.

[0160] 2. Ripple signal loss warning; Scenario: Multiple vehicles in a certain model project reported frequent anti-pinch issues at the 3% position point (top of the window); Execution process: Cloud-based cluster analysis → Identification of common problems → Identification of abnormal window travel, considering ripple signal loss → Issuance of travel compensation parameters + push of after-sales inspection suggestions. Result: Mass recalls were avoided, and maintenance costs were reduced by approximately 40%.

[0161] 3. Early warning of low temperature compensation failure; Scenario: Multiple vehicles experienced false alarms due to low temperatures; Execution process: Cloud clustering analysis → Identification of threshold compensation failure at low temperatures → Issuance of enhanced low temperature compensation parameters.

[0162] Result: Improved accuracy of vehicle anti-pinch detection.

[0163] 4. Anti-pinch function appears when normally clamping foreign objects; Scenario: A vehicle in a certain model project experiences an anti-pinch device at a random location; Cloud-based data collection and analysis → Identification as a non-common problem → No parameters will be sent to the vehicle that would affect the anti-pinch performance of other vehicles.

[0164] The beneficial effects of the technical solution in this application are as follows: 1. Group statistics and generalization capabilities: Single-vehicle terminals can only obtain limited samples of that vehicle, making it difficult to identify rare but important fault modes. The cloud can aggregate event data from a large number of vehicles under different models, geographical environments, and mileage, and perform cross-vehicle and cross-operating condition statistical analysis to extract more generalized fault features and compensation strategies.

[0165] 2. Self-updating capability of algorithm parameters: Referring to the algorithmic ideas of complex sequence models (such as LSTM / Transformer), the cloud provides the necessary computing resources for continuous iteration in offline / online stages.

[0166] 3. Adaptive parameter delivery and gray-scale verification: The cloud can generate differential parameter packages based on batch analysis (only transmitting the changes), and verify the reliability of new parameters through A / B zones and rollback mechanisms to reduce OTA risks.

[0167] 4. Long-term health management and strategy optimization: A vehicle group health database (including mileage, temperature, and maintenance records) is built in the cloud, which can be used for life prediction and offline optimization of the optimal threshold table.

[0168] 5. Security, Compliance and Audit: Centralized cloud management of digital signatures, version control and audit logs for parameter releases, improving the traceability and compliance of parameter updates.

[0169] like Figure 9 As shown, Figure 9 This application provides a logic block diagram of a threshold optimization device for vehicle anti-pinch scenarios applied to a cloud server. The device, 900, includes the following components: The category determination module 901 is used to determine the current anti-pinch category information of the vehicle based on the acquired anti-pinch scenario classification score. The anti-pinch scenario classification score is determined by analyzing the fault phenomena and causes of the fault phenomena in the motor's operating status data in the vehicle. The motor's operating status data is obtained by collecting data on the motor's operating status when the anti-pinch event is triggered at the target clamping position. The fault phenomenon types include at least one of voltage-related, current-related, anti-pinch position-related, signal quality-related, and user operation abnormality-related. The cause types include at least one of mechanical error-related, signal error-related, and sensor error-related. The anti-pinch category information includes correct anti-pinch and false anti-pinch. The threshold determination module 902 is used to determine the target anti-pinch judgment threshold corresponding to the target clamping position based on the motor operating status data sent by the vehicle terminal when the anti-pinch category information is false anti-pinch. The threshold sending module 903 is used to send the target anti-pinch judgment threshold to the vehicle terminal; wherein, the vehicle terminal is used to perform vehicle anti-pinch judgment based on the target anti-pinch judgment threshold.

[0170] In some embodiments, the threshold determination module 902 includes: The first acquisition submodule is used to determine the cause of the anti-pinch event; The first determining sub-block is used to determine the first-order and second-order difference values ​​of the motor's operating state data, and to determine the first-order and second-order difference thresholds corresponding to the target clamping position, when the cause of the event includes fault causes of mechanical error and / or signal error; wherein the fault causes of mechanical error and / or signal error represent fault causes that can be automatically excluded. The adjustment submodule is used to adjust the anti-pinch judgment threshold corresponding to the target clamping position according to preset rules when the first-order difference value is greater than the first-order difference threshold and the second-order difference value is greater than the second-order difference threshold, so as to obtain the target anti-pinch judgment threshold.

[0171] In some embodiments, the motor's operating state data includes at least drive current sequence data; the drive current sequence data includes a correspondence between multiple drive positions and drive currents; the first determining submodule includes: The first determining unit is used to determine the first-order difference value and the second-order difference value of the driving current relative to the driving position based on multiple correspondences.

[0172] In some embodiments, the first determining submodule includes: The second determining unit is used to determine the historical driving current sequence data of the target clamping position; The third determining unit is used to determine the first-order difference mean, first-order difference standard deviation, second-order difference mean, and second-order difference standard deviation of the historical driving current sequence data. The fourth determining unit is used to determine the first-order difference threshold corresponding to the target clamping position based on the mean of the first-order difference of the current and the standard deviation of the first-order difference of the current, and to determine the second-order difference threshold corresponding to the target clamping position based on the mean of the second-order difference of the current and the standard deviation of the second-order difference of the current.

[0173] In some embodiments, the up-adjustment submodule includes: The calculation unit is used to calculate the product of the anti-pinch judgment threshold corresponding to the target clamping position and the preset adjustment coefficient to obtain the target anti-pinch judgment threshold; the preset adjustment coefficient is greater than 1.

[0174] In some embodiments, the category determination module 901 includes at least one of the following: The second determination submodule is used to determine the current anti-pinch category information of the vehicle as false anti-pinch when the anti-pinch scenario classification score sent by the vehicle is greater than or equal to a preset score threshold. The third determination submodule is used to determine the anti-pinch category information as correct when the anti-pinch scene classification score is less than a preset score threshold.

[0175] like Figure 10 As shown, Figure 10 This application provides a threshold optimization device for vehicle anti-pinch scenarios, applied to a vehicle. The threshold optimization device 1000 for vehicle anti-pinch scenarios includes: The data acquisition module 1001 is used to collect data on the operating status of the drive motor that triggered the anti-pinch event when the anti-pinch event is triggered at the target clamping position on the vehicle side, and obtain the operating status data of the motor. The analysis and determination module 1002 is used to analyze the fault phenomena and causes of the motor's operating status data in the vehicle, and determine the scene classification score of the current anti-pinch scenario in the vehicle. The fault phenomenon type includes at least one of voltage-related, current-related, anti-pinch position-related, signal quality-related, and user operation abnormality-related types. The cause type includes at least one of mechanical error-related, signal error-related, and sensor error-related types. The scoring sending module 1003 is used to send the scene classification score to the cloud server; The execution module 1004 is used to perform vehicle anti-pinch determination based on the target anti-pinch determination threshold sent by the cloud server; the cloud server is used to determine the target anti-pinch determination threshold based on the anti-pinch scenario classification score and the motor's operating status data.

[0176] In some embodiments, the motor operating status data includes at least drive current sequence data, motor speed, and ambient temperature; the analysis and determination module 1002 includes: The fourth determination submodule is used to determine the target fault phenomenon and target fault cause at the vehicle end based on drive current sequence data, motor speed and ambient temperature; The table lookup submodule is used to look up tables to obtain the weighted score of the mapping formed by each target fault phenomenon and each target fault cause; The accumulation submodule is used to accumulate the weighted scores to obtain the scene classification score of the current anti-pinch scenario.

[0177] In some embodiments, the threshold optimization device 1000 applied to a vehicle for vehicle anti-pinch scenarios further includes: The second acquisition submodule is used to acquire anti-pinch feedback information input by the user; The rollback submodule is used to roll back the judgment threshold used in the vehicle anti-pinch determination from the target anti-pinch judgment threshold to the most recent anti-pinch judgment threshold when the anti-pinch feedback information indicates that the anti-pinch category information is normal anti-pinch.

[0178] The descriptions of the apparatus embodiments above are similar to those of the method embodiments above, and have similar beneficial effects. In some embodiments, the functions or modules included in the apparatus provided in this application can be used to perform the methods described in the method embodiments above. For technical details not disclosed in the apparatus embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0179] It should be noted that, in the embodiments of this application, if the above-described system monitoring method is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, or the part that contributes to the related technology, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, mobile hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware, software, or firmware, or any combination of hardware, software, and firmware.

[0180] Figure 11 This is a schematic diagram of the hardware entity of an electronic device provided in an embodiment of this application, such as... Figure 11As shown, the hardware entity of the electronic device 1100 includes a processor 1101 and a memory 1102, wherein the memory 1102 stores a computer program that can run on the processor 1101, and the processor 1101 executes the program to implement the steps in the method of any of the above embodiments.

[0181] The memory 1102 stores computer programs that can run on the processor. The memory 1102 is configured to store instructions and applications that can be executed by the processor 1101. It can also cache data to be processed or already processed (e.g., image data, audio data, voice communication data, and video communication data) in the processor 1101 and various modules in the electronic device 1100. It can be implemented by flash memory or random access memory (RAM).

[0182] When the processor 1101 executes the program, it implements the steps of the threshold optimization method for vehicle anti-pinch scenarios provided in any of the above embodiments. The processor 1101 typically controls the overall operation of the electronic device 1100.

[0183] The aforementioned processor can be at least one of the following: Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), Central Processing Unit (CPU), Controller, Microcontroller, and Microprocessor. It is understood that other electronic devices can also implement the functions of the aforementioned processor, and this application does not specifically limit the specific implementation.

[0184] The aforementioned computer storage media / memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM), etc.; or it can be various terminals that include one or any combination of the above-mentioned memories, such as mobile phones, computers, tablet devices, personal digital assistants, etc.

[0185] This application provides a computer-readable storage medium storing a computer program thereon. The computer-readable storage medium stores one or more programs, which can be executed by one or more processors. The computer program implements the threshold optimization method for vehicle anti-pinch scenarios as described above.

[0186] This application provides a computer program product, including a computer program or instructions. When the computer program or instructions are executed by a processor, they implement the threshold optimization method for vehicle anti-pinch scenarios as described above.

[0187] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, devices, or computer program products. Therefore, this application can take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0188] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable electronic device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable electronic device, generate instructions for implementing the process in the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0189] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable electronic device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0190] These computer program instructions may also be loaded onto a computer or other programmable electronic device, causing a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable device for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0191] 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 technical scope 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 threshold optimization method for vehicle anti-pinch scenarios, characterized in that, Applied to cloud server; the method includes: Based on the acquired anti-pinch scenario classification score, the current anti-pinch category information of the vehicle is determined. The anti-pinch scenario classification score is determined by analyzing the fault phenomena and causes of the motor's operating status data in the vehicle. The motor's operating status data is obtained by collecting data on the motor's operating status when an anti-pinch event is triggered at the target clamping position. The fault phenomenon types include at least one of voltage-related, current-related, anti-pinch position-related, signal quality-related, and user operation anomaly-related. The cause types include at least one of mechanical error-related, signal error-related, and sensor error-related. The anti-pinch category information includes correct anti-pinch and false anti-pinch. If the anti-pinch category information is false anti-pinch, the target anti-pinch judgment threshold corresponding to the target clamping position is determined based on the motor operating status data sent by the vehicle terminal. The target anti-pinch determination threshold is sent to the vehicle terminal; wherein, the vehicle terminal is used to perform vehicle anti-pinch determination based on the target anti-pinch determination threshold.

2. The method according to claim 1, characterized in that, The step of determining the target anti-pinch judgment threshold corresponding to the target clamping position based on the motor's operating status data sent by the vehicle includes: Determine the cause of the anti-pinch event; In cases where the cause of the event includes the fault cause of the mechanical error class and / or the signal error class, the first-order difference value and the second-order difference value of the motor's operating state data are determined, and the first-order difference threshold and the second-order difference threshold corresponding to the target clamping position are determined. If the first-order difference value is greater than the first-order difference threshold and the second-order difference value is greater than the second-order difference threshold, the anti-pinch judgment threshold corresponding to the target clamping position is increased according to a preset rule to obtain the target anti-pinch judgment threshold.

3. The method according to claim 2, characterized in that, The motor's operating status data includes at least drive current sequence data; the drive current sequence data includes multiple correspondences between drive positions and drive currents; The determination of the first-order and second-order difference values ​​of the motor's operating state data includes: Based on the multiple correspondences, the first-order difference value and the second-order difference value of the driving current relative to the driving position are determined respectively.

4. The method according to claim 3, characterized in that, Determining the first-order difference threshold and the second-order difference threshold corresponding to the target clamping position includes: Determine the historical drive current sequence data of the target clamping position; Determine the first-order difference mean, first-order difference standard deviation, second-order difference mean, and second-order difference standard deviation of the historical drive current sequence data. Based on the mean of the first-order difference of the current and the standard deviation of the first-order difference of the current, the first-order difference threshold corresponding to the target clamping position is determined, and based on the mean of the second-order difference of the current and the standard deviation of the second-order difference of the current, the second-order difference threshold corresponding to the target clamping position is determined.

5. The method according to claim 2, characterized in that, The step of increasing the anti-pinch determination threshold corresponding to the target clamping position according to a preset rule to obtain the target anti-pinch determination threshold includes: The anti-pinch judgment threshold corresponding to the target clamping position is calculated by multiplying it with a preset adjustment coefficient to obtain the target anti-pinch judgment threshold; the preset adjustment coefficient is greater than 1.

6. The method according to any one of claims 1 to 5, characterized in that, The determination of the current anti-pinch category information of the vehicle based on the anti-pinch scenario classification score sent by the vehicle includes at least one of the following: If the anti-pinch scenario classification score sent by the vehicle terminal is greater than or equal to a preset score threshold, the current anti-pinch category information of the vehicle terminal is determined to be false anti-pinch. If the anti-pinch scenario classification score is less than the preset score threshold, the anti-pinch category information is determined to be correct anti-pinch.

7. A threshold optimization method for vehicle anti-pinch scenarios, characterized in that, Applied to the vehicle end; the method includes: When an anti-pinch event is triggered at the target clamping position on the vehicle side, data on the operating status of the drive motor that triggered the anti-pinch event is collected to obtain the motor's operating status data. Analyze the fault phenomena and causes of the motor's operating status data in the vehicle to determine the scene classification score of the current anti-pinch scenario in the vehicle; the fault phenomenon type includes at least one of voltage-related, current-related, anti-pinch position-related, signal quality-related, and user operation abnormality-related; the cause type includes at least one of mechanical error-related, signal error-related, and sensor error-related. Send the scenario classification score to the cloud server; Based on the target anti-pinch judgment threshold sent by the cloud server, the vehicle anti-pinch judgment is performed; the cloud server is used to determine the target anti-pinch judgment threshold based on the anti-pinch scenario classification score and the motor's operating status data.

8. The method according to claim 7, characterized in that, The motor's operating status data includes at least drive current sequence data, motor speed, and ambient temperature; the determination of the scene classification score for the current anti-pinch scenario on the vehicle side based on the motor's operating status data includes: Based on the drive current sequence data, the motor speed, and the ambient temperature, the target fault phenomenon and the target fault cause at the vehicle end are determined; The weighted score of the mapping formed by each of the target fault phenomena and each of the target fault causes is obtained by looking up the table. The weighted scores are accumulated to obtain the scene classification score of the current anti-pinch scenario.

9. The method according to claim 7, characterized in that, The method further includes: Obtain anti-pinch feedback information input by the user; If the anti-pinch feedback information indicates that the anti-pinch category information is normal anti-pinch, the judgment threshold used in the vehicle anti-pinch determination is reverted from the target anti-pinch determination threshold to the most recent anti-pinch determination threshold.

10. A threshold optimization device for vehicle anti-pinch scenarios, characterized in that, Applied to cloud server; the device includes: The category determination module is used to determine the current anti-pinch category information of the vehicle based on the acquired anti-pinch scenario classification score. The anti-pinch scenario classification score is determined by analyzing the fault phenomena and causes of the motor's operating status data in the vehicle. The motor's operating status data is obtained by collecting data on the motor's operating status when an anti-pinch event is triggered at the target clamping position. The fault phenomenon types include at least one of voltage-related, current-related, anti-pinch position-related, signal quality-related, and user operation anomaly-related. The cause types include at least one of mechanical error-related, signal error-related, and sensor error-related. The anti-pinch category information includes correct anti-pinch and false anti-pinch. The threshold determination module is used to determine the target anti-pinch judgment threshold corresponding to the target clamping position based on the motor operating status data sent by the vehicle terminal when the anti-pinch category information is false anti-pinch. A threshold sending module is used to send the target anti-pinch determination threshold to the vehicle terminal; wherein, the vehicle terminal is used to perform vehicle anti-pinch determination based on the target anti-pinch determination threshold.

11. A threshold optimization device for vehicle anti-pinch scenarios, characterized in that, Applied to the vehicle end; the device includes: The data acquisition module is used to collect data on the operating status of the drive motor that triggered the anti-pinch event when the anti-pinch event is triggered at the target clamping position on the vehicle side, and obtain the operating status data of the motor. The analysis and determination module is used to analyze the fault phenomena and causes of the motor's operating status data in the vehicle, and to determine the scene classification score of the current anti-pinch scenario in the vehicle. The fault phenomenon type includes at least one of voltage-related, current-related, anti-pinch position-related, signal quality-related, and user operation abnormality-related. The cause type includes at least one of mechanical error-related, signal error-related, and sensor error-related. The scoring sending module is used to send the scene classification score to the cloud server; The execution module is used to perform vehicle anti-pinch determination based on the target anti-pinch determination threshold sent by the cloud server; the cloud server is used to determine the target anti-pinch determination threshold based on the anti-pinch scenario classification score and the motor's operating status data.

12. An electronic device, characterized in that, The device includes a processor and a memory, the memory storing a computer program that can run on the processor. When the processor executes the computer program, it implements the threshold optimization method for vehicle anti-pinch scenarios as described in any one of claims 1 to 6, or the threshold optimization method for vehicle anti-pinch scenarios as described in any one of claims 7 to 9.

Images