Method, apparatus, vehicle and readable storage medium for detecting hands-free status

By identifying vehicle scenarios and dynamically optimizing the steering wheel torque threshold, the problem of low accuracy in hands-free state detection in existing technologies has been solved, achieving higher detection accuracy and safety.

CN118270002BActive Publication Date: 2026-11-10BYD CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202410009299.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-01-02
Publication Date
2026-11-10
Estimated Expiration
2044-01-02

AI Technical Summary

Technical Problem

In existing technologies, using fixed preset upper and lower thresholds for hands-free state detection results in low detection accuracy and cannot cover changes in steering wheel torque values ​​under different scenarios.

Method used

By identifying the vehicle's scenario, the system obtains the hands-off alarm torque value based on the steering wheel torque value and a preset first steering wheel torque threshold. Then, it determines the second steering wheel torque threshold for the corresponding scenario based on the hands-off alarm torque value, dynamically optimizing the detection process to adapt to changes in different scenarios.

Benefits of technology

It improves the accuracy of hands-free state detection and alarm prompts, enhancing driving safety and driving experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118270002B_ABST
    Figure CN118270002B_ABST
Patent Text Reader

Abstract

This application discloses a method, device, vehicle, and readable storage medium for detecting hands-off state. The method includes: identifying a vehicle scene; obtaining a hands-off alarm torque value for the scene based on the steering wheel torque value and a preset first steering wheel torque threshold; determining a second steering wheel torque threshold corresponding to the scene based on the hands-off alarm torque value; and when the vehicle is identified as being in a target scene, determining whether the driver has taken their hands off the steering wheel based on the current steering wheel torque value and the second steering wheel torque threshold corresponding to the target scene. If so, a hands-off alarm is triggered. The technical solution disclosed in this application learns the corresponding steering wheel torque threshold based on the hands-off alarm torque values ​​for different scenes, and uses the learned steering wheel torque threshold for the corresponding scene to detect hands-off state when the vehicle is in the corresponding scene, thereby improving the accuracy of hands-off state detection.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of vehicle driving safety technology, and in particular to a method, device, vehicle, and readable storage medium for detecting hands-free status. Background Technology

[0002] During assisted driving, the vehicle will detect whether the driver is hands-free to ensure driving safety and driving experience.

[0003] Currently, a steering wheel torque sensor is typically used to detect the steering wheel torque value, which is then compared to preset upper and lower threshold values. If the steering wheel torque value falls within the threshold range, the hands-off state is considered genuine, and a hands-off warning is issued to alert the driver to take over the vehicle. However, using fixed preset upper and lower threshold values ​​for hands-off state detection results in relatively low detection accuracy.

[0004] In conclusion, improving the accuracy of hands-free state detection is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] In view of this, the purpose of this application is to provide a method, apparatus, vehicle, and readable storage medium for detecting hands-free status, so as to improve the accuracy of hands-free status detection.

[0006] To achieve the above objectives, this application provides the following technical solution:

[0007] A method for detecting hands-off state includes: identifying a vehicle scenario; obtaining a hands-off alarm torque value for the scenario based on a steering wheel torque value in the scenario and a preset first steering wheel torque threshold; the hands-off alarm torque value being the steering wheel torque value corresponding to a hands-off alarm; determining a second steering wheel torque threshold corresponding to the scenario based on the hands-off alarm torque value in the scenario; and when the vehicle is identified as currently in a target scenario, determining whether the driver has taken their hands off the steering wheel based on the current steering wheel torque value and the second steering wheel torque threshold corresponding to the target scenario, and if so, issuing a hands-off alarm.

[0008] Optionally, after triggering the hands-off alarm, the method further includes: obtaining the hands-off alarm torque value under the target scenario; determining a new steering wheel torque threshold corresponding to the target scenario based on the hands-off alarm torque value under the target scenario; and updating the second steering wheel torque threshold corresponding to the target scenario using the new steering wheel torque threshold corresponding to the target scenario.

[0009] Optionally, obtaining the hands-off alarm torque value in the scenario based on the steering wheel torque value in the scenario and a preset first steering wheel torque threshold includes: determining the hands-off alarm time based on the steering wheel torque value in the scenario and the preset first steering wheel torque threshold, and obtaining the steering wheel torque value for a first preset duration before the hands-off alarm and the steering wheel torque value for a second preset duration during the hands-off alarm.

[0010] Optionally, determining the second steering wheel torque threshold corresponding to the scenario based on the hands-off alarm torque value in the scenario includes: calculating the first mean and first standard deviation of the steering wheel torque value corresponding to the first preset duration before the hands-off alarm; calculating the second mean and second standard deviation of the steering wheel torque value corresponding to the first preset duration before the hands-off alarm and the steering wheel torque value corresponding to the second preset duration during the hands-off alarm; and determining the second steering wheel torque threshold corresponding to the scenario based on the first mean, the first standard deviation, the second mean, and the second standard deviation.

[0011] Optionally, identifying the vehicle's scenario includes: acquiring the steering wheel torque value and the vehicle's driving lane; using the steering wheel torque value to obtain the mean absolute value of the steering wheel torque value; determining whether the mean absolute value of the steering wheel torque value is greater than a first preset value and whether the vehicle's driving lane is a straight road; if the mean absolute value of the steering wheel torque value is greater than the first preset value and the vehicle's driving lane is a straight road, then determining that the scenario is a scenario where the steering wheel torque sensor is installed off-center in the vehicle; the steering wheel torque sensor is used to detect the steering wheel torque value.

[0012] Optionally, determining whether the vehicle driving lane is a straight road includes: obtaining the steering wheel angle and the radius of curvature of the vehicle driving lane; determining whether the steering wheel angle is within a preset angle range and whether the radius of curvature of the vehicle driving lane is greater than a second preset value; if so, then determining that the vehicle driving lane is a straight road.

[0013] Optionally, it further includes: when the scenario is determined to be a steering wheel torque sensor installation misalignment scenario in the vehicle, determining whether the driver is holding the steering wheel at the 6 o'clock position and determining whether to trigger a hands-off alarm based on the steering wheel torque threshold corresponding to the steering wheel torque sensor installation misalignment scenario in the vehicle; if so, determining that the scenario is a false alarm scenario, and before performing the step of obtaining the hands-off alarm torque value in the scenario based on the steering wheel torque value in the scenario and the preset first steering wheel torque threshold, replacing the preset first steering wheel torque threshold with the determined second steering wheel torque threshold corresponding to the steering wheel torque sensor installation misalignment scenario in the vehicle;

[0014] Once it is determined that the scenario is not a scenario where the steering wheel torque sensor is installed off in the vehicle, it is determined whether the driver is holding the steering wheel at the 6 o'clock position and whether the hands-off alarm is triggered based on the preset first steering wheel torque threshold. If so, the scenario is determined to be a false alarm scenario, and the step of obtaining the hands-off alarm torque value under the scenario based on the steering wheel torque value under the scenario and the preset first steering wheel torque threshold is executed.

[0015] Identify that the vehicle is currently in a target scenario, and determine whether the driver has taken his hands off the steering wheel based on the current torque value of the steering wheel and the steering wheel torque threshold corresponding to the target scenario. This includes: determining whether the driver is holding the steering wheel at the 6 o'clock position; if so, determining whether the driver has taken his hands off the steering wheel based on the current torque value of the steering wheel and the second steering wheel torque threshold corresponding to the false alarm scenario.

[0016] Optionally, determining whether the driver is holding the steering wheel at the 6 o'clock position includes: acquiring visual information about the position of the driver's hands on the steering wheel, and determining whether the driver is holding the steering wheel at the 6 o'clock position based on the visual information about the position.

[0017] Optionally, the system determines whether the driver has taken their hands off the steering wheel based on the current torque value of the steering wheel and the second steering wheel torque threshold corresponding to the target scenario. If so, a hands-off alarm is triggered, including: determining whether the current torque value of the steering wheel is within a threshold range determined by the second steering wheel torque threshold corresponding to the target scenario within a third preset time period; if the current torque value of the steering wheel is within the threshold range within the third preset time period, a first-level hands-off alarm is triggered, and it is determined whether the current torque value of the steering wheel is within the threshold range within a fourth preset time period; if the current torque value of the steering wheel is within the threshold range within the fourth preset time period, a second-level hands-off alarm is triggered, and it is determined whether the current torque value of the steering wheel is within the threshold range within a fifth preset time period; if the current torque value of the steering wheel is within the threshold range within the fifth preset time period, the driver exits the assisted driving mode and continues to trigger the second-level hands-off alarm.

[0018] Optionally, after obtaining the steering wheel torque value, the method further includes performing multi-layer low-pass filtering on the steering wheel torque value.

[0019] A hands-off state detection device includes: an identification module for identifying a vehicle scenario and obtaining a hands-off alarm torque value for the scenario based on a steering wheel torque value in the scenario and a preset first steering wheel torque threshold; the hands-off alarm torque value is the steering wheel torque value corresponding to a hands-off alarm; a first determination module for determining a second steering wheel torque threshold corresponding to the scenario based on the hands-off alarm torque value in the scenario; and an alarm judgment module for determining whether the driver has taken their hands off the steering wheel when the vehicle is identified as being in a target scenario, based on the current steering wheel torque value and the second steering wheel torque threshold corresponding to the target scenario, and if so, issuing a hands-off alarm.

[0020] A vehicle includes: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the hands-free state detection method as described in any of the preceding claims.

[0021] A readable storage medium storing a computer program that, when executed by a processor, implements the steps of the off-hand state detection method as described in any of the preceding claims.

[0022] This application provides a method, device, vehicle, and readable storage medium for detecting hands-off state. The method includes: identifying a vehicle scenario; obtaining a hands-off alarm torque value for the scenario based on the steering wheel torque value in the scenario and a preset first steering wheel torque threshold; the hands-off alarm torque value is the steering wheel torque value corresponding to the hands-off alarm; determining a second steering wheel torque threshold corresponding to the scenario based on the hands-off alarm torque value in the scenario; when it is identified that the vehicle is currently in a target scenario, determining whether the driver has taken their hands off the steering wheel based on the current steering wheel torque value and the second steering wheel torque threshold corresponding to the target scenario; if so, issuing a hands-off alarm.

[0023] The technical solution disclosed in this application identifies vehicle scenarios, obtains the corresponding hand-off alarm torque value based on the steering wheel torque value in the scenario and a preset first steering wheel torque threshold, and determines the corresponding second steering wheel torque threshold based on the corresponding hand-off alarm torque value in the scenario. This enables the learning of second steering wheel torque thresholds corresponding to different scenarios, and when the vehicle is in the corresponding scenario, the learned second steering wheel torque thresholds for the corresponding scenario are used to detect the hand-off state, thereby improving the accuracy of hand-off state detection.

[0024] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0025] Figure 1 A flowchart illustrating a method for detecting a hands-free state provided in an embodiment of this application;

[0026] Figure 2 The flowchart of another method for detecting the release state provided in this application;

[0027] Figure 3 A flowchart illustrating the identification of a steering wheel torque sensor installation offset scenario and the determination of the corresponding steering wheel torque threshold provided in this application embodiment;

[0028] Figure 4 A flowchart illustrating whether a vehicle's driving lane is a straight road, as provided in this application embodiment;

[0029] Figure 5 A flowchart illustrating the process of identifying false alarm scenarios and determining the steering wheel torque threshold corresponding to the false alarm scenarios provided in this application embodiment;

[0030] Figure 6 A flowchart illustrating another method for detecting the release state provided in this application embodiment;

[0031] Figure 7 A flowchart of the offset optimization module provided in this application embodiment;

[0032] Figure 8 A flowchart of the false alarm optimization module provided in this application embodiment;

[0033] Figure 9 This is a schematic diagram of the structure of a release state detection device provided in an embodiment of this application;

[0034] Figure 10 This is a structural schematic diagram of a vehicle provided in an embodiment of this application. Detailed Implementation

[0035] Currently, hands-off detection typically involves installing a steering wheel torque sensor in the vehicle. This sensor detects the torque value of the steering wheel and compares it to preset upper and lower threshold values. If the torque value falls within the threshold range, the hands-off state is considered genuine, and a hands-off warning is issued to alert the driver to take over. However, using preset and fixed threshold values ​​for hands-off detection cannot cover variations in steering wheel torque values ​​across different scenarios, resulting in relatively low accuracy in hands-off detection.

[0036] Therefore, this application provides a method, apparatus, device, and readable storage medium for detecting the release state, in order to improve the accuracy of the release state detection.

[0037] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0038] See Figure 1 The diagram illustrates a flowchart of a method for detecting a released state provided in an embodiment of this application. This method for detecting a released state may include:

[0039] S11: Identify the vehicle's scenario and obtain the hands-off alarm torque value based on the steering wheel torque value in the scenario and the preset first steering wheel torque threshold; the hands-off alarm torque value is the steering wheel torque value corresponding to the hands-off alarm.

[0040] It should be noted that the executing entity of the hands-free state detection method provided in this application embodiment can be a vehicle, specifically a driver assistance system in the vehicle (specifically, a driver assistance domain controller within the driver assistance system). This application embodiment uses the driver assistance system in the vehicle as an example for illustration. The vehicle includes an electric power steering system, which includes a steering wheel torque sensor. This steering wheel torque sensor is used to detect the steering wheel torque value in real time and continuously send the detected steering wheel torque value to the vehicle bus. Correspondingly, the driver assistance system can continuously obtain the steering wheel torque value from the vehicle bus.

[0041] After the vehicle's driver assistance function is activated, the driver assistance system can acquire relevant information and identify the vehicle's scene based on that information. When the vehicle is in the identified scene for the first time, it can obtain the steering wheel torque value for the corresponding scene from the vehicle bus, and can also obtain a preset first steering wheel torque threshold. The preset first steering wheel torque threshold includes a preset positive first steering wheel torque threshold TH0 and a preset negative first steering wheel torque threshold TL0 (the preset positive and negative first steering wheel torque thresholds are the same in magnitude but opposite in direction). The preset first steering wheel torque threshold is a default positive and negative threshold for the steering wheel torque after the hands are released during the vehicle development phase. Specifically, it can be taken from the steering wheel torque data after the hands are released in straight-line scenarios during the development process, and can generally be 0.2Nm-0.4Nm (this refers to the absolute value of the preset steering wheel torque threshold). In order to improve the detection effect of the hands-free state, it can be specifically taken as 0.4Nm, that is, the preset positive first steering wheel torque threshold can be 0.4Nm, and the preset negative first steering wheel torque threshold can be -0.4Nm.

[0042] Once the vehicle is in the identified scene for the first time and the steering wheel torque value and the preset first steering wheel torque threshold are obtained for the corresponding scene, it can be determined whether the driver has taken their hands off the steering wheel based on the obtained steering wheel torque value and the preset first steering wheel torque threshold. When it is determined that the driver has taken their hands off the steering wheel, not only can a hands-off alarm be triggered, but also the hands-off alarm torque value for the corresponding scene can be obtained from the obtained steering wheel torque value for the corresponding scene based on the preset first steering wheel torque threshold. The hands-off alarm torque value is the steering wheel torque value corresponding to the hands-off alarm, specifically the steering wheel torque value near the hands-off alarm.

[0043] S12: Determine the second steering wheel torque threshold corresponding to the scenario based on the hands-off alarm torque value in the scenario.

[0044] After obtaining the hands-off alarm torque value for a given scenario, a second steering wheel torque threshold corresponding to that scenario can be determined based on the obtained hands-off alarm torque value (i.e., the second steering wheel torque threshold for a given scenario is determined by self-learning the hands-off alarm torque values ​​for different scenarios). This allows for the optimization of the preset first steering wheel torque threshold based on the hands-off alarm torque values ​​for different scenarios, thereby obtaining a second steering wheel torque threshold suitable for the given scenario. In other words, the second steering wheel torque threshold corresponding to each scenario is obtained through the above method, thus improving the accuracy of hands-off state detection. Specifically, the second steering wheel torque threshold corresponding to a scenario includes a positive second steering wheel torque threshold THs and a negative second steering wheel torque threshold TLs.

[0045] As can be seen from the above, the hands-free state detection method of this application is implemented in two stages. The first stage is to set a preset first steering wheel torque threshold based on the hands-free state torque data collection results during the vehicle development stage. The second stage is to optimize the default threshold of the first stage based on subsequent vehicle hands-free state detection data through self-learning, so as to improve the rationality and accuracy of the steering wheel torque threshold, thereby improving the accuracy of hands-free state detection.

[0046] S13: When it is detected that the vehicle is currently in the target scene, determine whether the driver has taken their hands off the steering wheel based on the current torque value of the steering wheel and the second steering wheel torque threshold corresponding to the target scene. If yes, proceed to step S14; otherwise, return to step S13.

[0047] S14: Activate the hand-off alarm.

[0048] Based on the above, relevant information from the vehicle can be obtained, and the vehicle's scenario can be identified based on the obtained information. When the vehicle is identified as being in a target scenario (here, the target scenario refers to any scenario for which a corresponding second steering wheel torque threshold is determined), the second steering wheel torque threshold corresponding to that target scenario can be obtained. Then, it can be determined whether the driver has taken their hands off the steering wheel based on the current steering wheel torque value and the second steering wheel torque threshold corresponding to the target scenario. For example, if the current steering wheel torque value is within the threshold range determined by the second steering wheel torque threshold corresponding to the target scenario (i.e., within the threshold range determined by the negative and positive steering wheel torque thresholds corresponding to the target scenario), it is determined that the driver has taken their hands off the steering wheel; or if the current steering wheel torque value is within the threshold range determined by the second steering wheel torque threshold corresponding to the target scenario for a certain period of time (e.g., within 15 seconds), it is determined that the driver has taken their hands off the steering wheel, etc.

[0049] If it is determined that the driver has not taken his hands off the steering wheel, and if the vehicle is still in the target scenario, the system can continue to determine whether the driver has taken his hands off the steering wheel based on the current torque value of the steering wheel and the second steering wheel torque threshold corresponding to the target scenario. If the vehicle is not currently in the target scenario, the system can identify the new scenario in which the vehicle is currently located. If the second steering wheel torque threshold corresponding to the new scenario is determined based on steps S11-S12, the system can obtain the second steering wheel torque threshold corresponding to the new scenario and determine whether the driver has taken his hands off the steering wheel based on the current torque threshold of the steering wheel and the second steering wheel torque threshold corresponding to the new scenario (i.e., perform steps similar to step S13). If the second steering wheel torque threshold corresponding to the new scenario is not determined based on steps S11-S12, the system can obtain the hands-off alarm torque value in the scenario based on the steering wheel torque value of the new scenario and the preset first steering wheel torque threshold, and determine the second steering wheel torque threshold corresponding to the new scenario based on the hands-off alarm torque value in the new scenario (i.e., perform steps similar to steps S11-S12 to determine the second steering wheel torque threshold corresponding to the scenario).

[0050] If it is determined that the driver has taken their hands off the steering wheel, a hands-off warning can be triggered, such as an audible and visual alarm or a voice alarm, to prompt the driver to take over the steering wheel and thus ensure the safety of driving the vehicle.

[0051] Compared to using preset upper and lower thresholds for hands-off detection, this application's embodiment determines the steering wheel torque threshold for different scenarios based on the hands-off alarm torque value. Subsequently, when the vehicle is in the corresponding scenario, hands-off detection is performed according to the corresponding steering wheel torque threshold, instead of using preset upper and lower thresholds. This improves the accuracy of hands-off detection and hands-off alarm prompts, thereby enhancing driving safety and the driver's driving experience.

[0052] The technical solution disclosed in this application identifies vehicle scenarios, obtains the corresponding hand-off alarm torque value based on the steering wheel torque value in the scenario and a preset first steering wheel torque threshold, and determines the corresponding second steering wheel torque threshold based on the corresponding hand-off alarm torque value in the scenario. This enables the learning of second steering wheel torque thresholds corresponding to different scenarios, and when the vehicle is in the corresponding scenario, the learned second steering wheel torque thresholds for the corresponding scenario are used to detect the hand-off state, thereby improving the accuracy of hand-off state detection.

[0053] See Figure 2 The diagram illustrates a flowchart of another method for detecting a hands-free state provided in this application. An embodiment of this application provides a method for detecting a hands-free state that, after triggering a hands-free alarm, may further include:

[0054] S15: Obtain the release alarm torque value under the target scenario;

[0055] S16: Determine the new steering wheel torque threshold corresponding to the target scenario based on the hands-off alarm torque value under the target scenario;

[0056] S17: Update the second steering wheel torque threshold corresponding to the target scene using the new steering wheel torque threshold corresponding to the target scene.

[0057] In this embodiment, after determining that the driver has taken their hands off the steering wheel based on the current steering wheel torque value and the second steering wheel torque threshold corresponding to the target scenario, and issuing a hands-off alarm, the hands-off alarm torque value for the target scenario can be obtained from the steering wheel torque value in the target scenario (specifically, the steering wheel torque value compared with the second steering wheel torque threshold corresponding to the target scenario to determine whether the driver has taken their hands off the steering wheel, i.e., the steering wheel torque value in step S13). Then, the second steering wheel torque threshold corresponding to the target scenario can be updated based on the obtained hands-off alarm torque value in the target scenario. Specifically, a new steering wheel torque threshold corresponding to the target scenario can be determined first based on the hands-off alarm torque value in the target scenario (similar to step S12). Then, the existing second steering wheel torque threshold corresponding to the target scenario can be replaced with the new steering wheel torque threshold corresponding to the target scenario to achieve self-learning and updating of the second steering wheel torque threshold in different scenarios, i.e., continuously iterating and updating the second steering wheel torque threshold in the corresponding scenario to improve the accuracy of the second steering wheel torque threshold in the corresponding scenario.

[0058] The above methods enable dynamic learning of steering wheel torque values ​​under different scenarios, thereby dynamically optimizing the second steering wheel torque threshold under different scenarios. This further improves the accuracy of hands-off detection and allows for more accurate hands-off alarm prompts.

[0059] This application provides a method for detecting hands-off state, which obtains a hands-off alarm torque value based on a steering wheel torque value in a given scenario and a preset first steering wheel torque threshold. The method may include:

[0060] The time for the hands-off alarm is determined based on the steering wheel torque value in the scenario and the preset first steering wheel torque threshold. The steering wheel torque value is obtained for the first preset duration before the hands-off alarm and the steering wheel torque value for the second preset duration during the hands-off alarm.

[0061] In this embodiment of the application, when obtaining the hand-off alarm torque value in a scenario based on the steering wheel torque value in the scenario and the preset first steering wheel torque threshold, the hand-off alarm time can be determined based on the steering wheel torque value in the scenario and the preset first steering wheel torque threshold. For example, if the steering wheel torque value in the scenario is within the threshold range determined by the preset first steering wheel torque threshold within a certain time length (e.g., within 15 seconds), then a hand-off alarm is triggered.

[0062] Then, based on the hands-off alarm time in the corresponding scenario, the steering wheel torque value for the first preset duration before the hands-off alarm (the first preset duration can be the same as the certain time length for judging the hands-off alarm, for example, 15s) and the steering wheel torque value for the second preset duration during the hands-off alarm (the second preset duration can be set according to actual needs, for example, 5s) can be obtained. The obtained steering wheel torque value for the first preset duration before the hands-off alarm and the steering wheel torque value for the second preset duration during the hands-off alarm are determined as the hands-off alarm torque value in the scenario. Then, based on the obtained steering wheel torque threshold for the first preset duration before the hands-off alarm and the steering wheel torque threshold for the second preset duration during the hands-off alarm, the second steering wheel torque threshold corresponding to the scenario can be determined.

[0063] By implementing the above method, the steering wheel torque values ​​before and during the hands-off alarm in the corresponding scenario are used as the hands-off alarm torque values ​​for the corresponding scenario to determine the second steering wheel torque threshold for the corresponding scenario. This improves the rationality and diversity of the hands-off alarm torque values ​​obtained in each scenario, thereby improving the rationality and accuracy of the determination of the second steering wheel torque threshold for the corresponding scenario, and ultimately improving the accuracy of hands-off detection.

[0064] It should be noted that in the above embodiments, the method of obtaining the hands-off alarm torque value in the target scenario (i.e., step S15) is similar to the above method. Specifically, obtaining the hands-off alarm torque value in the target scenario may include: when a hands-off alarm is triggered in the target scenario (i.e., when the vehicle is identified as being in the target scenario, and the driver is determined to have taken off the steering wheel based on the current torque value of the steering wheel and the second steering wheel torque threshold corresponding to the target scenario, and a hands-off alarm is triggered), determining the hands-off alarm time, and obtaining the steering wheel torque value for a first preset duration before the hands-off alarm and the steering wheel torque value for a second preset duration during the hands-off alarm, so as to improve the rationality and diversity of the hands-off alarm torque value acquisition in the target scenario, thereby further improving the accuracy of hands-off detection.

[0065] This application provides a method for detecting hands-free state, which determines a second steering wheel torque threshold corresponding to a scenario based on the hands-free alarm torque value in the scenario. The method may include:

[0066] Calculate the first mean and first standard deviation of the steering wheel torque value corresponding to the first preset time before the hands-off alarm;

[0067] Calculate the second mean and second standard deviation of the steering wheel torque value for the first preset duration before the hands-off alarm and the steering wheel torque value for the second preset duration during the hands-off alarm.

[0068] The second steering wheel torque threshold corresponding to the scenario is determined based on the first mean, the first standard deviation, the second mean, and the second standard deviation.

[0069] In this application, after obtaining the steering wheel torque value for a first preset duration before the hands-off alarm and the steering wheel torque value for a second preset duration during the hands-off alarm in the corresponding scenario, and using these data as the hands-off alarm torque value for the corresponding scenario, when determining the second steering wheel torque threshold corresponding to the scenario based on the hands-off alarm torque value in the scenario, specifically, the first mean and the first standard deviation of the steering wheel torque value for the first preset duration before the hands-off alarm can be calculated, and the second mean and the second standard deviation of the steering wheel torque value for the first preset duration before the hands-off alarm and the steering wheel torque value for the second preset duration during the hands-off alarm can be calculated. Then, the second steering wheel torque threshold corresponding to the scenario can be determined based on the first mean, the first standard deviation, the second mean, and the second standard deviation.

[0070] The specific process for determining the second steering wheel torque threshold corresponding to a scenario based on the first mean, first standard deviation, second mean, and second standard deviation can be as follows: A final mean is obtained by weighting the first mean, the weight corresponding to the first preset duration before the hand-off alarm, the second mean, and the weights corresponding to the first preset duration before and the second preset duration during the hand-off alarm. A final standard deviation is then obtained by weighting the first standard deviation, the weight corresponding to the first preset duration before the hand-off alarm, the second standard deviation, and the weights corresponding to the first preset duration before and the second preset duration during the hand-off alarm. The weights corresponding to the first preset duration before and the second preset duration during the hand-off alarm can be obtained in advance through experiments. Then, the positive second steering wheel torque threshold THs = final mean + coefficient * final standard deviation and the negative second steering wheel torque threshold TLs = final mean - coefficient * final standard deviation are used to obtain the positive second steering wheel torque threshold THs and the negative second steering wheel torque threshold TLs corresponding to the scenario, respectively. The coefficient can be obtained through prior experiments or set based on experience; for example, the coefficient could be 1.5. Alternatively, other methods can be used to determine the steering wheel torque threshold corresponding to the scenario.

[0071] By calculating the first mean and first standard deviation of the steering wheel torque value for a first preset duration before the hands-off alarm, the second mean and second standard deviation of the steering wheel torque value for a first preset duration before the hands-off alarm and the steering wheel torque value for a second preset duration during the hands-off alarm, and then calculating the steering wheel torque threshold corresponding to the scenario based on the first mean, first standard deviation, second mean and second standard deviation, the accuracy of the steering wheel torque threshold calculation corresponding to the scenario can be improved, thereby improving the accuracy of hands-off state detection.

[0072] It should be noted that, in the above embodiments, the specific implementation process of determining the new steering wheel torque threshold corresponding to the target scenario based on the hands-off alarm torque value in the target scenario (i.e., the above step S16) is similar to the specific implementation method of determining the steering wheel torque threshold corresponding to the scenario based on the hands-off alarm torque value in the scenario, and will not be repeated here.

[0073] For details, please refer to [link / reference]. Figure 3 This document illustrates a flowchart illustrating the identification of a steering wheel torque sensor installation offset scenario and the determination of the corresponding steering wheel torque threshold, as provided in an embodiment of this application. An embodiment of this application provides a hands-off state detection method for identifying vehicle scenarios, which may include:

[0074] Obtain the steering wheel torque value and the vehicle's driving lane, and use the steering wheel torque value to obtain the mean and absolute value of the steering wheel torque value;

[0075] Determine whether the absolute value of the average steering wheel torque is greater than the first preset value and whether the vehicle is traveling in a straight road.

[0076] If the average absolute value of the steering wheel torque is greater than the first preset value and the vehicle is traveling in a straight lane, then the scenario is determined to be the scenario where the steering wheel torque sensor is installed off-center in the vehicle; the steering wheel torque sensor is used to detect the steering wheel torque value.

[0077] This application embodiment considers that the electric power steering system in a vehicle may experience steering wheel torque sensor misalignment due to factory installation, manufacturing tolerances, or other reasons. This misalignment causes the steering wheel torque value detected by the steering torque sensor to continuously deviate to the left or right, manifesting as an overall off-center steering wheel torque value on the bus signal, continuously shifting in the positive or negative direction. Prolonged misalignment of the steering wheel torque value can lead to the absolute value of the torque signal continuously exceeding a preset steering wheel torque threshold when the hands are off, resulting in an inability to output the correct hands-off state and thus no hands-off alarm. This affects the ADS (Autonomous Driving System) vehicle control and triggers a hands-off alarm, leading to missed alarms. To improve the accuracy of hands-off state detection and alarm, this application embodiment, when identifying vehicle scenarios, first identifies whether the vehicle is in a scenario where the steering wheel torque sensor is installed off-center. If it is determined that the vehicle is in a scenario where the steering wheel torque sensor is installed off-center, a second steering wheel torque threshold corresponding to this scenario can be determined based on the above process, thereby improving the accuracy of the steering wheel torque sensor and hands-off state detection corresponding to the scenario where the steering wheel torque sensor is installed off-center. The scenario of steering wheel torque sensor installation misalignment refers to the situation where the steering wheel torque sensor is not located in the center position due to reasons such as factory installation, but is biased to the left or right.

[0078] Specifically, the steering wheel torque value and the vehicle's driving lane can be obtained (the vehicle's driving lane can be obtained through visual perception), and the mean absolute value of the steering wheel torque value can be obtained using the obtained steering wheel torque value. Then, it can be determined whether the mean absolute value of the steering wheel torque value is greater than a first preset value and whether the vehicle's driving lane is a straight road. The first preset value is a value greater than 0, and the specific value can be obtained in advance through experiments, for example, it can be 0.3 Nm.

[0079] If the mean absolute value of the steering wheel torque is determined to be greater than the first preset value and the vehicle is traveling in a straight lane, then it is determined that the mean absolute value of the steering wheel torque is greater than the first preset value due to the installation misalignment of the steering wheel torque sensor. In this case, the scenario can be determined as the steering wheel torque sensor installation misalignment scenario. Then, the hands-off alarm torque value under the steering wheel torque sensor installation misalignment scenario can be obtained based on the steering wheel torque value under the steering wheel torque sensor installation misalignment scenario and the preset first steering wheel torque threshold. Based on the hands-off alarm torque value under the steering wheel torque sensor installation misalignment scenario, the second steering wheel torque threshold TH1 (specifically, the positive second steering wheel torque threshold corresponding to the steering wheel torque sensor installation misalignment scenario) and TL1 (specifically, the negative second steering wheel torque threshold corresponding to the steering wheel torque sensor installation misalignment scenario) corresponding to the steering wheel torque sensor installation misalignment scenario are determined. The second positive steering wheel torque threshold TH1 is used to replace the preset first positive steering wheel torque threshold TH0, and the second negative steering wheel torque threshold TL1 is used to replace the preset first negative steering wheel torque threshold TL0. Subsequently, when the vehicle is identified as being in a scenario where the steering wheel torque sensor is installed off-center (generally, if the vehicle is not repaired, it will remain in this scenario, and the specific method for identifying this scenario is similar to the one described above, so it will not be repeated here), the second steering wheel torque thresholds TH1 and TL1 corresponding to this scenario can be used to detect the hands-off state, thereby improving the accuracy of the hands-off state detection. Furthermore, the second steering wheel torque thresholds TH1 and TL1 corresponding to this scenario can be iteratively updated to further improve their accuracy, thus further enhancing the accuracy of the hands-off state detection.

[0080] If the absolute value of the average steering wheel torque value is determined to be no greater than the first preset value or the vehicle is traveling in a non-straight road, then the scenario is determined not to be the scenario of the steering wheel torque sensor being installed off-center in the vehicle.

[0081] By first identifying whether the vehicle is in a scenario where the steering wheel torque sensor is installed off-center, and then calculating the corresponding second steering wheel torque threshold when the vehicle is in a scenario where the steering wheel torque sensor is installed off-center, the accuracy of the steering wheel torque threshold determination can be improved, thereby improving the accuracy of hands-off state detection and hands-off alarm.

[0082] See Figure 4This document illustrates a flowchart illustrating whether a vehicle's driving lane is a straight road, as provided in an embodiment of this application. The method for detecting a hands-free state, as provided in this embodiment, to determine whether a vehicle's driving lane is a straight road, may include:

[0083] Obtain the steering wheel angle and the radius of curvature of the vehicle's driving lane;

[0084] Determine whether the steering wheel angle is within the preset angle range and whether the radius of curvature of the vehicle's driving lane is greater than the second preset value;

[0085] If so, then the vehicle's driving lane is determined to be a straight road.

[0086] In this embodiment of the application, the specific process of determining whether the vehicle driving lane is a straight road can be as follows: obtaining the steering wheel angle and the radius of curvature of the vehicle driving lane, wherein the radius of curvature of the vehicle driving lane can be obtained through visual perception; determining whether the steering wheel angle is within a preset angle range and whether the radius of curvature of the vehicle driving lane is greater than a second preset value, wherein the preset angle range and the second preset value can be obtained in advance through experiments, for example, the preset angle range can be ±5° and the second preset value can be 500m; if the steering wheel angle is within the preset angle range and the radius of curvature of the vehicle driving lane is greater than the second preset value, then the vehicle driving lane is determined to be a straight road; if the steering wheel angle is not within the preset angle range or the radius of curvature of the vehicle driving lane is not greater than the second preset value, then the vehicle driving lane is determined to be a non-straight road.

[0087] By combining the steering wheel angle and the radius of curvature of the vehicle's driving lane to determine whether the driving lane is straight, the accuracy of straight lane determination can be improved, thereby improving the accuracy of hands-free state detection.

[0088] See Figure 5 This document illustrates a flowchart illustrating the process of identifying false alarm scenarios and determining the steering wheel torque threshold corresponding to the false alarm scenarios, as provided in an embodiment of this application. The hands-free state detection method provided in this embodiment may further include:

[0089] Once the scenario is determined to be a steering wheel torque sensor installation misalignment scenario in the vehicle, it is determined whether the driver is holding the steering wheel at the 6 o'clock position and whether the hands-off alarm is triggered based on the first steering wheel torque threshold corresponding to the steering wheel torque sensor installation misalignment scenario in the vehicle. If so, the scenario is determined to be a false alarm scenario. Before executing the step of obtaining the hands-off alarm torque value in the scenario based on the steering wheel torque value in the scenario and the preset first steering wheel torque threshold, the preset first steering wheel torque threshold is replaced with the second steering wheel torque threshold corresponding to the determined steering wheel torque sensor installation misalignment scenario in the vehicle.

[0090] Once the scenario is determined to be a non-vehicle steering wheel torque sensor installation misalignment scenario, it is determined whether the driver is holding the steering wheel at the 6 o'clock position and whether the hands-off alarm is triggered based on the preset first steering wheel torque threshold. If so, the scenario is determined to be a false alarm scenario, and the step of obtaining the hands-off alarm torque value under the scenario based on the steering wheel torque value under the scenario and the preset first steering wheel torque threshold is executed.

[0091] Identifying that the vehicle is currently in a target scenario, and determining whether the driver has taken their hands off the steering wheel based on the current steering wheel torque value and the second steering wheel torque threshold corresponding to the target scenario, may include:

[0092] Determine whether the driver is holding the steering wheel at the 6 o'clock position. If so, determine whether the driver has taken their hands off the steering wheel based on the current torque value of the steering wheel and the second steering wheel torque threshold corresponding to the false alarm scenario.

[0093] This application embodiment takes into account the possibility of false alarms where the system still triggers a hands-off alarm even when the driver's hands are at the lowest position on the steering wheel. In order to improve the accuracy of hands-off state detection, this application embodiment can identify whether the vehicle is in a false alarm scenario, and when the vehicle is identified to be in a false alarm scenario, it determines the steering wheel torque threshold corresponding to the false alarm scenario based on the above-mentioned self-learning process.

[0094] Specifically, after determining that the vehicle scenario is a steering wheel torque sensor installation misalignment scenario, it can further determine whether the driver's hands are at the 6 o'clock position on the steering wheel, and also determine whether to trigger a hands-off alarm based on the second steering wheel torque threshold corresponding to the steering wheel torque sensor installation misalignment scenario. Here, the 6 o'clock position on the steering wheel is defined as the center of the steering wheel being the center of a clock, the front of the steering wheel being the 12 o'clock position, and the horizontal direction being determined according to the clock's time scale. The back of the steering wheel is the 6 o'clock position, which is the lowest point of the steering wheel. Furthermore, as can be seen from the above, in the case of a steering wheel torque sensor installation misalignment scenario, the initial identification of a false alarm scenario is based on the steering wheel torque threshold corresponding to the steering wheel torque sensor installation misalignment scenario to determine whether to trigger a hands-off alarm, thereby improving the accuracy of the hands-off alarm determination. If the driver holds the steering wheel at the 6 o'clock position and the second steering wheel torque threshold corresponding to the scenario of steering wheel torque sensor installation misalignment in the vehicle is determined to trigger the hands-off alarm, it indicates that the driver is lightly touching the steering wheel at the 6 o'clock position and the steering wheel torque value is low. In this case, it can be further determined that the vehicle's scenario is a false alarm scenario, that is, not only is the steering wheel torque sensor installed in the vehicle misaligned, but it is also in a false alarm scenario. In this scenario, before obtaining the hand-off alarm torque value based on the steering wheel torque value and the preset first steering wheel torque threshold, the preset first steering wheel torque threshold can be replaced with the second steering wheel torque threshold corresponding to the steering wheel torque sensor installation offset scenario (i.e., the second steering wheel torque threshold corresponding to the steering wheel torque sensor installation offset scenario is used as the initial steering wheel torque threshold for the false alarm scenario to participate in the self-learning of the steering wheel torque threshold corresponding to the false alarm scenario). Then, the hand-off alarm torque threshold for the false alarm scenario can be obtained based on the steering wheel torque value and the second steering wheel torque threshold corresponding to the steering wheel torque sensor installation offset scenario. Based on the hand-off alarm torque value for the false alarm scenario, the second steering wheel torque thresholds TH2 (specifically, the positive second steering wheel torque threshold) and TL2 (specifically, the negative second steering wheel torque threshold) corresponding to the false alarm scenario are determined to improve the accuracy of the steering wheel torque threshold determination for the false alarm scenario, thereby facilitating the improvement of the accuracy of hand-off state detection. During the above judgment process, if it is determined that the driver is not holding the steering wheel at the 6 o'clock position or the hands-off alarm is not triggered, then the scenario is determined to be a non-false alarm scenario. The second steering wheel torque threshold corresponding to the steering wheel torque sensor installation offset scenario in the vehicle can continue to be updated through self-learning according to the steering wheel torque sensor installation offset scenario in the vehicle.

[0095] Once the vehicle's scenario is determined to be a scenario other than the one where the steering wheel torque sensor is misaligned, it is determined whether the driver's hand is at the 6 o'clock position, and whether a hands-off alarm is triggered based on a preset first steering wheel torque threshold. If the driver's hand is at the 6 o'clock position and the hands-off alarm is triggered according to the preset first steering wheel torque threshold, it indicates that the driver is lightly touching the steering wheel at the 6 o'clock position and the steering wheel torque value is low. In this case, the vehicle's scenario can be determined to be a false alarm scenario. Then, the hands-off alarm torque value under the scenario can be obtained based on the steering wheel torque value under the false alarm scenario and the preset first steering wheel torque threshold. Based on the hands-off alarm torque value under the false alarm scenario, the second steering wheel torque threshold TH2 (specifically, the positive second steering wheel torque threshold corresponding to the false alarm scenario) and TL2 (specifically, the negative second steering wheel torque threshold corresponding to the false alarm scenario) corresponding to the false alarm scenario are determined to improve the accuracy of the determination of the second steering wheel torque threshold corresponding to the false alarm scenario, thereby facilitating the improvement of the accuracy of hands-off state detection.

[0096] It should be noted that, in determining whether a hands-off alarm has been triggered in the two scenarios described above, if a hands-off alarm is confirmed to have been triggered, it can also be determined whether the number of consecutive triggers of the hands-off alarm exceeds a preset number. The preset number can be 3 and can be adjusted. Only if the number of consecutive triggers exceeds the preset number is it determined to be a false alarm scenario, thus improving the accuracy of false alarm scenario identification. Specifically, consecutive triggers of the hands-off alarm refer to a situation where the driver manually cancels the hands-off alarm after it has been triggered once, and then triggers the hands-off alarm again.

[0097] Based on the above, the system identifies the vehicle's current location within a target scenario. It then determines whether the driver has taken their hands off the steering wheel based on the current steering wheel torque value and the corresponding second steering wheel torque threshold for the target scenario. Specifically, it checks if the driver's hands are at the 6 o'clock position. If the driver is confirmed to be at the 6 o'clock position, the vehicle is identified as being in the target scenario corresponding to the false alarm scenario. The system can then further determine whether the driver has taken their hands off the steering wheel based on the current steering wheel torque value and the corresponding second steering wheel torque thresholds TH2 and TL2. If the driver has taken their hands off the steering wheel, a hands-off warning is issued to improve the accuracy of hands-off detection in this scenario. Furthermore, the steering wheel torque thresholds corresponding to the above scenario can be updated and iterated upon in subsequent iterations to further improve the accuracy of determining the steering wheel torque thresholds for this scenario.

[0098] This application provides a method for detecting hands-free status, determining whether the driver's hands are at the 6 o'clock position on the steering wheel, which may include:

[0099] Obtain visual information about the driver's hand position on the steering wheel, and determine whether the driver's hands are at the 6 o'clock position based on the visual information.

[0100] In this embodiment, determining whether the driver's hands are at the 6 o'clock position on the steering wheel can be achieved by using a vision detection device inside the vehicle to acquire visual information about the driver's hand position on the steering wheel, and then sending this visual information to the driver assistance system. The driver assistance system can then acquire this visual information and determine whether the driver's hands are at the 6 o'clock position based on it, thereby improving the convenience and accuracy of determining the driver's hand position at the 6 o'clock position.

[0101] This application provides a method for detecting a hands-off state, which determines whether the driver has taken their hands off the steering wheel based on the current torque value of the steering wheel and a second steering wheel torque threshold corresponding to the target scenario. If so, a hands-off alarm is triggered. The method may include:

[0102] Determine whether the current torque value of the steering wheel is within the threshold range determined by the second steering wheel torque threshold corresponding to the target scenario within the third preset time period;

[0103] If the current torque value of the steering wheel is within the threshold range for the third preset time period, a first-level hands-off alarm will be triggered, and it will be determined whether the current torque value of the steering wheel is within the threshold range for the fourth preset time period.

[0104] If the current torque value of the steering wheel is within the threshold range for the fourth preset time period, a level two hands-off alarm will be triggered, and it will be determined whether the current torque value of the steering wheel is within the threshold range for the fifth preset time period.

[0105] If the current torque value of the steering wheel is within the threshold range for the fifth preset time period, then exit the assisted driving mode and continue to issue the second-level hands-off warning.

[0106] In this embodiment of the application, the process of determining whether the driver has taken their hands off the steering wheel based on the current torque value of the steering wheel and the second steering wheel torque threshold corresponding to the target scenario, and if so, issuing a hands-off alarm, can be as follows:

[0107] (1) First, determine whether the current torque value of the steering wheel is within the threshold range determined by the second steering wheel torque threshold corresponding to the target scenario within the third preset time period. The third preset time period can be set by relevant personnel based on experience or by conducting experiments in advance. The third preset time period can be the same as the first preset time period mentioned above. The threshold range is specifically determined by the negative steering wheel torque threshold and the positive steering wheel torque threshold corresponding to the target scenario.

[0108] (2) If the current torque value of the steering wheel is not within the range defined by the second steering wheel torque threshold corresponding to the target scenario within the third preset time period, a hands-off alarm will not be triggered. If the current torque value of the steering wheel is within the range defined by the second steering wheel torque threshold corresponding to the target scenario within the third preset time period, a level one hands-off alarm will be triggered, such as a level one audible and visual alarm. Furthermore, while triggering the level one hands-off alarm, it is also determined whether the current torque value of the steering wheel is within the threshold range defined by the second steering wheel torque threshold corresponding to the target scenario within the fourth preset time period. The fourth preset time period can also be set by relevant personnel based on experience or prior testing, for example, it can be 15 seconds.

[0109] (3) If it is determined that the current torque value of the steering wheel is not within the threshold range determined by the second steering wheel torque threshold corresponding to the target scene within the fourth preset time period, the second-level hands-off alarm can be omitted. If it is determined that the current torque value of the steering wheel is within the threshold range determined by the second steering wheel torque threshold corresponding to the target scene within the fourth preset time period, the second-level hands-off alarm can be activated. The first-level hands-off alarm and the second-level hands-off alarm have different forms; for example, the second-level hands-off alarm can be a second-level sound, light, and vibration alarm. Furthermore, while activating the second-level hands-off alarm, it is also determined whether the current torque value of the steering wheel is within the threshold range determined by the second steering wheel torque threshold corresponding to the target scene within the fifth preset time period. The fifth preset time period can also be set by relevant personnel based on experience or pre-testing, for example, it can be 10 seconds.

[0110] (4) If it is determined that the current torque value of the steering wheel is not within the threshold range determined by the second steering wheel torque threshold corresponding to the target scenario within the fifth preset time period, the second-level hands-off alarm can be terminated. If it is determined that the current torque value of the steering wheel is within the threshold range determined by the second steering wheel torque threshold corresponding to the target scenario within the fifth preset time period, the second-level hands-off alarm can continue and the driver assistance function can be deactivated to ensure driving safety.

[0111] Taking a third preset duration of 15s, a fourth preset duration of 15s, a fifth preset duration of 10s, a first-level hands-off alarm as a first-level audible and visual alarm, and a second-level hands-off alarm as an audible, visual, and vibration alarm as an example, the above process can be specifically as follows: continuously monitor and obtain the current torque value of the steering wheel. If the current torque value of the steering wheel is within the threshold range determined by the steering wheel torque threshold corresponding to the target scenario for 15s, then a first-level audible and visual alarm is triggered. If the current torque value of the steering wheel continues to be within the threshold range determined by the steering wheel torque threshold corresponding to the target scenario for 15s, then a second-level audible, visual, and vibration alarm is triggered. After the second-level audible, visual, and vibration alarm continues for 10s, the system exits the assisted driving function and continues to issue alarms.

[0112] Different levels of hands-off warnings alert drivers to take over the vehicle in a timely manner to ensure the safety of both the vehicle and the driver.

[0113] The hands-free state detection method provided in this application embodiment, after obtaining the steering wheel torque value, may further include:

[0114] Multi-layer low-pass filtering is applied to the steering wheel torque value.

[0115] In this embodiment of the application, after each acquisition of the steering wheel torque value, the assisted driving system can first perform multi-layer low-pass filtering on the steering wheel torque value to try to filter out torque disturbance signals caused by system instability, and then use the multi-layer low-pass filtered steering wheel torque value for subsequent processing to improve the quality of the steering wheel torque value, thereby facilitating the improvement of the accuracy of hands-free state detection.

[0116] For a clearer explanation of the scheme in this application, please refer to [link / reference needed]. Figure 6 The document illustrates a flowchart of another hands-free state detection method provided in an embodiment of this application. The driver assistance system may include a hands-free detection module, which is not only connected to the electric power steering system but also includes a default threshold module, an offset optimization module, and a false alarm optimization module. Specifically, the hands-free detection module obtains the steering wheel torque value from the electric power steering system and performs a low-pass filter on the obtained steering wheel torque value to attempt to filter out torque disturbance signals caused by system instability, thereby improving the quality of the steering wheel torque value.

[0117] After filtering and processing by various optimization modules within the hands-off detection module, the system determines the hands-off state by combining the continuously updated self-learned steering wheel torque thresholds of each module in different scenarios. The hands-off detection module continuously monitors the steering wheel torque value. If the current steering wheel torque value remains within the threshold range defined by the steering wheel torque threshold for the target scenario for 15 seconds, a level one audible and visual alarm is triggered. If the current steering wheel torque value continues to remain within the threshold range defined by the steering wheel torque threshold for the target scenario for 15 seconds, a level two audible, visual, and vibration alarm is triggered. After the level two audible, visual, and vibration alarm continues for 10 seconds, the system exits the assisted driving function but continues to issue alarms.

[0118] The default threshold module defines preset first steering wheel torque thresholds TH0 and TL0. The default values ​​are generally taken from the steering wheel torque data after releasing the steering wheel in straight-line scenarios during the R&D process, and are generally 0.2Nm-0.4Nm. It is recommended to use larger values ​​here.

[0119] The offset optimization module addresses the scenario of steering wheel torque sensor installation misalignment in vehicles. Specifically, due to factory installation issues, the torque value output by the steering wheel torque sensor in the electric power steering system may be continuously biased to the left or right. This manifests on the bus signal as the steering wheel torque value being out of center, continuously shifting in the positive or negative direction. Prolonged steering wheel torque value misalignment can cause the absolute value of the torque signal to consistently exceed the preset first steering wheel torque thresholds TL0 and TH0 when the hands are off the steering wheel, preventing the correct hands-off state from being output and thus failing to trigger a hands-off alarm. This affects the ADS vehicle control system's ability to trigger a hands-off alarm, resulting in missed alarms. See [link to details] for more information. Figure 7 The diagram illustrates the workflow of the offset optimization module provided in this application embodiment. The offset optimization module can acquire the steering wheel torque value and perform multi-layer low-pass filtering, and acquire the vehicle's driving lane. If it is determined that the road is not curved and the system triggers a hands-off alarm, the steering wheel torque value in the 15 seconds before the hands-off alarm and the 5 seconds during the alarm can be searched and recorded, and these steering wheel torque values ​​can be used as the hands-off alarm torque values ​​for the corresponding scenario. Then, the hands-off alarm torque values ​​collected in this scenario are filtered, and the mean 1 and standard deviation 1 of the 15 seconds before the hands-off alarm, and the mean 2 and standard deviation 2 of the 15 seconds before the hands-off alarm and the 5 seconds during the duration of the hands-off alarm are calculated. For the mean 1 and mean 2, weights 1 and 2 are used to calculate the final mean A, and for the standard deviation 1 and standard deviation 2, weights 1 and 2 are used to calculate the final standard deviation A. The second steering wheel torque thresholds TH1 and TL1 corresponding to the scenario are obtained using TH1 = mean A + 1.5 * standard deviation A and TL1 = mean A - 1.5 * standard deviation A. Subsequently, all similar hands-off alarm data can be calculated iteratively, and the values ​​of TH1 and TL1 can be iterated. That is, after each training, the system can output the upper and lower limit thresholds of steering wheel torque TH1 and TL1 of the offset optimization module. When the system determines that the steering wheel torque sensor is installed off-center, specifically when the hands-off torque continuously deviates from 0, the system uses TH1 and TL1 as the judgment thresholds and continuously learns and optimizes.

[0120] The false alarm optimization module addresses false alarm scenarios. Specifically, a false alarm scenario refers to an occasional low steering wheel torque value triggered when the driver lightly touches the steering wheel at the 6 o'clock position with one hand. This value is lower than the preset steering wheel torque thresholds TL0 and TH0 (in scenarios without steering wheel torque sensor installation misalignment), or lower than the steering wheel torque thresholds TL1 and TH1 corresponding to scenarios with steering wheel torque sensor installation misalignment (in such scenarios). This triggers a hands-off alarm, resulting in a false alarm. See [link to documentation] for details. Figure 8The diagram illustrates the workflow of the false alarm optimization module provided in this application embodiment. It can acquire steering wheel torque values ​​and perform multi-layer low-pass filtering. If an in-cabin vision detection module exists and can detect the position or orientation of the driver's hands relative to the steering wheel, this status flag is input into the false alarm optimization module. If the system determines that the driver's hands are at the 6 o'clock position on the steering wheel and triggers the hands-off alarm three times consecutively, the system automatically searches for and records the steering wheel torque values ​​for the 15 seconds before the alarm and the 5 seconds during the alarm, and uses these steering wheel torque values ​​as the hands-off alarm torque values ​​for the corresponding scenario. Then, the hands-off alarm torque value in this scenario is filtered, and the mean 1 and standard deviation 1 of the 15 seconds before the hands-off alarm, and the mean 2 and standard deviation 2 of the 15 seconds before the hands-off alarm and the 5 seconds during the hands-off alarm are calculated. For the mean 1 and mean 2, the final mean A is calculated using weight 1 and weight 2. For the standard deviation 1 and standard deviation 2, the final standard deviation A is calculated using weight 1 and weight 2. The second steering wheel torque thresholds TH2 and TL2 corresponding to the scenario are obtained using TH2 = mean A + 1.5 * standard deviation A and TL2 = mean A - 1.5 * standard deviation A. Subsequently, all similar hands-off alarm data can be calculated in this loop, iterating the values ​​of TH2 and TL2. That is, after each training, the system can output the upper and lower limit thresholds of steering wheel torque TH2 and TL2 of the false alarm optimization module. When the system determines that a false alarm occurs, specifically when the steering wheel torque value is low and the driver's hands are at the 6 o'clock position relative to the steering wheel, the system uses TH2 and TL2 as judgment thresholds and continuously learns and optimizes them.

[0121] This application also provides a device for detecting a hands-off state, see [link to relevant documentation]. Figure 9 It shows a schematic diagram of the structure of a release state detection device provided in an embodiment of this application, which may include:

[0122] The recognition module 91 is used to recognize the vehicle scene and obtain the hands-off alarm torque value under the scene based on the steering wheel torque value under the scene and the preset first steering wheel torque threshold; the hands-off alarm torque value is the steering wheel torque value corresponding to the hands-off alarm.

[0123] The first determining module 92 is used to determine the second steering wheel torque threshold corresponding to the scenario based on the hands-off alarm torque value in the scenario.

[0124] The alarm judgment module 93 is used to determine whether the driver has taken his hands off the steering wheel when the vehicle is currently in the target scene, based on the current torque value of the steering wheel and the second steering wheel torque threshold corresponding to the target scene. If so, an alarm is triggered for taking hands off the steering wheel.

[0125] The release state detection device provided in this application embodiment may further include:

[0126] The acquisition module is used to acquire the release alarm torque value under the target scenario;

[0127] The second determining module is used to determine the new steering wheel torque threshold corresponding to the target scenario based on the hands-off alarm torque value in the target scenario.

[0128] The update module is used to update the second steering wheel torque threshold corresponding to the target scene using the new steering wheel torque threshold corresponding to the target scene.

[0129] This application provides a device for detecting a hands-free state, wherein the identification module 91 may include:

[0130] The first acquisition unit is used to determine the time of the hands-off alarm based on the steering wheel torque value in the scenario and the preset first steering wheel torque threshold, and to acquire the steering wheel torque value for a first preset duration before the hands-off alarm and the steering wheel torque value for a second preset duration during the hands-off alarm.

[0131] This application provides a release state detection device, wherein the first determining module 92 may include:

[0132] The first calculation unit is used to calculate the first mean and the first standard deviation of the steering wheel torque value corresponding to the first preset time before the hands-off alarm;

[0133] The second calculation unit is used to calculate the steering wheel torque value for a first preset duration before the hands-off alarm and the steering wheel torque value for a second preset duration during the hands-off alarm;

[0134] The first determining unit is used to determine the second steering wheel torque threshold corresponding to the scenario based on the first mean, the first standard deviation, the second mean, and the second standard deviation.

[0135] This application provides a device for detecting a hands-free state, wherein the identification module 91 may include:

[0136] The second acquisition unit is used to acquire the steering wheel torque value and the vehicle driving lane, and to obtain the mean and absolute value of the steering wheel torque value using the steering wheel torque value.

[0137] The first judgment unit is used to determine whether the average absolute value of the steering wheel torque is greater than a first preset value and whether the vehicle driving lane is a straight road.

[0138] The second determining unit is used to determine the scenario as the steering wheel torque sensor installation offset scenario if the average absolute value of the steering wheel torque value is greater than the first preset value and the vehicle driving lane is a straight road; the steering wheel torque sensor is used to detect the steering wheel torque value.

[0139] This application provides a release state detection device, wherein the first judgment unit may include:

[0140] The acquisition sub-unit is used to acquire the steering wheel angle and the radius of curvature of the vehicle's driving lane;

[0141] The first judgment subunit is used to determine whether the steering wheel angle is within the preset angle range and whether the radius of curvature of the vehicle driving lane is greater than the second preset value.

[0142] The determination subunit is used to determine that the vehicle driving lane is a straight road if the steering wheel angle is within a preset angle range and the radius of curvature of the vehicle driving lane is greater than a second preset value.

[0143] The hand-release state detection device provided in this application embodiment may further include, in the identification module 91:

[0144] The second judgment unit is used to determine whether the driver is holding the steering wheel at the 6 o'clock position after determining that the scenario is a steering wheel torque sensor installation offset scenario in the vehicle, and to determine whether to trigger the hands-off alarm based on the second steering wheel torque threshold corresponding to the steering wheel torque sensor installation offset scenario in the vehicle; if so, the scenario is determined to be a false alarm scenario, and before executing the step of obtaining the hands-off alarm torque value in the scenario based on the steering wheel torque value in the scenario and the preset first steering wheel torque threshold, the preset first steering wheel torque threshold is replaced with the determined second steering wheel torque threshold corresponding to the steering wheel torque sensor installation offset scenario in the vehicle.

[0145] The third judgment unit is used to determine whether the driver is holding the steering wheel at the 6 o'clock position and whether the hands-off alarm is triggered based on the preset first steering wheel torque threshold after determining that the scenario is not a steering wheel torque sensor installation offset scenario in the vehicle; if so, the scenario is determined to be a false alarm scenario, and the step of obtaining the hands-off alarm torque value in the scenario based on the steering wheel torque value in the scenario and the preset first steering wheel torque threshold is executed.

[0146] The alarm detection module 93 may include:

[0147] The fourth judgment unit is used to determine whether the driver is holding the steering wheel at the 6 o'clock position. If so, it determines whether the driver has taken his hands off the steering wheel based on the current torque value of the steering wheel and the second steering wheel torque threshold corresponding to the false alarm scenario.

[0148] The release state detection device provided in this application embodiment may include, as follows: the second judgment unit, the third judgment unit, and the fourth judgment unit may each include:

[0149] The second judgment subunit is used to obtain visual information about the position of the driver's hands on the steering wheel, and to determine whether the driver's hands are at the 6 o'clock position based on the visual information.

[0150] This application provides a device for detecting a hands-free state, wherein the alarm module 93 may include:

[0151] The fifth judgment unit is used to determine whether the current torque value of the steering wheel is within the threshold range determined by the second steering wheel torque threshold corresponding to the target scene within the third preset time period;

[0152] The sixth judgment unit is used to issue a first-level hands-off alarm if the current torque value of the steering wheel is within the threshold range for the third preset time period, and to determine whether the current torque value of the steering wheel is within the threshold range for the fourth preset time period.

[0153] The seventh judgment unit is used to issue a second-level hands-off alarm if the current torque value of the steering wheel is within the threshold range for the fourth preset time period, and to determine whether the current torque value of the steering wheel is within the threshold range for the fifth preset time period.

[0154] The exit and alarm unit is used to exit the assisted driving and continue to issue a level 2 hands-free alarm if the current torque value of the steering wheel is within the threshold range for a fifth preset time period.

[0155] This application also provides a vehicle, see [link to example]. Figure 10 It shows a structural schematic diagram of a vehicle provided in an embodiment of this application, which may include:

[0156] Memory 101 is used to store computer programs;

[0157] When processor 102 executes a computer program stored in memory 101, it can perform the following steps:

[0158] The system identifies the vehicle's current scenario and obtains the hand-off alarm torque value based on the steering wheel torque value and a preset first steering wheel torque threshold. The hand-off alarm torque value corresponds to the steering wheel torque value for the hand-off alarm. Based on the hand-off alarm torque value, the system determines the second steering wheel torque threshold corresponding to the scenario. When the system detects that the vehicle is currently in the target scenario, it determines whether the driver has taken their hands off the steering wheel based on the current steering wheel torque value and the second steering wheel torque threshold corresponding to the target scenario. If so, a hand-off alarm is triggered.

[0159] This application embodiment also provides a readable storage medium storing a computer program, which, when executed by a processor, can perform the following steps:

[0160] The system identifies the vehicle's current scenario and obtains the hand-off alarm torque value based on the steering wheel torque value and a preset first steering wheel torque threshold. The hand-off alarm torque value corresponds to the steering wheel torque value for the hand-off alarm. Based on the hand-off alarm torque value, the system determines the second steering wheel torque threshold corresponding to the scenario. When the system detects that the vehicle is currently in the target scenario, it determines whether the driver has taken their hands off the steering wheel based on the current steering wheel torque value and the second steering wheel torque threshold corresponding to the target scenario. If so, a hand-off alarm is triggered.

[0161] For a description of the hands-free state detection device, vehicle, and readable storage medium provided in this application embodiment, please refer to the detailed description of the corresponding part in the hands-free state detection method provided in this application embodiment, and it will not be repeated here.

[0162] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any readable storage medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "readable storage medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of readable storage media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the readable storage medium could even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0163] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0164] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0165] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0166] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "joining," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise expressly limited. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0167] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for detecting a released state, characterized in that, include: The vehicle scene identification process includes: acquiring the steering wheel torque value and the vehicle's driving lane; using the steering wheel torque value to obtain the mean absolute value of the steering wheel torque value; determining whether the mean absolute value of the steering wheel torque value is greater than a first preset value and whether the vehicle's driving lane is a straight road; if the mean absolute value of the steering wheel torque value is greater than the first preset value and the vehicle's driving lane is a straight road, then the scene is determined to be a scenario where the steering wheel torque sensor is installed off-center in the vehicle; the steering wheel torque sensor is used to detect the steering wheel torque value. The hands-off alarm torque value in the scenario is obtained based on the steering wheel torque value in the scenario and a preset first steering wheel torque threshold; the hands-off alarm torque value is the steering wheel torque value corresponding to the hands-off alarm. Based on the hands-off alarm torque value in the scenario, determine the second steering wheel torque threshold corresponding to the scenario; When the vehicle is identified as being in a target scenario, the system determines whether the driver has taken their hands off the steering wheel based on the current torque value of the steering wheel and the second steering wheel torque threshold corresponding to the target scenario. If so, a hands-off alarm is triggered.

2. The method for detecting the release state according to claim 1, characterized in that, After triggering the release alarm, it also includes: Obtain the release alarm torque value under the target scenario; Determine the new steering wheel torque threshold corresponding to the target scenario based on the hands-off alarm torque value under the target scenario; The second steering wheel torque threshold corresponding to the target scene is updated using the new steering wheel torque threshold corresponding to the target scene.

3. The method for detecting the release state according to claim 1, characterized in that, The hand-off alarm torque value for the scenario is obtained based on the steering wheel torque value in the scenario and a preset first steering wheel torque threshold, including: The time for the hands-off alarm is determined based on the steering wheel torque value in the scenario and the preset first steering wheel torque threshold. The steering wheel torque value is obtained for a first preset duration before the hands-off alarm and for a second preset duration during the hands-off alarm.

4. The method for detecting the release state according to claim 3, characterized in that, Based on the hands-off alarm torque value in the described scenario, determine the corresponding second steering wheel torque threshold for the scenario, including: Calculate the first mean and first standard deviation of the steering wheel torque value corresponding to the first preset time before the hands-off alarm; Calculate the second mean and the second standard deviation of the steering wheel torque value for the first preset duration before the hands-off alarm and the steering wheel torque value for the second preset duration during the hands-off alarm. The second steering wheel torque threshold corresponding to the scenario is determined based on the first mean, the first standard deviation, the second mean, and the second standard deviation.

5. The method for detecting the release state according to claim 1, characterized in that, Determining whether the vehicle's driving lane is a straight road includes: Obtain the steering wheel angle and the radius of curvature of the vehicle's driving lane; Determine whether the steering wheel angle is within a preset angle range and whether the radius of curvature of the vehicle's driving lane is greater than a second preset value; If so, then the vehicle's driving lane is determined to be a straight road.

6. The method for detecting the release state according to claim 1, characterized in that, Also includes: Once the scenario is determined to be the scenario of the steering wheel torque sensor installation offset in the vehicle, it is determined whether the driver is holding the steering wheel at the 6 o'clock position and whether to trigger the hands-off alarm based on the second steering wheel torque threshold corresponding to the scenario of the steering wheel torque sensor installation offset in the vehicle. If so, the scenario is determined to be a false alarm scenario, and before performing the step of obtaining the hands-off alarm torque value under the scenario based on the steering wheel torque value under the scenario and the preset first steering wheel torque threshold, the preset first steering wheel torque threshold is replaced with the second steering wheel torque threshold corresponding to the determined scenario of the steering wheel torque sensor installation offset in the vehicle. Once it is determined that the scenario is not a scenario where the steering wheel torque sensor is installed off in the vehicle, it is determined whether the driver is holding the steering wheel at the 6 o'clock position and whether to trigger the hands-off alarm based on the preset first steering wheel torque threshold. If so, the scenario is determined to be a false alarm scenario, and the step of obtaining the hands-off alarm torque value under the scenario based on the steering wheel torque value under the scenario and the preset first steering wheel torque threshold is executed; Identifying that the vehicle is currently in a target scenario, and determining whether the driver has taken their hands off the steering wheel based on the current steering wheel torque value and the second steering wheel torque threshold corresponding to the target scenario, including: Determine whether the driver is holding the steering wheel at the 6 o'clock position. If so, determine whether the driver has taken their hands off the steering wheel based on the current torque value of the steering wheel and the second steering wheel torque threshold corresponding to the false alarm scenario.

7. The method for detecting the release state according to claim 6, characterized in that, Determining whether the driver's hands are at the 6 o'clock position on the steering wheel includes: Obtain visual information about the driver's hand position on the steering wheel, and determine whether the driver's hand is at the 6 o'clock position on the steering wheel based on the visual information.

8. The method for detecting the release state according to claim 1, characterized in that, Based on the current torque value of the steering wheel and the second steering wheel torque threshold corresponding to the target scenario, it is determined whether the driver has taken their hands off the steering wheel. If so, a hands-off alarm is triggered, including: Determine whether the current torque value of the steering wheel is within the threshold range determined by the second steering wheel torque threshold corresponding to the target scenario within the third preset time period; If the current torque value of the steering wheel is within the threshold range for the third preset time period, a first-level hands-off alarm is triggered, and it is determined whether the current torque value of the steering wheel is within the threshold range for the fourth preset time period. If the current torque value of the steering wheel is within the threshold range for the fourth preset time period, a second-level hands-off alarm is triggered, and it is determined whether the current torque value of the steering wheel is within the threshold range for the fifth preset time period. If the current torque value of the steering wheel is within the threshold range within the fifth preset time period, then the driver assistance system will be discontinued and the level 2 hands-free alarm will continue.

9. The method for detecting the release state according to claim 1, characterized in that, After obtaining the steering wheel torque value, the following is also included: The steering wheel torque value is subjected to multi-layer low-pass filtering.

10. A device for detecting a release state, characterized in that, include: The identification module is used to identify vehicle scenarios, including: acquiring steering wheel torque values ​​and the vehicle's driving lane; obtaining the mean absolute value of the steering wheel torque values ​​using the steering wheel torque values; determining whether the mean absolute value of the steering wheel torque values ​​is greater than a first preset value and whether the vehicle's driving lane is a straight road; if the mean absolute value of the steering wheel torque values ​​is greater than the first preset value and the vehicle's driving lane is a straight road, then determining that the scenario is a scenario where the steering wheel torque sensor is installed off-center in the vehicle; the steering wheel torque sensor is used to detect the steering wheel torque values; acquiring the hands-off alarm torque value in the scenario based on the steering wheel torque value in the scenario and a preset first steering wheel torque threshold; the hands-off alarm torque value is the steering wheel torque value corresponding to the hands-off alarm. The first determining module is used to determine the second steering wheel torque threshold corresponding to the scenario based on the hands-off alarm torque value in the scenario. The alarm detection module is used to determine whether the driver has taken his hands off the steering wheel when the vehicle is currently in a target scene, based on the current torque value of the steering wheel and the second steering wheel torque threshold corresponding to the target scene. If so, an alarm is triggered to indicate whether the driver has taken his hands off the steering wheel.

11. A vehicle, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the off-hand state detection method as described in any one of claims 1 to 9 when executing the computer program.

12. A readable storage medium, characterized in that, The readable storage medium stores a computer program that, when executed by a processor, implements the steps of the off-hand state detection method as described in any one of claims 1 to 9.

Citation Information

Patent Citations

  • Device and a method for driver releasing warning

    CN109050397A

  • Vehicle and controlling method of the vehicle

    KR1020170069675A