Steering wheel hand-off detection method, eps system, and vehicle

Abstract

The application relates to the field of automobile technology and discloses a steering wheel hand-off detection method, an EPS system and an automobile. The method is applied to an EPS system with a preset detection model, and the method comprises the following steps: acquiring a torque signal and an angle signal of a to-be-detected steering wheel; determining a current hand force value and a current hand force change rate applied to the to-be-detected steering wheel according to the torque signal and the angle signal; judging whether the current hand force value is smaller than a preset hand force threshold value and whether the current hand force change rate is smaller than a preset hand force change rate threshold value; and if yes, determining that the to-be-detected steering wheel is currently in a hand-off state. Since the current hand force value and the current hand force change rate applied to the to-be-detected steering wheel can be used to judge whether the steering wheel is in a hand-off state by the EPS system, compared with the existing method of assembling a capacitive steering wheel, the hand-off detection can be directly realized based on the EPS system, and the production cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of automotive technology, and in particular to a method for detecting steering wheel hands-off operation, an EPS system, and an automobile. Background Technology

[0002] Currently, with the development of driver assistance technology, more and more vehicles are equipped with electric power steering (EPS) systems, making driver hands-off detection increasingly important. When a driver takes their hands off the steering wheel while driving, it is necessary to monitor and warn them of this behavior.

[0003] The existing method typically involves equipping a steering wheel with a capacitor to detect hands-free driving, but this method is costly, which in turn increases the overall production cost of the vehicle.

[0004] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of this invention is to provide a steering wheel hands-off detection method, an EPS system, and an automobile, aiming to solve the technical problem that the high cost of hands-off detection caused by equipping a capacitive steering wheel in the prior art.

[0006] To achieve the above objectives, the present invention provides a method for detecting steering wheel hand-off, the method being applied to an EPS system with a preset detection model, the method comprising the following steps:

[0007] Acquire the torque and angle signals of the steering wheel to be tested;

[0008] The current hand force applied to the steering wheel to be detected and the current rate of change of hand force are determined based on the torque signal and the angle signal.

[0009] Determine whether the current hand force value is less than a preset hand force threshold and whether the current hand force change rate is less than a preset hand force change rate threshold. The preset hand force threshold and the preset hand force change rate threshold are determined based on the preset detection model.

[0010] If so, it is determined that the steering wheel to be tested is currently in a hands-free state.

[0011] Optionally, the step of determining the current hand force applied to the steering wheel to be detected and the current rate of change of hand force based on the torque signal and the angle signal includes:

[0012] The current hand force applied to the steering wheel to be tested is determined based on the torque signal;

[0013] The rate of change of the current hand force applied to the steering wheel to be detected is determined based on the current hand force value and the angle signal.

[0014] Optionally, before the step of determining whether the current hand force value is less than a preset hand force threshold and whether the current hand force change rate is less than a preset hand force change rate threshold, the method further includes:

[0015] The current rack force of the steering wheel to be tested is determined based on the current hand force value;

[0016] The degree of road bumpiness is determined based on the current rack force, and the preset hand force threshold and preset hand force change rate threshold corresponding to the degree of road bumpiness are determined based on the preset detection model.

[0017] Optionally, before the step of acquiring the torque signal and angle signal of the steering wheel to be detected, the method further includes:

[0018] Determine the typical road surface bump level and the atypical road surface bump level from the various road surface bump levels;

[0019] Determine the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the steering wheel to be tested under the typical road surface bumpiness;

[0020] The initial detection model is calibrated based on the standard preset hand force threshold and the standard preset hand force change rate threshold.

[0021] The preset detection model is obtained by self-learning the degree of bumpiness of the atypical road surface through the calibrated initial detection model.

[0022] Optionally, the step of self-learning the atypical road surface bumpiness level using the calibrated initial detection model to obtain the preset detection model includes:

[0023] The learning hand force value and the learning hand force change rate of the steering wheel under the atypical road surface bump level are obtained;

[0024] The hand force value to be learned and the rate of change of the hand force to be learned are filtered;

[0025] The target typical road surface bumpiness degree corresponding to the atypical road surface bumpiness degree is determined based on the filtered learning hand force value and the rate of change of the filtered learning hand force.

[0026] The initial detection model is self-learned by using the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the typical road surface bumpiness level to obtain the preset detection model.

[0027] Optionally, the step of self-learning the initial detection model using a standard preset hand force threshold and a standard preset hand force change rate threshold corresponding to the target typical road surface bumpiness to obtain the preset detection model includes:

[0028] The initial detection model is self-learned using the standard preset hand force threshold corresponding to the typical road surface bumpiness, the filtered hand force value to be learned, the standard preset hand force change rate threshold, and the filtered hand force change rate to be learned.

[0029] The current confidence value is obtained based on the learning results, and when the current confidence value is within a preset confidence interval, the initial detection model after self-learning is used as the preset detection model.

[0030] Optionally, before the step of acquiring the torque signal and angle signal of the steering wheel to be detected, the method further includes:

[0031] When the EPS system is in lane keeping function enabled, typical and atypical requested torque and angle are determined from each requested torque and angle.

[0032] Determine the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the steering wheel to be tested under the typical requested torque and angle;

[0033] The initial detection model is calibrated based on the standard preset hand force threshold and the standard preset hand force change rate threshold.

[0034] The preset detection model is obtained by self-learning the atypical requested torque and angle through the calibrated initial detection model.

[0035] In addition, to achieve the above objectives, the present invention also proposes an EPS system, which includes: a signal acquisition module, a hand force determination module, a hand force judgment module, and a state judgment module.

[0036] The signal acquisition module is used to acquire the torque signal and angle signal of the steering wheel to be detected;

[0037] The hand force determination module is used to determine the current hand force value and the current hand force change rate applied to the steering wheel to be detected based on the torque signal and the angle signal.

[0038] The hand force judgment module is used to determine whether the current hand force value is less than a preset hand force threshold and whether the current hand force change rate is less than a preset hand force change rate threshold. The preset hand force threshold and the preset hand force change rate threshold are determined based on the preset detection model.

[0039] The state determination module is used to determine if the steering wheel to be detected is currently in a hands-free state.

[0040] Optionally, the hand force determination module is further configured to determine the current hand force value applied to the steering wheel to be detected based on the torque signal;

[0041] The hand force determination module is also used to determine the rate of change of the current hand force applied to the steering wheel to be detected based on the current hand force value and the angle signal.

[0042] In addition, to achieve the above objectives, the present invention also proposes an automobile that includes the EPS system as described above.

[0043] This invention provides a method for detecting steering wheel hands-off issues, an EPS system, and a vehicle. The method is applied to an EPS system with a preset detection model. The method includes the following steps: acquiring torque and angle signals of the steering wheel to be detected; determining the current hand force value and the current hand force change rate applied to the steering wheel to be detected based on the torque and angle signals; determining whether the current hand force value is less than a preset hand force threshold and whether the current hand force change rate is less than a preset hand force change rate threshold, wherein the preset hand force threshold and the preset hand force change rate threshold are determined based on the preset detection model; if so, determining that the steering wheel to be detected is currently in a hands-off state. Because this invention can determine the current hand force value and the current hand force change rate applied to the steering wheel to be tested by the EPS system based on the acquired torque signal and angle signal, and then determine whether the current hand force value is less than a preset hand force threshold and whether the current hand force change rate is less than a preset hand force change rate threshold, wherein the preset hand force threshold and the preset hand force change rate can be determined based on a preset detection model, if so, it is determined that the steering wheel to be tested is in a hands-free state. Compared with the existing method that requires the installation of a capacitor steering wheel, this invention can directly realize hands-free detection based on the EPS system, reducing production costs. Attached Figure Description

[0044] Figure 1 This is a flowchart illustrating the first embodiment of the steering wheel hands-off detection method of the present invention;

[0045] Figure 2 This is a flowchart illustrating the second embodiment of the steering wheel hands-off detection method of the present invention;

[0046] Figure 3 This is a schematic diagram illustrating the definition of the confidence value interval in the steering wheel hands-off detection method of the present invention;

[0047] Figure 4 This is a flowchart illustrating the third embodiment of the steering wheel hands-off detection method of the present invention;

[0048] Figure 5This is a diagram showing the interaction between the EPS system and related systems in the steering wheel hands-off detection method of the present invention;

[0049] Figure 6 This is a structural block diagram of the first embodiment of the EPS system of the present invention.

[0050] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0051] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0052] It should be noted that with the development of driver assistance technology, more and more vehicles are equipped with electric power steering (EPS) systems. As a result, driver hands-off detection is becoming increasingly important. When a driver takes both hands off the steering wheel while driving, it is necessary to monitor and warn the driver of this behavior.

[0053] The existing method typically involves equipping a steering wheel with a capacitor to detect hands-free driving, but this method is costly, which in turn increases the overall production cost of the vehicle.

[0054] To address the aforementioned shortcomings, this embodiment provides a steering wheel hands-off detection method, applied to an EPS system with a preset detection model. The preset detection model can be a model used for hands-off detection and can be set within the EPS system.

[0055] In its specific implementation, this embodiment acquires the torque and angle signals of the steering wheel to be tested; determines the current hand force applied to the steering wheel and the current rate of change of hand force based on the torque and angle signals; judges whether the current hand force is less than a preset hand force threshold and whether the current rate of change of hand force is less than a preset rate of change of hand force threshold, wherein both the preset hand force threshold and the preset rate of change of hand force can be determined based on a preset detection model; if so, it is determined that the steering wheel to be tested is currently in a hands-free state. Since this embodiment can determine whether it is in a hands-free state based on the current hand force applied to the steering wheel and the current rate of change of hand force through the EPS system, compared with the existing method that requires the installation of a capacitor-equipped steering wheel, this embodiment can directly realize hands-free detection based on the EPS system, reducing production costs.

[0056] For ease of understanding, the following is combined with Figures 1 to 5 The steering wheel hands-off detection method provided in the embodiments of this application will be described in detail.

[0057] This invention provides a method for detecting steering wheel hand-off, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the steering wheel hands-off detection method of the present invention.

[0058] In this embodiment, the steering wheel hands-off detection method includes the following steps:

[0059] Step S10: Obtain the torque signal and angle signal of the steering wheel to be tested.

[0060] It should be noted that the method in this embodiment can be applied to scenarios where the driver's hands are released from the steering wheel, or other scenarios requiring hands-free detection. This method can be executed by an EPS system, or by other electronic devices that perform the same or similar functions. The following description uses the aforementioned EPS system to illustrate this embodiment and the subsequent embodiments.

[0061] Understandably, the steering wheel to be tested can be the steering wheel of the vehicle driven by the driver, the torque signal can be the signal corresponding to the force or torque applied by the driver when turning the steering wheel to be tested, which can be obtained in real time by a torque sensor; the angle signal can be the signal corresponding to the rotation angle applied by the driver when turning the steering wheel to be tested, which can be obtained in real time by an angle sensor.

[0062] In practical use, the EPS system can provide corresponding steering assistance to the vehicle based on the torque signal collected by the torque sensor, and at the same time provide corresponding rotation angle to the vehicle based on the angle signal collected by the angle sensor, thereby improving the driver's driving comfort.

[0063] Step S20: Determine the current hand force value and the current hand force change rate applied to the steering wheel to be detected based on the torque signal and the angle signal.

[0064] It should be understood that the aforementioned current hand force value can be the value corresponding to the force currently applied by the driver to the steering wheel to be detected, and the aforementioned current hand force change rate can be the change rate corresponding to the driver's current hand force value over a certain period of time.

[0065] Furthermore, in order to obtain the aforementioned current hand force value and current hand force change rate, in this embodiment, step S20 includes:

[0066] Step S21: Determine the current hand force applied to the steering wheel to be tested based on the torque signal;

[0067] Step S22: Determine the rate of change of the current hand force applied to the steering wheel to be detected based on the current hand force value and the angle signal.

[0068] It should also be understood that the EPS system can directly obtain the current hand force value based on the torque signal, determine the current rotation angle of the steering wheel to be detected based on the angle signal, and when the current rotation angle reaches a specific angle, the ratio of the difference between the current hand force value corresponding to the start of rotation and the current hand force value corresponding to the end of rotation to the time required for the rotation is used as the current hand force change rate. The specific angle can be set according to the actual situation, and this embodiment does not limit it.

[0069] In its specific implementation, the EPS system can acquire the torque signal generated by the torque sensor and the angle signal generated by the angle sensor, and obtain the current hand force value applied by the driver to the steering wheel to be detected based on the torque signal, obtain the current rotation angle of the steering wheel to be detected based on the angle signal, and determine the current hand force change rate based on the current rotation angle and the current hand force value.

[0070] Step S30: Determine whether the current hand force value is less than a preset hand force threshold and whether the current hand force change rate is less than a preset hand force change rate threshold. The preset hand force threshold and the preset hand force change rate threshold are determined based on the preset detection model.

[0071] Step S40: If yes, then it is determined that the steering wheel to be detected is currently in a hands-free state.

[0072] It should be noted that the aforementioned preset hand force threshold and preset hand force change rate threshold can both be used to determine whether the hand is released, and can be preset in the preset detection model.

[0073] This embodiment determines whether the steering wheel to be tested is in a hands-free state by measuring the difference between the current hand force value and a preset hand force threshold, and the difference between the current hand force change rate and a preset hand force change rate threshold. This is because, under normal circumstances, when a driver is driving normally, they continuously apply a certain amount of hand force to maintain control of the steering wheel and make fine adjustments to the steering wheel in real time. Once the hand leaves the steering wheel, the hand force value and the hand force change rate will rapidly decrease to a very small or zero value. Therefore, if the current hand force value is less than the preset hand force threshold and the current hand force change threshold is less than the preset hand force change rate threshold, it can be determined that the steering wheel to be tested is currently in a hands-free state. Similarly, if the current hand force value is greater than the preset hand force threshold and the current hand force change threshold is greater than the preset hand force change rate threshold, it can be determined that the steering wheel to be tested is currently in a hands-on state.

[0074] In its implementation, the EPS system can determine whether the user is in a release state by the relationship between the current hand force value and the preset hand force threshold, as well as the relationship between the current hand force change rate and the preset hand force change rate threshold.

[0075] Furthermore, since factors such as different vehicle speeds and varying road surface roughness can affect the judgment results, in this embodiment, preset hand force thresholds and preset hand force change rate thresholds can be pre-tested for different scenarios (different vehicle speeds, varying road surface roughness, etc.). The initial detection model is calibrated by associating the preset hand force thresholds and preset hand force change rate thresholds obtained from the tests with the corresponding scenarios, thereby obtaining the aforementioned preset detection model. In actual use, the aforementioned preset detection model can select the corresponding preset hand force thresholds and preset hand force change rate thresholds based on the identified current scenario before making a judgment, thus improving the comprehensiveness of the application scenarios.

[0076] For example, when considering the impact of different vehicle speeds, this embodiment can pre-obtain the preset hand force threshold and preset hand force change rate threshold corresponding to different vehicle speeds, and calibrate the initial detection model using the preset hand force threshold and preset hand force change rate threshold to obtain the above-mentioned preset detection model; for example, calibration can be performed on each vehicle speed from 0 to 120 km / h to obtain the above-mentioned preset detection model.

[0077] For example, when considering the impact of different road surface bump levels, this embodiment can also pre-obtain preset hand force thresholds and preset hand force change rate thresholds corresponding to different road surface bump levels, and calibrate the initial detection model using the preset hand force thresholds and preset hand force change rate thresholds to obtain the aforementioned preset detection model; for example, the road surface bump level can be divided into 0 to 100, where 0 is a non-bumpy road surface, and calibration can be performed for each road surface bump level to obtain the aforementioned preset detection model.

[0078] For example, this embodiment can also take into account vehicle speed and road surface bumpiness together. Specifically, it can calibrate different road surface bumpiness corresponding to different vehicle speeds to obtain the above-mentioned preset detection model.

[0079] It should be emphasized that in the actual detection process, the EPS system can directly obtain the current vehicle speed through the vehicle speed sensor. The preset detection model then selects the corresponding preset hand force threshold and preset hand force change rate based on the current vehicle speed for judgment. Of course, the current vehicle speed can also be obtained through other methods, and this embodiment does not limit this.

[0080] In obtaining the degree of road bumpiness, this embodiment can acquire it in real time using a torque sensor. The specific process is as follows: before step S30 above, it also includes:

[0081] Step S31: Determine the current rack force of the steering wheel to be tested based on the current hand force value.

[0082] It should be noted that the aforementioned current rack force can refer to the reaction force generated on the steering rack by the force applied by the driver to the steering wheel to be tested when the vehicle is turning; in this embodiment, the above-mentioned preset detection model can store the mapping relationship between different current hand force values ​​and the corresponding current rack force, wherein the mapping relationship can be set according to the characteristics of the EPS system of different vehicles, and this embodiment does not impose any restrictions on the specific settings.

[0083] Step S32: Determine the road surface bump level based on the current rack force, and determine the preset hand force threshold and preset hand force change rate threshold corresponding to the road surface bump level based on the preset detection model;

[0084] Understandably, since different current rack forces correspond to different degrees of road bumps, the EPS system can determine the degree of road bumps on the road surface where the vehicle is currently traveling based on the current rack force, and determine the corresponding preset hand force threshold and preset hand force change rate threshold from the preset detection model. It then determines whether the current hand force value is less than the preset hand force threshold and whether the current hand force change rate is less than the preset hand force change rate threshold. If so, it is determined to be a hands-free state.

[0085] This embodiment can pre-calibrate the initial detection model according to preset hand force thresholds and preset hand force change rate thresholds corresponding to different vehicle speeds and different road surface bump levels, thus obtaining a preset detection model. In actual use, the EPS system can acquire the torque signal generated by the torque sensor and the angle signal generated by the angle sensor. Based on the torque signal, it obtains the current hand force value applied by the driver to the steering wheel to be tested, and based on the angle signal, it obtains the current rotation angle of the steering wheel to be tested. Based on the current rotation angle and the current hand force value, it determines the current hand force change rate, and then based on the preset detection model, it determines the preset hand force value and the preset hand force change rate threshold. It then determines whether the current hand force value is less than the preset hand force threshold and whether the current hand force change rate is less than the preset hand force change rate threshold. If so, it is determined to be a hands-off state. Compared with the existing method that requires the installation of a capacitor-loaded steering wheel, this embodiment can directly realize hands-off detection based on the EPS system, thereby reducing production costs.

[0086] refer to Figure 2 , Figure 2 This is a flowchart illustrating the second embodiment of the steering wheel hands-off detection method of the present invention.

[0087] Considering that calibrating all scenarios during calibration would be time-consuming and inefficient, this embodiment selects some typical scenarios from all scenarios to calibrate the initial detection model in order to improve calibration efficiency and ensure accuracy. Then, it learns from all scenarios through self-learning to obtain the aforementioned preset detection model.

[0088] To facilitate understanding, this embodiment first explains the self-learning scenario at different vehicle speeds. Specifically, in this embodiment, the speed range from 0 to 120 km / h can be pre-divided according to a certain gradient. If each gradient is 10 km / h, then speeds of 0 to 10 km / h, 10 to 20 km / h, 20 to 30 km / h, 30 to 40 km / h, 40 to 50 km / h, 50 to 60 km / h, 60 to 70 km / h, 70 to 80 km / h, 80 to 90 km / h, 90 to 100 km / h, 100 to 110 km / h, and 110 km / h can be obtained. There are 12 speed ranges from 10 km / h to 120 km / h. Then, 10 km / h, 20 km / h, 30 km / h, 40 km / h, 50 km / h, 60 km / h, 70 km / h, 80 km / h, 90 km / h, 100 km / h, 110 km / h and 120 km / h are selected as typical vehicle speeds. The standard preset hand force threshold and standard preset hand force change rate threshold corresponding to the typical vehicle speed are obtained by measurement. The initial detection model is calibrated. The initial detection model can be the model before self-learning. This embodiment does not limit the specific model.

[0089] After calibration, the preset hand force threshold and preset hand force change rate threshold corresponding to atypical vehicle speeds can be learned through self-learning, thereby obtaining the above-mentioned preset detection model. The atypical vehicle speed can be any speed other than the typical vehicle speed, ranging from 0 to 120 km / h.

[0090] Furthermore, to determine whether learning was successful, this embodiment can use a confidence value for judgment, referring to... Figure 3 To explain, Figure 3 This is a schematic diagram illustrating the definition of the confidence value interval in the steering wheel hands-off detection method of the present invention, as shown below. Figure 3 As shown, in this embodiment, the confidence value can be divided into 0 to 15, where the range of 0 to 4 is defined as the steering wheel being in a hands-on state (i.e., ...). Figure 3 (Not yet released from hand), corresponding to the current hand force value being greater than a preset hand force threshold, and the current hand force change rate being greater than a preset hand force change rate threshold; definitions 11 to 15 represent the steering wheel being released from hand (i.e., not yet released from hand). Figure 3 "Disengagement" refers to the situation where the current hand force value is less than a preset hand force threshold and the current hand force change rate is less than a preset hand force change rate threshold; definitions 4 to 11 indicate that the steering wheel to be detected is in the critical range (i.e., ...). Figure 3 The critical interval corresponds to the situation where the current hand force value is greater than the preset hand force threshold and the current hand force change rate is less than the preset hand force change rate threshold, or the situation where the current hand force value is less than the preset hand force threshold and the current hand force change rate is greater than the preset hand force change rate threshold.

[0091] Furthermore, during calibration, the initial detection model can directly output a confidence value between 0 and 4 or between 11 and 15 when it receives the standard preset hand force threshold and preset hand force change rate threshold corresponding to typical vehicle speeds. During self-learning, it can determine whether the output confidence value at each non-typical vehicle speed is between 0 and 4 or between 11 and 15. If so, it indicates that the self-learning of that interval is complete. If the output confidence value is between 4 and 11, it indicates that the self-learning of that interval is not complete and can be repeated. As the maturity of the learning increases, it can gradually approach 4 or 11. If it is still in the critical interval after repeated learning a certain number of times, the interval can be calibrated until the output confidence value at all vehicle speeds is no longer in the critical interval, at which point the initial detection model can be used as the preset detection model.

[0092] In actual use, the EPS system can directly obtain confidence values ​​based on the relationship between the current hand force value and the preset hand force threshold, as well as the relationship between the current hand force change rate and the preset hand force change rate threshold. The confidence values ​​are then transmitted to the associated systems (prompt systems, alarm systems, etc.). The associated systems determine whether the hand is in a release state based on the received confidence values ​​and issue a release alarm when the hand is in a release state.

[0093] As another implementation method, the EPS system can also determine whether it is in a released state based on the obtained confidence value, and generate an alarm signal to be transmitted to the associated system to trigger a release alarm when it is in a released state.

[0094] Furthermore, this embodiment will further describe the self-learning scenario under different road surface bump levels, and the specific process is as follows: Figure 2 As shown, in this embodiment, before step S10, the following steps are also included:

[0095] Step S01: Determine the typical road surface bump level and the atypical road surface bump level from the various road surface bump levels.

[0096] It should be noted that this embodiment can also divide the road bumpiness level from 0 to 100 according to a certain gradient. If 10 is used as a gradient, then 10 intervals can be obtained: 0 to 10, 10 to 20, 20 to 30, 30 to 40, 40 to 50, 50 to 60, 60 to 70, 70 to 80, 80 to 90, and 90 to 100. Then, 10, 20, 30, 40, 50, 60, 70, 80, 90, and 100 are selected as the above-mentioned typical road bumpiness levels, and the remaining intervals from 0 to 100 excluding the typical road bumpiness levels are selected as the above-mentioned atypical road bumpiness levels.

[0097] Step S02: Determine the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the steering wheel to be tested under the typical road surface bump level.

[0098] Understandably, the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the steering wheel under the above typical road bump level can be obtained by measurement; that is, under the typical road bump level, the current hand force value corresponding to whether the steering wheel under test is in a hands-free state or not is used as the standard preset hand force threshold, and the current hand force change rate is used as the standard hand force change threshold.

[0099] Step S03: Calibrate the initial detection model based on the standard preset hand force threshold and the standard preset hand force change rate threshold;

[0100] Step S04: The atypical road surface bumpiness is self-learned using the calibrated initial detection model to obtain the preset detection model.

[0101] It should be understood that the above EPS system can calibrate the initial detection model by using the standard preset hand force threshold and the standard preset hand force change rate threshold under various typical road surface bump levels, so that the confidence value output by the initial detection model under typical road surface bump levels is 0 to 4 or 11 to 15.

[0102] After calibration, the preset hand force threshold and preset hand force change rate threshold corresponding to the atypical road surface bumpiness are learned through self-learning, thereby obtaining the above-mentioned preset detection model.

[0103] Furthermore, in order to improve accuracy during self-learning, step S04 above includes:

[0104] Step S041: Obtain the learning hand force value and the learning hand force change rate of the steering wheel under the atypical road surface bump level.

[0105] It should be noted that the aforementioned hand force value to be learned can be the current hand force value corresponding to the torque signal collected by the torque sensor under atypical road surface bumps, and the aforementioned hand force change rate to be learned can be the current hand force change rate corresponding to the angle signal collected by the angle sensor and the hand force value to be learned.

[0106] Step S042: Filter the hand force value to be learned and the rate of change of the hand force to be learned;

[0107] Step S043: Determine the target typical road bump level corresponding to the atypical road bump level based on the filtered learning hand force value and the rate of change of the filtered learning hand force.

[0108] Understandably, during bumpy conditions, the learnable hand force value and the rate of change of learnable hand force obtained by the EPS system contain a significant amount of noise and interference. Therefore, the learnable hand force value and the rate of change of learnable hand force can be filtered to obtain filtered learnable hand force value and filtered learnable hand force rate of change. The filtering method can be mean filtering, median filtering, etc., or other methods; this embodiment does not limit this.

[0109] It should be understood that after obtaining the filtered learning hand force value and the filtered learning hand force change rate, the atypical road bump level can be equated to a typical road bump level based on the filtered learning hand force value and the filtered learning hand force change rate, and this typical road bump level can be used as the aforementioned target typical road bump level. For example, when learning the atypical road bump level corresponding to 53 in the interval 50 to 60, after filtering the learning hand force value and the learning hand force change rate corresponding to the atypical road bump level 53, the atypical road bump level 53 can be equated to the typical road bump level 50 based on the filtered learning hand force value and the filtered learning hand force change rate, and the typical road bump level 50 can be used as the aforementioned target typical road bump level.

[0110] Step S044: The initial detection model is self-learned by using the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the typical road surface bumpiness of the target to obtain the preset detection model.

[0111] It should also be understood that after determining the typical road surface bumpiness level as typical road surface bumpiness level 50, a preset detection model can be obtained by self-learning through the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the typical road surface bumpiness level 50 during calibration.

[0112] Further, step S044 includes: performing self-learning on the initial detection model using the standard preset hand force threshold corresponding to the target typical road surface bumpiness, the filtered hand force value to be learned, the standard preset hand force change rate threshold, and the filtered hand force change rate to be learned; obtaining the current confidence value based on the learning result, and using the self-learned initial detection model as the preset detection model when the current confidence value is within a preset confidence interval.

[0113] It should be noted that, based on the above example, the EPS system can perform self-learning on the initial detection model according to the filtered hand force value to be learned, the hand force change rate to be learned, and the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the typical road bumpiness level 50.

[0114] Understandably, the aforementioned current confidence value can be the confidence value output by this self-learning iteration, and the aforementioned preset confidence interval can be the interval corresponding to 0 to 4 or 11 to 15. For example, during the self-learning process, the initial detection model can output a confidence value for each self-learning result of atypical road surface bumpiness. If the output confidence value is in the interval of 0 to 4 or 11 to 15, it indicates that the self-learning for that atypical road surface bumpiness is complete. Learning continues until all current confidence values ​​output in the 50 to 60 interval are in the 0 to 4 or 11 to 15 interval, indicating that the self-learning for that interval is complete. If the output current confidence value is a value between 4 and 11, it indicates that the self-learning for that interval is not complete and can be repeated. As the learning maturity increases, it can gradually approach 4 or 11. If it remains in the critical interval after a certain number of repetitions, the interval can be calibrated until the current confidence values ​​output for all road surface bumpiness levels are no longer in the critical interval, at which point the initial detection model can be used as the preset detection model.

[0115] It is important to emphasize that in actual use, EPS can first determine whether the vehicle is on a non-abnormally bumpy road surface based on the current rack force. If it is (i.e., the road bumpiness is 0), the current hand force value can be directly compared with the preset hand force threshold, and the current hand force change rate can be compared with the preset hand force change rate threshold without filtering. If it is on an abnormally bumpy road surface, the current hand force value and the current hand force change rate need to be filtered. Based on the filtered current hand force value and the current hand force change rate, the equivalent typical road bumpiness level under that road bumpiness level is determined, and then a comparison is made to determine whether to let go.

[0116] Before actual use, the EPS system described in this embodiment can pre-calibrate the initial detection model by selecting typical vehicle speeds from various vehicle speeds. Then, the calibrated initial detection model can be used to learn from other atypical vehicle speeds until the output confidence values ​​at all vehicle speeds are within the range of 0 to 4 or 11 to 15. The self-learned initial detection model can then be used as the preset detection model. Similarly, the initial detection model can be calibrated by selecting typical road surface roughness from various road surface roughness levels. Then, the calibrated initial detection model can be used to learn from other atypical road surface roughness levels until the output confidence values ​​at all vehicle speeds are within the range of 0 to 4 or 11 to 15. Because this embodiment uses a calibration-then-learning approach, it is unnecessary to calibrate all cases, thus improving calibration efficiency.

[0117] It should also be emphasized that if vehicle speed and road surface roughness are taken into account, the specific calibration and self-learning process is similar to the above process, and will not be described in detail here.

[0118] refer to Figure 4 , Figure 4This is a flowchart illustrating the third embodiment of the steering wheel hands-off detection method of the present invention.

[0119] Considering that most vehicles are now equipped with Lane Keeping Assist (LKA), when LKA is activated and the driver's hands are off the steering wheel, the EPS system (especially Dual Pinion Electric Power Steering (DP-EPS) and Rack Electric Power Steering (R-EPS), as the power steering mechanism of DP-EPS and R-EPS is located on the steering gear) executes the LKA angle or torque request. Because the vehicle's steering column, intermediate shaft, and steering wheel all have inertia, the torque and angle sensors will detect the torque value and torque change rate, which may cause the EPS system to fail to accurately detect whether the driver has taken their hands off the steering wheel. Therefore, if... Figure 4 As shown, in this embodiment, in order to improve the accuracy of EPS system recognition, the EPS system can also calibrate and self-learn the suddenly requested LKA request through the initial detection model to prevent the influence of inertia on the detection results. The specific process is as follows: before the above step S10, it also includes:

[0120] Step S05: When the EPS system is in the lane keeping function enabled state, determine the typical requested torque and angle and the atypical requested torque and angle from each requested torque and angle;

[0121] Step S06: Determine the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the steering wheel to be tested under the typical requested torque and angle;

[0122] Step S07: Calibrate the initial detection model based on the standard preset hand force threshold and the standard preset hand force change rate threshold.

[0123] It should be noted that the requested torque and angle mentioned above can be the requested torque and angle for lane keeping detected by the EPS system when the LKA function is enabled. Different steering wheel inertia correspond to different requested torque and angle. In specific implementation, it can also be consistent with the degree of road bumps, and each requested torque and angle can be divided according to a certain gradient. Typical requested torque and typical requested angle are selected from them, and the standard hand force threshold and standard hand force change rate threshold corresponding to the typical requested torque and typical requested angle are determined. The initial detection model is then calibrated so that the confidence value of the output of the initial detection model is 0 to 4 or 11 to 15.

[0124] Step S08: The non-typical requested torque and angle are self-learned using the calibrated initial detection model to obtain the preset detection model.

[0125] Understandably, after calibration, the preset hand force threshold and preset hand force change rate threshold corresponding to atypical request torque and atypical request angle are learned through self-learning, thereby obtaining the above-mentioned preset detection model.

[0126] It is important to emphasize that during self-learning, the learning hand force value and the rate of change of learning hand force under atypical request torque and atypical request angle can also be filtered to obtain the filtered learning hand force value and the filtered learning hand force rate of change. Then, based on the filtered learning hand force value and the filtered learning hand force rate of change, the atypical request torque and atypical request angle are equivalent to a typical request torque and typical request angle. The typical request torque and typical request angle are used as the target typical request torque and typical request angle for self-learning until all the output current confidence values ​​are in the range of 0 to 4 or 11 to 15, and the preset detection model can be obtained.

[0127] Furthermore, during actual testing, the EPS system can determine whether the LKA function is enabled. If so, the current hand force value and the current hand force change rate are filtered. Based on the filtered current hand force value and the current hand force change rate, the equivalent typical requested torque and typical requested angle under the requested torque and requested angle are determined, and then compared to determine whether to release the hand. This can prevent the influence of the inertia of the vehicle's steering column, intermediate shaft, and steering wheel, and improve the accuracy of the test.

[0128] It should also be emphasized that if vehicle speed, road surface roughness, and LKA are considered together, the specific calibration and self-learning process is similar to the process described above. For ease of understanding, please refer to [link / reference]. Figure 5 , Figure 5 This is a diagram showing the interaction between the EPS system and related systems in the steering wheel hands-off detection method of the present invention, as shown below. Figure 5 As shown, the EPS system can obtain the current vehicle speed (i.e., speed measurement system) from the speed measurement system. Figure 5 At medium vehicle speed, the hand force value and the rate of change of hand force are obtained from the torque sensor and angle sensor. Figure 5 The driver's hand force (torque value / time) is calibrated and self-learned for different vehicle speed scenarios; the current rack force (i.e., torque value / time) is obtained from the torque sensor. Figure 5 The system calibrates and learns under different road surface roughness levels using the medium road load (rack force); it obtains the steering wheel inertia from the torque and angle sensors, and retrieves the requested torque and angle (i.e., ...) from the LKA. Figure 5The LKA torque / angle request (torque value / time) is used to calibrate and self-learn for different torques and angles. The specific process will not be described in detail in this embodiment.

[0129] Reference Figure 6 , Figure 6 This is a structural block diagram of the first embodiment of the EPS system of the present invention.

[0130] like Figure 6 As shown, the EPS system proposed in this embodiment of the invention includes: a signal acquisition module 601, a hand force determination module 602, a hand force judgment module 603, and a state judgment module 604;

[0131] The signal acquisition module 601 is used to acquire the torque signal and angle signal of the steering wheel to be detected;

[0132] The hand force determination module 602 is used to determine the current hand force value and the current hand force change rate applied to the steering wheel to be detected based on the torque signal and the angle signal;

[0133] The hand force judgment module 603 is used to determine whether the current hand force value is less than a preset hand force threshold and whether the current hand force change rate is less than a preset hand force change rate threshold. The preset hand force threshold and the preset hand force change rate threshold are determined based on the preset detection model.

[0134] The state determination module 604 is used to determine, if yes, that the steering wheel to be detected is currently in a hands-free state.

[0135] This embodiment can pre-calibrate the initial detection model according to preset hand force thresholds and preset hand force change rate thresholds corresponding to different vehicle speeds and different road surface bump levels, thus obtaining a preset detection model. In actual use, the EPS system can acquire the torque signal generated by the torque sensor and the angle signal generated by the angle sensor. Based on the torque signal, it obtains the current hand force value applied by the driver to the steering wheel to be tested, and based on the angle signal, it obtains the current rotation angle of the steering wheel to be tested. Based on the current rotation angle and the current hand force value, it determines the current hand force change rate, and then based on the preset detection model, it determines the preset hand force value and the preset change rate threshold. It then determines whether the current hand force value is less than the preset hand force threshold and whether the current hand force change rate is less than the preset hand force change rate threshold. If so, it is determined to be a hands-off state. Compared with the existing method that requires the installation of a capacitor-loaded steering wheel, this embodiment can directly realize hands-off detection based on the EPS system, thereby reducing production costs.

[0136] In one implementation, the hand force determination module 602 is further configured to determine the current hand force value applied to the steering wheel to be tested based on the torque signal; and to determine the current hand force change rate applied to the steering wheel to be tested based on the current hand force value and the angle signal.

[0137] As one implementation, the hand force judgment module 603 is further configured to determine the current rack force of the steering wheel to be detected based on the current hand force value; determine the road bump level based on the current rack force, and determine the preset hand force threshold and preset hand force change rate threshold corresponding to the road bump level based on the preset detection model; and determine whether the current hand force value is less than the preset hand force threshold and whether the current hand force change rate is less than the preset hand force change rate threshold.

[0138] Based on the first embodiment of the EPS system of the present invention described above, a second embodiment of the EPS system of the present invention is proposed.

[0139] The signal acquisition module 601 is further configured to determine the typical road bump level and the atypical road bump level from various road bump levels; determine the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the steering wheel to be detected under the typical road bump level; calibrate the initial detection model based on the standard preset hand force threshold and the standard preset hand force change rate threshold; and perform self-learning on the atypical road bump level through the calibrated initial detection model to obtain the preset detection model.

[0140] As one implementation, the signal acquisition module 601 is further configured to acquire the learning hand force value and the learning hand force change rate of the steering wheel under the atypical road bump level; filter the learning hand force value and the learning hand force change rate; determine the target typical road bump level corresponding to the atypical road bump level based on the filtered learning hand force value and the filtered learning hand force change rate; and perform self-learning on the initial detection model through the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the target typical road bump level to obtain the preset detection model.

[0141] As one implementation, the signal acquisition module 601 is further configured to perform self-learning on the initial detection model using the standard preset hand force threshold corresponding to the typical road surface bumpiness, the filtered hand force value to be learned, the standard preset hand force change rate threshold, and the filtered hand force change rate to be learned; obtain the current confidence value based on the learning result, and use the self-learned initial detection model as the preset detection model when the current confidence value is within a preset confidence interval.

[0142] Based on the above embodiments of the EPS system of the present invention, a third embodiment of the EPS system of the present invention is proposed.

[0143] In this embodiment, the signal acquisition module 601 is further configured to, when the EPS system is in the lane keeping function enabled state, determine typical and atypical requested torques and angles and atypical requested torques and angles from each requested torque and angle; determine the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the steering wheel to be detected under the typical requested torque and angle; calibrate the initial detection model based on the standard preset hand force threshold and the standard preset hand force change rate threshold; and perform self-learning on the atypical requested torques and angles through the calibrated initial detection model to obtain the preset detection model.

[0144] Other embodiments or specific implementations of the EPS system of the present invention can be referred to the above-described method embodiments, and will not be repeated here.

[0145] Furthermore, embodiments of the present invention also propose an automobile that includes the EPS system as described above.

[0146] Other embodiments or specific implementations of the vehicle of the present invention can be referred to the above system embodiments, and will not be repeated here.

[0147] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0148] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0149] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory / random access memory, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0150] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for detecting steering wheel release from hands, characterized in that, The method is applied to an EPS system with a preset detection model, and the method includes the following steps: Acquire the torque and angle signals of the steering wheel to be tested; The current hand force applied to the steering wheel to be detected and the current rate of change of hand force are determined based on the torque signal and the angle signal. Determine whether the current hand force value is less than a preset hand force threshold and whether the current hand force change rate is less than a preset hand force change rate threshold. The preset hand force threshold and the preset hand force change rate threshold are determined based on the preset detection model. If so, it is determined that the steering wheel to be tested is currently in a hands-free state; Before the step of acquiring the torque signal and angle signal of the steering wheel to be detected, the method further includes: Determine the typical road surface bump level and the atypical road surface bump level from the various road surface bump levels; Determine the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the steering wheel to be tested under the typical road surface bumpiness; The initial detection model is calibrated based on the standard preset hand force threshold and the standard preset hand force change rate threshold. The preset detection model is obtained by self-learning the degree of bumpiness of the atypical road surface through the calibrated initial detection model.

2. The steering wheel hands-off detection method as described in claim 1, characterized in that, The step of determining the current hand force applied to the steering wheel to be detected and the current rate of change of hand force based on the torque signal and the angle signal includes: The current hand force applied to the steering wheel to be tested is determined based on the torque signal; The rate of change of the current hand force applied to the steering wheel to be detected is determined based on the current hand force value and the angle signal.

3. The steering wheel hands-off detection method as described in claim 2, characterized in that, Before the steps of determining whether the current hand force value is less than a preset hand force threshold and whether the current hand force change rate is less than a preset hand force change rate threshold, the method further includes: The current rack force of the steering wheel to be tested is determined based on the current hand force value; The degree of road bumpiness is determined based on the current rack force, and the preset hand force threshold and preset hand force change rate threshold corresponding to the degree of road bumpiness are determined based on the preset detection model.

4. The steering wheel hands-off detection method as described in claim 1, characterized in that, The step of self-learning the atypical road surface bumpiness level using the calibrated initial detection model to obtain the preset detection model includes: The learning hand force value and the learning hand force change rate of the steering wheel under the atypical road surface bump level are obtained; The hand force value to be learned and the rate of change of the hand force to be learned are filtered; The target typical road bumpiness degree corresponding to the atypical road bumpiness degree is determined based on the filtered learning hand force value and the rate of change of the filtered learning hand force. The initial detection model is self-learned by using the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the typical road surface bumpiness level to obtain the preset detection model.

5. The steering wheel hands-off detection method as described in claim 4, characterized in that, The step of self-learning the initial detection model by using the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the typical road surface bumpiness level to obtain the preset detection model includes: The initial detection model is self-learned using the standard preset hand force threshold corresponding to the typical road surface bumpiness, the filtered hand force value to be learned, the standard preset hand force change rate threshold, and the filtered hand force change rate to be learned. The current confidence value is obtained based on the learning results, and when the current confidence value is within a preset confidence interval, the initial detection model after self-learning is used as the preset detection model.

6. The steering wheel hands-off detection method as described in claim 1, characterized in that, Before the step of acquiring the torque signal and angle signal of the steering wheel to be detected, the method further includes: When the EPS system is in lane keeping function enabled, typical and atypical requested torque and angle are determined from each requested torque and angle. Determine the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the steering wheel to be tested under the typical requested torque and angle; The initial detection model is calibrated based on the standard preset hand force threshold and the standard preset hand force change rate threshold. The preset detection model is obtained by self-learning the atypical requested torque and angle through the calibrated initial detection model.

7. An EPS system, characterized in that, The EPS system includes: a signal acquisition module, a hand force determination module, a hand force judgment module, and a status judgment module; The signal acquisition module is used to acquire the torque signal and angle signal of the steering wheel to be detected; The hand force determination module is used to determine the current hand force value and the current hand force change rate applied to the steering wheel to be detected based on the torque signal and the angle signal. The hand force judgment module is used to determine whether the current hand force value is less than a preset hand force threshold and whether the current hand force change rate is less than a preset hand force change rate threshold. The preset hand force threshold and the preset hand force change rate threshold are determined based on a preset detection model. The state determination module is used to determine, if yes, that the steering wheel to be detected is currently in a hands-free state. The signal acquisition module is further configured to determine the typical road bump level and the atypical road bump level from various road bump levels; determine the standard preset hand force threshold and the standard preset hand force change rate threshold corresponding to the steering wheel to be tested under the typical road bump level; calibrate the initial detection model based on the standard preset hand force threshold and the standard preset hand force change rate threshold; and perform self-learning on the atypical road bump level using the calibrated initial detection model to obtain the preset detection model.

8. The EPS system as described in claim 7, characterized in that, The hand force determination module is also used to determine the current hand force value applied to the steering wheel to be detected based on the torque signal; The hand force determination module is also used to determine the rate of change of the current hand force applied to the steering wheel to be detected based on the current hand force value and the angle signal.

9. A car, characterized in that, The vehicle includes the EPS system as described in claim 7 or 8.

Images