Zero self-learning method of vehicle steering system software and control equipment

Abstract

The invention provides a zero-position self-learning method and control equipment for vehicle steering system software, and relates to the technical field of vehicle control, and the method comprises the steps: determining whether a target vehicle meets a preset self-learning enabling condition or not at the current moment through obtaining operation data of the target vehicle in a preset time period before the current time; only when the current moment meets the preset self-learning enabling condition, the vehicle speed signal is segmented, at least one target vehicle speed section is determined, a corner signal corresponding to each target vehicle speed section is determined, and the target steering wheel zero angle of each target vehicle speed section is determined by combining the historical steering wheel zero angle of each target vehicle speed section. Vehicle operation data are independently analyzed according to vehicle speed sections, the steering wheel zero angle is automatically updated when specific working conditions are met, the accuracy of the target steering wheel zero angle is improved, steering characteristic differences at different driving speeds can be adapted, zero deviation caused by long-term use is effectively eliminated, and the driving safety is improved. And the control precision and stability of the steering system are improved.

Description

Technical Field

[0001] This invention relates to the field of vehicle control technology, and more specifically, to a zero-position self-learning method and control device for vehicle steering system software. Background Technology

[0002] During long-term vehicle use, various practical operating conditions, such as dynamic changes in tire pressure, load increases and decreases, and uneven load distribution, can cause deviations between the software and mechanical zero points of the steering system, preventing them from maintaining a synchronized state. Specifically, when the vehicle is stationary and the front wheels are aligned, the steering wheel needs to be turned to the right by a certain angle to match the mechanical zero point; however, when fully loaded, to maintain a straight line, the steering wheel needs to be turned to the left by a certain angle to compensate for the steering offset caused by load changes. This zero-point deviation directly affects the accuracy of lateral control, becoming a key issue restricting the lateral control performance of autonomous driving and advanced driver assistance systems.

[0003] In existing technologies, steering wheel zero-position calibration mainly employs zero-position self-learning algorithms. The core logic involves collecting sensor data such as steering angle, vehicle speed, and lateral acceleration during vehicle operation, and automatically correcting the software zero position based on preset thresholds or statistical models. However, this type of algorithm has a key flaw in commercial applications: it lacks a complete self-learning condition determination mechanism, initiating the zero-position learning process only when basic driving conditions are met, and lacks in-depth verification of steering stability. Specifically, when the vehicle is in a straight line and at a stable speed, the steering wheel may fluctuate frequently within a small range due to slight road bumps, mechanical clearances in the steering system, etc. At this time, the steering state has not met the conditions required for stable calibration, but existing algorithms do not identify or filter such unstable states, directly initiating learning and misjudging the fluctuation signal as a valid steering deviation, thus continuously adjusting the zero-position learning value, leading to repeated oscillations in the learning value and a significant decrease in prediction accuracy. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of the prior art by providing a zero-position self-learning method and control device for vehicle steering system software. This method involves acquiring the operating data of the target vehicle over a preset time period prior to the current time, determining whether the target vehicle meets preset self-learning enable conditions at the current moment, and segmenting the vehicle speed signal to determine at least one target vehicle speed segment. It also determines the steering angle signal corresponding to each target vehicle speed segment and, in conjunction with the historical steering wheel zero-position angle of each target vehicle speed segment, determines the target steering wheel zero-position angle for each target vehicle speed segment.

[0005] To achieve the above objectives, the technical solutions adopted in the embodiments of this application are as follows: In a first aspect, embodiments of this application provide a zero-position self-learning method for vehicle steering system software, the method comprising: Obtain the target vehicle's operating data for a preset time period prior to the current moment, wherein the operating data includes: vehicle speed signal and corresponding steering wheel angle signal; Obtain the historical steering wheel zero angle for multiple preset vehicle speed ranges that are stored in advance for the steering system software in the target vehicle; Based on the operational data, it is determined whether the target vehicle meets the preset self-learning enable conditions at the current moment; If the current time satisfies the preset self-learning enable condition, the vehicle speed signal is segmented according to the vehicle speed range of the multiple preset vehicle speed segments to obtain at least one target vehicle speed segment corresponding to the vehicle speed signal. Determine the angle signal corresponding to the at least one target vehicle speed segment from the steering wheel angle signal; The target steering wheel zero angle for each target speed segment is determined based on the steering angle signal corresponding to each target speed segment and the historical steering wheel zero angle for each target speed segment. Update the historical steering wheel zero angle for each target vehicle speed segment to the target steering wheel zero angle.

[0006] In an optional implementation, the operating data further includes: multiple detection signals corresponding to the vehicle speed signal, lane line data of the current driving vehicle identified by the camera perception system, driving control data, and lateral function status information; The step of determining whether the target vehicle meets the preset self-learning enable conditions at the current moment based on the operational data includes: Based on the multiple detection signals and the driving control data, the vehicle status of the target vehicle is detected; If the vehicle status detection result indicates that the target vehicle has no communication failure and is in a steady-state driving state, then the driving scenario detection is performed on the target vehicle based on the lane line data. If the driving scenario detection result indicates that the target vehicle is in a straight driving scenario, then driving behavior detection is performed on the target vehicle based on the driving control data; If the driving behavior detection result indicates that the target vehicle is not in a state of driver over-control, then lane parallelism detection is performed on the target vehicle based on the lane line data; If the lane parallel detection result indicates that the target vehicle is in a lane parallel driving state, then lateral function state detection is performed based on the lateral function state information. If the lateral function status detection result indicates that the lateral function detection has passed, then it is determined that the current moment meets the preset self-learning enable condition.

[0007] In an optional implementation, the plurality of detection signals further include: longitudinal acceleration signal, lateral acceleration signal and yaw rate signal acquired by the inertial navigation sensor; The step of detecting the vehicle status of the target vehicle based on the multiple detection signals and the driving control data includes: Based on the multiple detection signals and the driving control data, the communication status of the target vehicle is detected; If the target vehicle does not have a communication fault, then the target vehicle is determined to be in a steady-state driving state based on the longitudinal acceleration value of the longitudinal acceleration signal, the lateral acceleration value of the lateral acceleration signal, and the yaw rate value of the yaw rate signal.

[0008] In an optional implementation, the lane line data includes: the polynomial coefficients of the target vehicle's current driving road, and a confidence level; the step of detecting the driving scene of the target vehicle based on the lane line data includes: Based on the polynomial coefficients of the current driving road, the current driving road is determined to be a straight driving road; If the current driving road is a straight driving road, then based on the confidence level of the current driving road, determine whether the current driving road meets the preset lane line clear condition; If the current driving road meets the preset lane line clear condition, then the current driving scenario of the target vehicle is determined to be a straight driving scenario with clear lane lines.

[0009] In an optional implementation, the driving control data includes: hand torque, steering wheel speed, and steering wheel angle; the step of detecting driving behavior of the target vehicle based on the driving control data includes: Based on the hand torque and the steering wheel speed, determine whether the target vehicle is under driver control. If the target vehicle is not under driver control, then it is determined that the target vehicle is not under driver over-control. If the target vehicle is under driver control, then the driver determines whether the target vehicle has entered a turning mode based on the hand torque and the steering wheel rotation angle. If the duration for which the target vehicle switches to the steering wheel centering state after entering the turning mode is less than or equal to a preset duration, then the target vehicle is determined to be in a state of driver over-control. If the duration of the target vehicle switching to the steering wheel centering state is greater than the preset duration, then it is determined that the target vehicle is not in a state of driver over-control.

[0010] In an optional implementation, the step of performing lane parallelism detection on the target vehicle based on the lane line data includes: Based on the lane line data, determine the heading angle of the target vehicle relative to the lane line and the aiming deviation; Based on the heading angle and aiming deviation of the target vehicle relative to the lane line, determine whether the target vehicle is parallel to the center line of the road; If the target vehicle is parallel to the center line of the road, then the target vehicle is determined to be in a parallel driving state in the lane.

[0011] In an optional implementation, the step of detecting the lateral functional state based on the lateral functional state information includes: If the lateral function status information indicates that the lateral function of the target vehicle is not activated, then the lateral function detection of the target vehicle is determined to be successful. If the lateral function status information indicates that the lateral function of the target vehicle has been activated and the activation time is greater than a preset duration, and the corresponding performance of the steering function actuator corresponding to the lateral function meets the preset performance constraints, then the lateral function detection of the target vehicle is determined to be successful.

[0012] In an optional implementation, determining the target steering wheel zero angle for each target speed segment based on the steering angle signal corresponding to each target speed segment and the historical steering wheel zero angle for each target speed segment includes: Within a preset calculation period, the steering angle signal corresponding to each target vehicle speed segment is sampled to obtain multiple sampled steering wheel angle values ​​within the current calculation period; Based on the multiple sampled steering wheel angle values, the average steering wheel angle value is calculated as the calculated steering wheel angle value for the current calculation cycle; The weighting coefficients are determined based on the calculated steering wheel angle value, the preset learning angle threshold, and the current zero-drift level. Based on the actual steering wheel zero-drift value for each target speed segment, the control input zero-drift value of the lateral function in the target vehicle, and the calculated steering angle value, an adjustment coefficient is determined. The target weight coefficient is determined based on the weighting coefficient and the adjustment coefficient. Based on the target weighting coefficient, the historical steering wheel zero angle and the calculated steering angle value are weighted and calculated to obtain the target steering wheel zero angle for each target speed segment.

[0013] In an optional implementation, before determining the weighting coefficient based on the calculated steering wheel angle value, the preset learning angle threshold, and the current zero-drift level, the method further includes: If the lateral function of the target vehicle is active, the current zero-drift level is determined based on the historical steering wheel zero angle. If the lateral function of the target vehicle is not activated, the preset minimum zero-drift level is determined as the current zero-drift level.

[0014] Secondly, embodiments of this application also provide a zero-position self-learning device for vehicle steering system software, the device comprising: The acquisition module is used to acquire the operating data of the target vehicle for a preset time period before the current moment, wherein the operating data includes: vehicle speed signal and corresponding steering wheel angle signal; The acquisition module is also used to acquire historical steering wheel zero angles for multiple preset vehicle speed ranges stored in advance for the steering system software in the target vehicle. The judgment module is used to determine, based on the running data, whether the target vehicle meets the preset self-learning enable conditions at the current moment; The segmentation module is used to segment the vehicle speed signal according to the vehicle speed range of the multiple preset vehicle speed segments if the current time satisfies the preset self-learning enable condition, so as to obtain at least one target vehicle speed segment corresponding to the vehicle speed signal. The determining module is used to determine the angle signal corresponding to the at least one target vehicle speed segment from the steering wheel angle signal; The determining module is further configured to determine the target steering wheel zero angle for each target speed segment based on the turning angle signal corresponding to each target speed segment and the historical steering wheel zero angle for each target speed segment. The update module is used to update the historical steering wheel zero angle of each target vehicle speed segment to the target steering wheel zero angle.

[0015] Thirdly, embodiments of this application also provide a control device, including: a processor, a storage medium, and a bus. The storage medium stores program instructions executable by the processor. When the control device is running, the processor communicates with the storage medium via the bus, and the processor executes the program instructions to perform the steps of the zero-position self-learning method for vehicle steering system software as described in any of the first aspects.

[0016] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the zero-position self-learning method for vehicle steering system software as described in any of the first aspects.

[0017] The beneficial effects of this application are: This application provides a zero-position self-learning method and control device for vehicle steering system software. The method includes: acquiring operating data of a target vehicle for a preset time period prior to the current moment, wherein the operating data includes: vehicle speed signal and corresponding steering wheel angle signal; acquiring pre-stored historical steering wheel zero-position angles for multiple preset vehicle speed segments for the steering system software in the target vehicle; determining, based on the operating data, whether the target vehicle meets preset self-learning enabling conditions at the current moment; if the preset self-learning enabling conditions are met at the current moment, segmenting the vehicle speed signal according to the vehicle speed range of the multiple preset vehicle speed segments to obtain at least one target vehicle speed segment corresponding to the vehicle speed signal; determining the steering wheel angle signal corresponding to at least one target vehicle speed segment from the steering wheel angle signal; determining the target steering wheel zero-position angle for each target vehicle speed segment based on the steering wheel angle signal corresponding to each target vehicle speed segment and the historical steering wheel zero-position angle for each target vehicle speed segment; and updating the historical steering wheel zero-position angle for each target vehicle speed segment to the target steering wheel zero-position angle.

[0018] The method of this application obtains the operating data of the target vehicle for a preset time period before the current time, determines whether the target vehicle meets the preset self-learning enable conditions at the current time, and only if the preset self-learning enable conditions are met at the current time, segments the vehicle speed signal to determine at least one target vehicle speed segment, and determines the steering angle signal corresponding to each target vehicle speed segment. Combining the historical steering wheel zero angle of each target vehicle speed segment, the target steering wheel zero angle of each target vehicle speed segment is determined. By independently analyzing the vehicle operating data by vehicle speed segment, the steering wheel zero angle is automatically updated when specific operating conditions are met. This not only improves the accuracy of the target steering wheel zero angle, but also adapts to the differences in steering characteristics at different driving speeds, effectively eliminates zero deviation caused by long-term use, improves the control accuracy and stability of the steering system, reduces the need for manual calibration, and enhances the safety and driving comfort of vehicle steering. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 One of the flowcharts for a zero-position self-learning method for vehicle steering system software provided in this application embodiment; Figure 2 A second schematic flowchart illustrating a zero-position self-learning method for vehicle steering system software provided in an embodiment of this application; Figure 3A third schematic flowchart illustrating a zero-position self-learning method for vehicle steering system software provided in this application embodiment; Figure 4 A fourth flowchart illustrating a zero-position self-learning method for vehicle steering system software provided in this application embodiment; Figure 5 Fifth flowchart illustrating a zero-position self-learning method for vehicle steering system software provided in this application embodiment; Figure 6 A flowchart illustrating a zero-position self-learning method for vehicle steering system software provided in this application embodiment is shown in Figure 6. Figure 7 The seventh flowchart illustrates a zero-position self-learning method for vehicle steering system software provided in this application embodiment; Figure 8 This is the eighth flowchart illustrating a zero-position self-learning method for vehicle steering system software provided in this application embodiment; Figure 9 A schematic diagram of the functional modules of a zero-position self-learning device for vehicle steering system software provided in an embodiment of this application; Figure 10 This is a schematic diagram of a control device provided in an embodiment of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0022] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0023] In the description of this application, it should be noted that if the terms "upper", "lower", etc. appear to indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship that the product of this application is usually placed in, it is only for the convenience of describing this application and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0024] Furthermore, the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Additionally, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] It should be noted that, where there is no conflict, the features in the embodiments of this application can be combined with each other.

[0026] To obtain a precise steering wheel zero-position angle, this application provides a zero-position self-learning method for vehicle steering system software. By acquiring the target vehicle's operating data over a preset time period prior to the current time, it determines whether the target vehicle meets preset self-learning enable conditions at the current moment. Only when the preset self-learning enable conditions are met at the current moment is the vehicle speed signal segmented to determine at least one target vehicle speed segment, and the corresponding steering angle signal for each target vehicle speed segment is determined. Combining the historical steering wheel zero-position angle for each target vehicle speed segment, the target steering wheel zero-position angle for each target vehicle speed segment is determined. By independently analyzing vehicle operating data segment by segment, the steering wheel zero-position angle is automatically updated when specific operating conditions are met. This adapts to differences in steering characteristics at different driving speeds, effectively eliminates zero-position deviation caused by long-term use, improves the control accuracy and stability of the steering system, reduces the need for manual calibration, and enhances vehicle steering safety and driving comfort.

[0027] The zero-position self-learning method for vehicle steering system software provided in this application will be explained in detail below with reference to the accompanying drawings and specific examples. Figure 1 This is one of the flowcharts illustrating a zero-position self-learning method for vehicle steering system software provided in an embodiment of this application; as shown below. Figure 1 As shown, the method includes: S101. Obtain the operating data of the target vehicle for a preset time period before the current moment.

[0028] The operational data includes: vehicle speed signal and corresponding steering wheel angle signal.

[0029] In this embodiment, after the target vehicle is powered on, the system begins to collect the target vehicle's operating data and collects enough operating data for a preset time period. The operating data for the preset time period is used to determine whether the target vehicle meets the preset self-learning enable conditions at the current moment.

[0030] The running data for the preset time period consists of multiple frames of running data within the preset time period. Each frame of running data includes the vehicle speed signal and the corresponding steering wheel angle signal.

[0031] S102. Obtain the historical steering wheel zero angle for multiple preset speed ranges of the steering system software in the target vehicle, which are stored in advance.

[0032] Specifically, the steering system software in the target vehicle pre-stores the historical steering wheel zero angles for multiple preset vehicle speed ranges. This can be understood as the historical steering wheel zero angles for multiple preset vehicle speed ranges being either the steering wheel zero angles obtained through the target vehicle's historical self-learning or the steering wheel zero angles stored after the target vehicle's factory calibration. Each preset vehicle speed range corresponds to a historical steering zero angle.

[0033] S103. Based on the operating data, determine whether the target vehicle meets the preset self-learning enable conditions at the current moment.

[0034] S104. If the preset self-learning enable condition is met at the current moment, the vehicle speed signal is segmented according to the vehicle speed range of multiple preset vehicle speed segments to obtain at least one target vehicle speed segment corresponding to the vehicle speed signal.

[0035] Specifically, in order to ensure the accuracy of zero-point learning and avoid self-learning under complex operating conditions, the system determines whether the target vehicle meets the preset self-learning enable conditions at the current moment based on the collected operating data.

[0036] When it is determined that the target vehicle meets the preset self-learning enable conditions at the current moment, the vehicle speed signal is segmented according to the vehicle speed range of multiple preset vehicle speed segments to obtain at least one target vehicle speed segment corresponding to the vehicle speed signal.

[0037] For example, the speed ranges of multiple preset speed segments may include: a first speed segment, a second speed segment, a third speed segment, and a fourth speed segment. The first speed segment has an actual speed greater than 10 km / h and less than or equal to 26 km / h; the second speed segment has an actual speed greater than 26 km / h and less than or equal to 52 km / h; the third speed segment has an actual speed greater than 52 km / h and less than or equal to 78 km / h; and the fourth speed segment has a speed greater than 78 km / h. The speed signal is then segmented to obtain at least one target speed segment corresponding to the speed signal. For example, if the speed signal is 50 km / h, then the target speed segment corresponding to this speed signal is determined to be the second speed segment.

[0038] Because the steering resistance is high at low vehicle speeds, the self-learning value is inaccurate. Therefore, self-learning is not performed at low speeds. For example, if the vehicle speed signal is less than 10 km / h, the vehicle speed signal is not processed.

[0039] S105. Determine at least one steering wheel angle signal corresponding to a target vehicle speed range from the steering wheel angle signal.

[0040] Specifically, after determining at least one target speed segment, based on the speed signals in the at least one target speed segment, the steering wheel angle signal corresponding to each speed signal is determined as the angle signal corresponding to the at least one target speed segment.

[0041] S106. Determine the target steering wheel zero angle for each target speed segment based on the steering angle signal corresponding to each target speed segment and the historical steering wheel zero angle for each target speed segment.

[0042] S107. Update the historical steering wheel zero angle for each target speed segment to the target steering wheel zero angle.

[0043] Specifically, based on the steering angle signal corresponding to each target vehicle speed segment and the historical steering wheel zero-angle for each target vehicle speed segment, the steering wheel zero-angle for each target vehicle speed segment is self-learned to obtain the target steering wheel zero-angle for each target vehicle speed segment. Finally, the historical steering wheel zero-angle for each target vehicle speed segment is updated to the target steering wheel zero-angle. The updated target steering wheel zero-angle will serve as the benchmark for subsequent vehicle steering control, such as the calibration of the power steering curve of the electronic power steering system and the correction of steering commands of the autonomous driving system, thereby achieving dynamic optimization of zero-position parameters.

[0044] In summary, this application provides a zero-position self-learning method for vehicle steering system software. The method includes: acquiring operating data of a target vehicle over a preset time period prior to the current moment, wherein the operating data includes: vehicle speed signal and corresponding steering wheel angle signal; acquiring pre-stored historical steering wheel zero-position angles for multiple preset vehicle speed segments of the steering system software in the target vehicle; determining, based on the operating data, whether the target vehicle meets preset self-learning enabling conditions at the current moment; if the current moment meets the preset self-learning enabling conditions, segmenting the vehicle speed signal according to the speed range of the multiple preset vehicle speed segments to obtain at least one target vehicle speed segment corresponding to the vehicle speed signal; determining the steering wheel angle signal corresponding to at least one target vehicle speed segment from the steering wheel angle signal; determining the target steering wheel zero-position angle for each target vehicle speed segment based on the steering wheel angle signal corresponding to each target vehicle speed segment and the historical steering wheel zero-position angle of each target vehicle speed segment; and updating the historical steering wheel zero-position angle of each target vehicle speed segment to the target steering wheel zero-position angle.

[0045] The method of this application obtains the operating data of the target vehicle for a preset time period before the current time, determines whether the target vehicle meets the preset self-learning enable conditions at the current time, and only if the preset self-learning enable conditions are met at the current time, segments the vehicle speed signal to determine at least one target vehicle speed segment, and determines the steering angle signal corresponding to each target vehicle speed segment. Combining the historical steering wheel zero angle of each target vehicle speed segment, the target steering wheel zero angle of each target vehicle speed segment is determined. By independently analyzing the vehicle operating data by vehicle speed segment, the steering wheel zero angle is automatically updated when specific operating conditions are met. This not only improves the accuracy of the target steering wheel zero angle, but also adapts to the differences in steering characteristics at different driving speeds, effectively eliminates zero deviation caused by long-term use, improves the control accuracy and stability of the steering system, reduces the need for manual calibration, and enhances the safety and driving comfort of vehicle steering.

[0046] This application also provides another possible implementation of a zero-position self-learning method for vehicle steering system software. The operating data also includes: multiple detection signals corresponding to the vehicle speed signal, lane line data of the current driving vehicle identified by the camera perception system, driving control data, and lateral function status information. Figure 2 This is a second flowchart illustrating a zero-position self-learning method for vehicle steering system software provided in an embodiment of this application. Figure 2 As shown, based on the operational data, it is determined whether the target vehicle meets the preset self-learning enable conditions at the current moment, including: S201. Based on multiple detection signals and driving control data, perform vehicle status detection on the target vehicle.

[0047] S202. If the vehicle status detection result indicates that the target vehicle has no communication failure and is in a steady-state driving state, then the driving scenario detection of the target vehicle is performed based on the lane line data.

[0048] S203. If the driving scenario detection result indicates that the target vehicle is in a straight driving scenario, then the driving behavior of the target vehicle shall be detected based on the driving control data.

[0049] S204. If the driving behavior detection result indicates that the target vehicle is not in a state of driver over-control, then lane parallel detection is performed on the target vehicle based on the lane line data.

[0050] S205. If the lane parallel detection result indicates that the target vehicle is in a lane parallel driving state, then perform lateral function status detection based on the lateral function status information.

[0051] S206. If the horizontal function status detection result indicates that the horizontal function detection has passed, then it is determined that the preset self-learning enable condition is met at the current moment.

[0052] In this embodiment, a vehicle speed signal is obtained from the chassis system or the vehicle controller, as well as multiple detection signals corresponding to the vehicle speed signal. Based on the multiple detection signals and driving control data, the state of the target vehicle is detected to determine whether the target vehicle has a communication fault. If it is determined that the target vehicle has a communication fault, the steering wheel zero-position self-learning is stopped.

[0053] If it is determined that the target vehicle does not have a communication fault, further analysis is performed based on multiple detection data to determine whether the target vehicle is in a steady-state driving state, thus obtaining a vehicle state detection result. Only if the vehicle state detection result indicates that the target vehicle has no communication fault and is in a steady-state driving state at the current moment, driving scenario detection is performed on the target vehicle based on lane line data, thus obtaining a driving scenario detection result. If the vehicle state detection result indicates that the target vehicle has a communication fault or is not in a steady-state driving state at the current moment, then the steering wheel zero-position self-learning is stopped.

[0054] Driving scenario detection is performed on the target vehicle to determine whether it is currently in a straight-line driving scenario. If the driving scenario detection result indicates that the target vehicle is currently in a straight-line driving scenario, then driving behavior detection is performed on the target vehicle based on driving control data. If the driving scenario detection result indicates that the target vehicle is not currently in a straight-line driving scenario, then the steering wheel zero-position self-learning is stopped.

[0055] Driving behavior detection is performed on the target vehicle to determine whether the vehicle is currently in a state of driver over-control. If the driving behavior detection result indicates that the target vehicle is not currently in a state of driver over-control, then lane parallelism detection is performed on the target vehicle based on lane line data. If the driving behavior detection result indicates that the target vehicle is currently in a state of driver over-control, then the steering wheel zero-position self-learning is stopped.

[0056] Lane parallelism detection is performed on the target vehicle to determine whether it is currently in a lane-parallel driving state. If the lane parallelism detection result indicates that the target vehicle is in a lane-parallel driving state, then lateral function status detection is performed based on the lateral function status information. If the lane parallelism detection result indicates that the target vehicle is not in a lane-parallel driving state, then steering wheel zero-position self-learning is stopped.

[0057] Lateral function status information is used to perform lateral function status detection to determine whether the target vehicle's lateral function has passed the detection at the current moment. If the lateral function status detection result indicates that the lateral function has passed, then the target vehicle meets the preset self-learning enable conditions at the current moment, and steering wheel zero-position self-learning can begin. If the lateral function status detection result indicates that the lateral function has failed, then steering wheel zero-position self-learning stops.

[0058] The method provided in this application constructs a multi-dimensional, progressive self-learning enable condition judgment logic by introducing multiple detection signals, lane line data, driving control data, and lateral functional state information. First, vehicle state detection ensures the system is fault-free and driving stably. Then, lane line data confirms the straight-line driving scenario, driving control data eliminates driver overtaking interference, lane parallelism detection ensures driving trajectory compliance, and finally, lateral functional state detection verifies system adaptability. This process significantly improves the accuracy and reliability of self-learning trigger conditions, effectively avoiding erroneous learning in complex scenarios, fault states, or under human intervention. It ensures that zero-position self-learning only occurs when the vehicle is in a stable, realistic straight-line driving condition, thereby further improving the accuracy of the steering system's zero-position angle and enhancing the safety and stability of vehicle steering control.

[0059] This application also provides another possible implementation of a zero-position self-learning method for vehicle steering system software, wherein the multiple detection signals include: longitudinal acceleration signal, lateral acceleration signal and yaw rate signal collected by inertial navigation sensor. Figure 3 This is the third flowchart illustrating a zero-position self-learning method for vehicle steering system software provided in this application embodiment. Figure 3 As shown, based on multiple detection signals and driving control data, vehicle status detection is performed on the target vehicle, including: S301. Based on multiple detection signals and driving control data, perform communication status detection on the target vehicle.

[0060] S302. If the target vehicle does not have a communication fault, determine whether the target vehicle is in a steady-state driving state based on the longitudinal acceleration value of the longitudinal acceleration signal, the lateral acceleration value of the lateral acceleration signal, and the yaw rate value of the yaw rate signal.

[0061] In this embodiment, the inertial navigation sensor is an inertial measurement unit (IMU), which acquires longitudinal acceleration signals, lateral acceleration signals, and yaw rate signals. Communication status detection is then performed on the longitudinal acceleration signals, lateral acceleration signals, yaw rate signals, and vehicle speed signals, respectively.

[0062] Specifically, each longitudinal acceleration signal, lateral acceleration signal, yaw rate signal, and vehicle speed signal includes a specific value and whether it is invalid. Then, each longitudinal acceleration signal, lateral acceleration signal, yaw rate signal, and vehicle speed signal is checked to see if it is invalid and whether the duration is greater than 200ms.

[0063] Optionally, the driving control data includes: hand torque, steering wheel speed, and steering wheel angle. The driving control data is obtained from the Steering Angle Sensor (SAS). Each hand torque, steering wheel speed, and steering wheel angle includes a specific value and whether it is in an invalid state. The system detects whether each hand torque, steering wheel speed, and steering wheel angle is in an invalid state and whether the duration of this invalid state is greater than 200ms. If each hand torque, steering wheel speed, and steering wheel angle is not in an invalid state, or the duration of the invalid state is less than or equal to 200ms, and each longitudinal acceleration signal, lateral acceleration signal, yaw rate signal, and vehicle speed signal is not in an invalid state, or the duration of this invalid state is less than or equal to 200ms, then it is preliminarily determined that the target vehicle does not have a communication fault.

[0064] Additionally, it is necessary to confirm whether each hand torque, steering wheel speed, steering wheel rotation angle, longitudinal acceleration signal, lateral acceleration signal, yaw rate signal, and vehicle speed signal has been lost. If no message loss has occurred, it is determined that the target vehicle does not have a communication fault.

[0065] Specifically, if the target vehicle does not have a communication fault, the longitudinal acceleration value is compared with the preset longitudinal acceleration threshold, the lateral acceleration value is compared with the preset lateral acceleration threshold, and the yaw rate value is compared with the preset yaw rate threshold to determine whether the target vehicle is in a steady-state driving state.

[0066] The preset longitudinal acceleration threshold can be set to 2 m / s². 2 The preset lateral acceleration threshold can be set to 1 m / s². 2 The preset yaw rate threshold can be set to 0.02 rad / s. If the longitudinal acceleration value is less than or equal to 2 m / s², then... 2 Lateral acceleration value less than or equal to 1 m / s² 2 If the lateral acceleration value is less than or equal to 0.02 rad / s, then the target vehicle is determined to be in a steady-state driving state.

[0067] The method provided in this application first detects the vehicle's communication status to ensure reliable data transmission, and then combines key parameters such as longitudinal acceleration, lateral acceleration, and yaw rate to determine whether the vehicle is in a steady-state driving state, thus constructing a hierarchical and progressive vehicle state detection logic. This not only eliminates invalid data interference caused by communication failures, but also accurately identifies whether the vehicle is in a stable driving condition without dynamic changes such as sudden acceleration, sudden deceleration, or sharp turns. This provides a reliable basic state guarantee for subsequent self-learning, avoiding the need to initiate self-learning when the vehicle is in a non-steady state or when data is abnormal, further improving the accuracy and effectiveness of zero-point self-learning.

[0068] This application also provides another possible implementation of a zero-position self-learning method for vehicle steering system software, wherein the lane line data includes: the polynomial coefficients of the target vehicle's current driving road, and the confidence level. Figure 4 This is the fourth flowchart illustrating a zero-position self-learning method for vehicle steering system software provided in this application embodiment. Figure 4 As shown, based on lane line data, driving scene detection is performed on the target vehicle, including: S401. Based on the polynomial coefficients of the current driving road, determine that the target vehicle's current driving road is a straight driving road.

[0069] S402. If the current driving road is a straight driving road, determine whether the current driving road meets the preset lane line clear condition based on the confidence level of the current driving road.

[0070] S403. If the current driving road meets the preset lane line clear condition, then the target vehicle's current driving scenario is determined to be a straight driving scenario with clear lane lines.

[0071] In this embodiment, lane line data is provided by a camera perception system. The road radius of the target vehicle's current driving road is determined based on the polynomial coefficients of the current driving road. The road radius of the target vehicle's current driving road is compared with a preset road radius threshold. If the preset road radius threshold is set to 10000m, the road radius is greater than or equal to 10000m, and the duration is greater than or equal to 200ms, then the target vehicle's current driving road is determined to be a straight driving road.

[0072] The confidence level of the current driving road includes the confidence level of the left lane line and the confidence level of the right lane line. If the current driving road is a straight driving road, the confidence level of the left lane line is compared with a preset confidence threshold for the left lane line, and the confidence level of the right lane line is compared with a preset confidence threshold for the right lane line. If the preset confidence threshold for the left lane line is 30%, and the preset confidence threshold for the right lane line is also 30%, then if the confidence level of the left lane line is greater than or equal to 30% and the duration is greater than or equal to 200ms, and the confidence level of the right lane line is greater than or equal to 30% and the duration is greater than or equal to 200ms, then the current driving road is determined to meet the preset lane line clarity condition, and the current driving scenario of the target vehicle is determined to be a straight driving scenario with clear lane lines.

[0073] The method provided in this application determines whether a road is straight by using lane line polynomial coefficients, and then verifies its clarity by combining lane line confidence scores, forming a progressive scene detection logic from road type to lane line quality. This not only ensures that the self-learning scenario is a straight road, meeting the basic requirements of zero-position learning for driving trajectories, but also eliminates cases of blurred lane lines and unreliable recognition through confidence score filtering. This provides an accurate road reference benchmark for subsequent zero-position angle calculations, effectively avoiding learning biases caused by curved roads or unclear lane lines, and further improving the scene adaptability and result accuracy of self-learning.

[0074] This application also provides another possible implementation of a zero-position self-learning method for vehicle steering system software. Figure 5 This is the fifth flowchart illustrating a zero-position self-learning method for vehicle steering system software provided in this application embodiment. Figure 5 As shown, based on driving control data, driving behavior detection is performed on the target vehicle, including: S501. Determine whether the target vehicle is under driver control based on hand torque and steering wheel speed.

[0075] S502. If the target vehicle is not under the driver's control, then it is determined that the target vehicle is not under the driver's control.

[0076] In this embodiment, if the hand torque is less than or equal to a preset first hand torque threshold and the duration is greater than or equal to 200ms, and the steering wheel speed is less than or equal to a preset steering wheel speed threshold and the duration is greater than or equal to 200ms, then it is determined that the target vehicle is not under driver control, and therefore the target vehicle is not under driver over-control. The preset first hand torque threshold and the preset steering wheel speed threshold are related to the weight of the target vehicle and increase with the increase of the target vehicle's weight.

[0077] S503. If the target vehicle is under the driver's control, determine whether the target vehicle has entered turning mode based on the hand torque and steering wheel rotation angle.

[0078] S504. If the duration for which the target vehicle switches to the steering wheel centering state after entering the turning mode is less than or equal to the preset duration, then the target vehicle is determined to be in a state of driver over-control.

[0079] S505. If the duration of the target vehicle switching to the steering wheel centering state is greater than the preset duration, it is determined that the target vehicle is not in a state of driver over-control.

[0080] Specifically, if the target vehicle is under driver control, it means that the hand torque is greater than the preset first hand torque threshold, or the steering wheel speed is greater than the preset steering wheel speed threshold. Based on the hand torque and steering wheel rotation angle, it is determined whether the target vehicle has entered turning mode.

[0081] If the hand torque is greater than or equal to a preset second hand torque threshold, and the steering wheel rotation angle is greater than or equal to a first steering wheel rotation angle threshold, then the target vehicle is determined to have entered turning mode. Turning mode includes U-turn mode or large-angle turn mode. If the hand torque is less than or equal to a preset third hand torque threshold, and the steering wheel rotation angle is less than or equal to a second steering wheel rotation angle threshold, then the target vehicle is determined to have switched to the steering wheel centering state after entering turning mode. If the duration of the centering state is less than or equal to a preset duration of 3 seconds, then the target vehicle is determined to still be under the driver's control.

[0082] If the duration of the return-to-center state is greater than the preset duration of 3 seconds, it is determined that the target vehicle is not in a state of driver over-control.

[0083] Specifically, the preset second hand torque threshold is greater than the preset third hand torque threshold, and the preset third hand torque threshold is greater than the preset first hand torque threshold. The preset second hand torque threshold can be set to 1.8 Nm, the preset third hand torque threshold can be set to 0.5 Nm, and the preset first hand torque threshold can be set to 0.1 Nm. The second steering wheel rotation angle threshold is less than the first steering wheel rotation angle threshold.

[0084] The method provided in this application constructs a multi-parameter linked driving behavior detection logic using driving control data such as hand torque, steering wheel speed, and steering angle. First, it determines whether driver control is present based on hand torque and speed. Then, for cases where control is present, it analyzes whether a turning mode has been entered and the duration of the subsequent centering state, combining hand torque and steering angle, thereby accurately distinguishing the driver's over-control state. This effectively eliminates interference from the driver's active steering (such as turning or correcting direction) on zero-position self-learning, allowing self-learning only when the driver has no control or the steering wheel has stably returned to center for a sufficient duration after control. This ensures the learning process is not affected by temporary human intervention, further improving the stability and accuracy of zero-position self-learning.

[0085] This application also provides another possible implementation of a zero-position self-learning method for vehicle steering system software. Figure 6 This is the sixth flowchart illustrating a zero-position self-learning method for vehicle steering system software provided in this application embodiment. Figure 6 As shown, based on lane line data, lane parallelism detection is performed on the target vehicle, including: S601. Based on the lane line data, determine the heading angle of the target vehicle relative to the lane line and the aiming deviation.

[0086] S602. Determine whether the target vehicle is parallel to the center line of the road based on the heading angle and aiming deviation of the target vehicle relative to the lane line.

[0087] S603. If the target vehicle is parallel to the center line of the road, then the target vehicle is determined to be in a parallel driving state in the lane.

[0088] In this embodiment, based on lane line data, the heading angle and aiming deviation of the target vehicle relative to the lane line are determined. If the heading angle is less than or equal to a preset heading angle threshold and the duration is greater than or equal to 200ms, and the aiming deviation is less than a preset aiming deviation, then the target vehicle is determined to be parallel to the road centerline, and the target vehicle is determined to be in a lane-parallel driving state. The preset heading angle threshold can be set to 0.01rad, and the preset aiming deviation can be set to 0.09m.

[0089] The formula for calculating the aiming deviation is as follows:

[0090] in, (i=1, 2, 3) are the coefficients of the cubic polynomial of the lane lines. This is represented as the aiming distance. Expressed as the yaw rate of the entire vehicle, This indicates the aiming time.

[0091] The method provided in this application calculates the vehicle's heading angle and aiming deviation relative to the lane line using lane line data to determine whether the vehicle is parallel to the road centerline, thus constructing a lane parallelism detection logic based on quantified parameters. This accurately identifies whether the vehicle is traveling smoothly and in a straight line along the lane, rather than deviating from the lane or traveling at an angle. It ensures that zero-position self-learning is performed under ideal conditions where the vehicle trajectory remains consistent with the lane line, avoiding zero-position angle calculation errors caused by vehicle deviation or trajectory tilt, further improving the accuracy of the self-learning scenario and the reliability of the results.

[0092] This application also provides another possible implementation of a zero-position self-learning method for vehicle steering system software. Figure 7 This is the seventh flowchart illustrating a zero-position self-learning method for vehicle steering system software provided in this application embodiment, as shown below. Figure 7 As shown, horizontal functional status detection is performed based on horizontal functional status information, including: S701. If the lateral function status information indicates that the lateral function of the target vehicle is not activated, then the lateral function detection of the target vehicle is determined to be successful.

[0093] S702. If the lateral function status information indicates that the lateral function of the target vehicle has been activated and the activation time is greater than the preset duration, and the corresponding performance of the steering function actuator corresponding to the lateral function meets the preset performance constraints, then the lateral function detection of the target vehicle is determined to be successful.

[0094] In this embodiment, lateral function status detection is performed based on lateral function status information to determine whether the lateral function of the target vehicle is activated. If the lateral function status information indicates that the lateral function of the target vehicle is not activated, the lateral function detection of the target vehicle is determined to be successful, and the steering wheel angle self-learning can be performed.

[0095] If the lateral function status information indicates that the lateral function of the target vehicle has been activated, and the activation time is less than or equal to the preset duration of 3 seconds, then the lateral function detection of the target vehicle is determined to have failed, and the steering wheel angle self-learning will not be performed.

[0096] If the lateral function status information indicates that the lateral function of the target vehicle has been activated and the activation time is greater than the preset duration of 3 seconds, then it is determined whether the corresponding performance of the steering function actuator corresponding to the lateral function meets the preset performance constraints. Specifically, the difference between the requested turning angle and the actual turning angle of the steering function actuator is calculated, and it is determined whether the difference is less than the preset response threshold. If it is determined that the difference is less than the preset response threshold, then it is determined that the corresponding performance of the steering function actuator meets the preset performance constraints, and the lateral function detection of the target vehicle is determined to be passed, and the steering wheel angle self-learning can be performed.

[0097] If the difference is determined to be greater than or equal to the preset response threshold, it is determined that the corresponding performance of the steering function actuator does not meet the preset performance constraints, the lateral function test of the target vehicle is determined to be failed, and the steering wheel angle self-learning is not performed.

[0098] The method provided in this application constructs a targeted horizontal function status detection logic by distinguishing whether the horizontal function is activated and its activated state. For cases where the horizontal function is not activated, the detection is directly confirmed as passed; for cases where it is activated, the activation duration and actuator performance constraints are verified to ensure the horizontal function is in a stable and reliable working state. This avoids misjudgments when the horizontal function is not activated and eliminates interference when the function is newly activated or the actuator performance is substandard, ensuring that zero-level self-learning is performed under conditions of horizontal function adaptation and stability, further improving the environmental adaptability and result accuracy of self-learning.

[0099] This application also provides another possible implementation of a zero-position self-learning method for vehicle steering system software. Figure 8 This is the eighth flowchart illustrating a zero-position self-learning method for vehicle steering system software provided in this application embodiment. Figure 8 As shown, based on the steering angle signal corresponding to each target vehicle speed segment and the historical steering wheel zero-position angle for each target vehicle speed segment, the target steering wheel zero-position angle for each target vehicle speed segment is determined, including: S801. Within a preset calculation period, sample the steering angle signal corresponding to each target vehicle speed segment to obtain multiple sampled steering wheel angle values ​​within the current calculation period.

[0100] S802. Calculate the average steering wheel angle value based on multiple sampled steering wheel angle values, and use it as the calculated steering wheel angle value for the current calculation cycle.

[0101] In this embodiment, each target speed segment has a corresponding current zero-drift level. A preset calculation period for each target speed segment is determined based on the current zero-drift level. For example, if the current zero-drift level of a target speed segment is low, then the preset calculation period for that target speed segment is 20 sampling points; if the current zero-drift level of a target speed segment is medium, then the preset calculation period for that target speed segment is 40 sampling points; and if the current zero-drift level of a target speed segment is high, then the preset calculation period for that target speed segment is 60 sampling points. Based on the preset calculation period, the steering angle signal corresponding to each target speed segment is sampled to obtain multiple sampled steering wheel angle values ​​within the current calculation period.

[0102] After obtaining multiple sampled steering wheel angle values ​​for the current calculation cycle, the average steering wheel angle value is calculated and used as the calculated steering wheel angle value for the current calculation cycle.

[0103] It should be noted that the multiple sampled steering wheel angle values ​​in the current calculation cycle can be obtained by continuous sampling or non-continuous sampling. The calculated steering wheel angle value will only be output after a sufficient number of steering wheel angle values ​​have been sampled.

[0104] S803. Determine the weighting coefficient based on the calculated steering wheel angle value, the preset learning angle threshold, and the current zero-drift level.

[0105] S804. Based on the actual steering wheel zero-drift value for each target speed segment, the control input zero-drift value of the lateral function in the target vehicle, and the calculated steering angle value, determine the adjustment coefficient.

[0106] S805. Determine the target weight coefficient based on the weighting coefficient and adjustment coefficient.

[0107] S806. Based on the target weighting coefficient, perform a weighted calculation on the historical steering wheel zero angle and the calculated steering angle value to obtain the target steering wheel zero angle for each target vehicle speed segment.

[0108] Specifically, based on the target weighting coefficient, the historical steering wheel zero-position angle and the calculated steering angle value of the current calculation cycle are weighted and calculated to obtain the steering wheel zero-position angle of the current calculation cycle. The calculation formula is expressed as follows:

[0109] in, This represents the steering wheel zero-position angle for the current calculation cycle. This represents the steering wheel zero-position angle of the previous calculation cycle. This represents the sequence number of the target speed range. i This is represented by the sequence number of the calculation period. Represented as weighting coefficients, This is represented as an adjustment factor. This represents the calculated steering angle value for the current calculation cycle. It can be understood that if the current calculation cycle is the first calculation cycle after the target vehicle is powered on, then the steering wheel zero angle of the previous calculation cycle is the historical steering wheel zero angle, until the steering wheel zero angle of the last calculation cycle is obtained, and this is used as the target steering wheel zero angle for the target vehicle speed segment, thus obtaining the target steering wheel zero angle for each target vehicle speed segment.

[0110] It should be noted that before performing weighted calculations on the historical steering wheel zero angle and the calculated steering angle value based on the target weight coefficient, it is necessary to determine whether the historical steering wheel zero angle exceeds the preset maximum value of the steering wheel zero angle. If the historical steering wheel zero angle exceeds the preset maximum value of the steering wheel zero angle, the historical steering wheel zero angle is discarded and 0 is used for weighted calculation; if the historical steering wheel zero angle does not exceed the preset maximum value of the steering wheel zero angle, the historical steering wheel zero angle is used for weighted calculation.

[0111] The weighting coefficient is determined based on the calculated steering wheel angle value, the preset learning angle threshold, and the current zero-drift level. Specifically, if the current zero-drift level of a target speed segment is high, it is determined whether the calculated steering wheel angle value is less than the preset learning angle threshold. If the calculated steering wheel angle value is less than the preset learning angle threshold, the weighting coefficient is determined to be 0.05. If the current zero-drift level of a target speed segment is medium or low, the weighting coefficient is determined to be 1.

[0112] Based on the actual steering wheel drift value, the control input drift value of the lateral function in the target vehicle, and the calculated steering angle value for each target speed segment, adjustment coefficients are determined from a preset adjustment coefficient table. This table consists of the differences between the actual steering wheel drift value, the control input drift value of the lateral function, and the calculated steering angle value. Therefore, the adjustment coefficients can be determined based on these factors for each target speed segment. The target weighting coefficient is the product of the weighting coefficient and the adjustment coefficient.

[0113] It should be noted that when using this method for the first learning after the target vehicle rolls off the production line, or when using it again for steering wheel zero-position self-learning after the controller memory unit has been cleared, if there is a zero learning value within the current target speed segment, it indicates that the target speed segment has not collected enough sampling steering wheel angle values ​​for one calculation cycle. Therefore, no calculated steering wheel angle value is output, and the target steering wheel zero-position angle for that speed segment is not obtained. Instead, prediction is made based on the zero-position deviation between two adjacent speed segments. If only one of the target steering wheel zero-position angles in two adjacent speed segments is not zero, the non-zero value is used. For example: if the target steering wheel zero-position angle for that speed segment is expressed as... And the target steering wheel zero angle for the previous speed range is expressed as And the target steering wheel zero angle for the next speed range is expressed as Then the target steering wheel zero angle for the target speed segment is the target steering wheel zero angle for the next speed segment, expressed as: .

[0114] When the target steering wheel zero angle of two adjacent speed segments is not 0, the average of the target steering wheel zero angles of the two adjacent speed segments is used as the target steering wheel zero angle of the target speed segment. For example, if the target steering wheel zero angle of the target speed segment is expressed as... And the target steering wheel zero angle for the previous speed range is expressed as And the target steering wheel zero angle for the next speed range is expressed as Then, the target steering wheel zero angle for the target speed range is the average of the target steering wheel zero angles for two adjacent speed ranges, expressed as: .

[0115] If the current target vehicle speed segment is the minimum or maximum vehicle speed segment, the learning value of the adjacent vehicle speed segment that is not zero is used, as shown in the following example: If the target vehicle speed range, i.e., the first vehicle speed range, is represented as... And the target steering wheel zero angle for the second speed range is expressed as The target steering wheel zero angle for the first speed segment and the target steering wheel zero angle for the second speed segment are expressed as: .

[0116] If the target vehicle speed range, i.e., the maximum speed range, is represented as the target steering wheel zero angle, then... And the target steering wheel zero angle for the previous speed range is expressed as The target steering wheel zero angle for the target speed segment, i.e., the maximum speed segment, is the target steering wheel zero angle for the previous speed segment, expressed as: , where j is the maximum value. Because the controller uses floating-point or fixed-point numbers for calculations, 0.001 or other small values ​​are used here to determine whether the variable is zero.

[0117] If the target steering wheel zero angle is not 0 for only one of the multiple speed ranges, then the target steering wheel zero angle for the remaining speed ranges will be adjusted to the target steering wheel zero angle for that speed range.

[0118] It should be noted that when the target vehicle rolls off the production line (i.e., leaves the factory), the controller's initial value is set to zero. Afterwards, before each power-off of the target vehicle, the learned value—the target steering wheel zero-angle for each target speed range—is stored separately. If the target vehicle is powered off within 30 seconds of power-on, the target steering wheel zero-angle for each target speed range is not updated to prevent the recording of a new, unstable target steering wheel zero-angle or the previously learned target steering wheel zero-angle from being abnormally cleared to zero.

[0119] The method provided in this application uses the average value of the steering angle signal sampled at the target vehicle speed range as the basis for calculation. A weighting coefficient is determined by combining the calculated steering angle value, a preset threshold, and the zero-drift level. An adjustment coefficient is then determined based on the actual zero-drift value and the control input zero-drift value. Finally, the target zero-angle is determined by weighting the historical zero-angle and the calculated steering angle value using the target weighting coefficient. This method comprehensively considers the stability of the current sampled data, the impact of zero-drift factors, and the reference value of historical data, achieving dynamic and accurate correction of the zero-angle. It avoids deviations caused by fluctuations in a single data point and gradually optimizes the zero-angle reference, effectively improving the calculation accuracy and reliability of the steering wheel zero-angle at different vehicle speed ranges.

[0120] This application embodiment also provides another possible implementation of a zero-position self-learning method for vehicle steering system software. Before determining the weighting coefficient based on the calculated steering wheel angle value, the preset learning angle threshold, and the current zero-drift level, the method further includes: If the target vehicle's lateral function is active, the current zero-drift level is determined based on the historical steering wheel zero-position angle.

[0121] If the target vehicle's lateral function is not activated, then the preset minimum zero-drift level is set as the current zero-drift level.

[0122] In this embodiment, the historical steering wheel zero angle indicator refers to the historical steering wheel zero angle for each speed segment stored in the steering system software before the target vehicle is powered on, or it can be the steering wheel zero angle for each speed segment newly learned after the target vehicle is powered on. If the lateral function of the target vehicle is active, the historical steering wheel zero angle, the first zero angle threshold, and the second zero angle threshold are compared. If the historical steering wheel zero angle is determined to be less than the first zero angle threshold, the current zero drift level for the current speed segment is determined to be low. If the historical steering wheel zero angle is determined to be greater than or equal to the first zero angle threshold and less than the second zero angle threshold, the current zero drift level for the current speed segment is determined to be medium. If the historical steering wheel zero angle is determined to be greater than or equal to the second zero angle threshold, the current zero drift level for the current speed segment is determined to be high. Furthermore, the zero drift level can only increase unidirectionally within the current power-on cycle; the zero drift level is not stored after power-off and is re-evaluated after each power-on.

[0123] If the target vehicle's lateral function is not activated, then the current zero-drift level for each speed segment is determined to be the preset minimum zero-drift level, i.e., the low level.

[0124] The method provided in this application dynamically determines the current zero-drift level by combining the activation status of the vehicle's lateral function. When the lateral function is activated, the zero-drift level is accurately matched based on the historical steering wheel zero angle; when it is not activated, the preset lowest zero-drift level is directly used. This constructs a zero-drift level determination logic adapted to the lateral function status. This ensures the relevance and accuracy of the zero-drift level when the lateral function is activated, providing a reference basis that fits the actual working conditions for subsequent weighted coefficient calculation. Furthermore, when the lateral function is not activated, the determination process is simplified and basic reliability is ensured, avoiding deviations in the weighted coefficient due to misjudgment of the zero-drift level. This further improves the accuracy and adaptability of the target steering wheel zero angle calculation.

[0125] The following will continue to explain the zero-position self-learning device and control device for implementing the vehicle steering system software provided in any of the above embodiments of this application. The specific implementation process and the resulting technical effects are the same as those in the corresponding method embodiments. For the sake of brevity, parts not mentioned in this embodiment can be referred to the corresponding content in the method embodiments.

[0126] Figure 9 This is a functional module diagram of a zero-position self-learning device for vehicle steering system software provided in an embodiment of this application. Figure 9 As shown, the zero-position self-learning device 100 for the vehicle steering system software includes: The acquisition module 110 is used to acquire the running data of the target vehicle for a preset time period before the current moment. The running data includes: vehicle speed signal and corresponding steering wheel angle signal. The acquisition module 110 is also used to acquire the historical steering wheel zero angle of multiple preset vehicle speed ranges for the steering system software in the target vehicle that are stored in advance. The judgment module 120 is used to determine, based on the running data, whether the target vehicle meets the preset self-learning enable conditions at the current moment; The segmentation module 130 is used to segment the vehicle speed signal according to the vehicle speed range of multiple preset vehicle speed segments if the preset self-learning enable condition is met at the current time, so as to obtain at least one target vehicle speed segment corresponding to the vehicle speed signal. The determining module 140 is used to determine at least one angle signal corresponding to a target vehicle speed segment from the steering wheel angle signal; The determining module 140 is also used to determine the target steering wheel zero angle for each target speed segment based on the steering angle signal corresponding to each target speed segment and the historical steering wheel zero angle for each target speed segment; Update module 150 is used to update the historical steering wheel zero angle for each target vehicle speed segment to the target steering wheel zero angle.

[0127] Optionally, the operational data also includes: multiple detection signals corresponding to the vehicle speed signal, lane line data of the current driving vehicle identified by the camera perception system, driving control data, and lateral function status information; the judgment module 120 is also used to perform vehicle status detection on the target vehicle based on the multiple detection signals and driving control data; if the vehicle status detection result indicates that the target vehicle has no communication failure and is in a steady-state driving state, then the driving scenario detection is performed on the target vehicle based on the lane line data; if the driving scenario detection result indicates that the target vehicle is in a straight driving scenario, then the driving behavior detection is performed on the target vehicle based on the driving control data; if the driving behavior detection result indicates that the target vehicle is not in a driver over-control state, then the lane parallel detection is performed on the target vehicle based on the lane line data; if the lane parallel detection result indicates that the target vehicle is in a lane parallel driving state, then the lateral function status detection is performed based on the lateral function status information; if the lateral function status detection result indicates that the lateral function detection is passed, then it is determined that the preset self-learning enabling conditions are met at the current moment.

[0128] Optionally, the multiple detection signals also include: longitudinal acceleration signal, lateral acceleration signal and yaw rate signal collected by the inertial navigation sensor; the judgment module 120 is also used to detect the communication status of the target vehicle based on the multiple detection signals and driving control data; if the target vehicle does not have a communication fault, it determines whether the target vehicle is in a steady-state driving state based on the longitudinal acceleration value of the longitudinal acceleration signal, the lateral acceleration value of the lateral acceleration signal and the yaw rate value of the yaw rate signal.

[0129] Optionally, the lane line data includes: the polynomial coefficients of the current driving road of the target vehicle, and the confidence level; the judgment module 120 is further used to determine that the current driving road is a straight driving road based on the polynomial coefficients of the current driving road; if the current driving road is a straight driving road, then based on the confidence level of the current driving road, determine whether the current driving road meets the preset lane line clarity condition; if the current driving road meets the preset lane line clarity condition, then determine that the current driving scenario of the target vehicle is a straight driving scenario with clear lane lines.

[0130] Optionally, the driving control data includes: hand torque, steering wheel speed, and steering wheel angle; the judgment module 120 is also used to determine whether the target vehicle is under driver control based on the hand torque and steering wheel speed; if the target vehicle is not under driver control, it is determined that the target vehicle is not under driver over-control; if the target vehicle is under driver control, it is determined whether the target vehicle has entered turning mode based on the hand torque and steering wheel angle; if the duration of the target vehicle switching to the steering wheel centering state after entering turning mode is less than or equal to a preset duration, it is determined that the target vehicle is under driver over-control; if the duration of the target vehicle switching to the steering wheel centering state is greater than the preset duration, it is determined that the target vehicle is not under driver over-control.

[0131] Optionally, the judgment module 120 is also used to determine the heading angle and aiming deviation of the target vehicle relative to the lane line based on the lane line data; determine whether the target vehicle is parallel to the center line of the road based on the heading angle and aiming deviation of the target vehicle relative to the lane line; if the target vehicle is parallel to the center line of the road, then it is determined that the target vehicle is in a parallel driving state.

[0132] Optionally, the judgment module 120 is further configured to determine that the lateral function detection of the target vehicle has passed if the lateral function status information indicates that the lateral function of the target vehicle is not activated; and to determine that the lateral function detection of the target vehicle has passed if the lateral function status information indicates that the lateral function of the target vehicle has been activated and the activation time is greater than a preset duration, and the corresponding performance of the steering function actuator corresponding to the lateral function meets the preset performance constraints.

[0133] Optionally, the determining module 140 is further configured to sample the steering angle signal corresponding to each target vehicle speed segment within a preset calculation period to obtain multiple sampled steering wheel angle values ​​within the current calculation period; calculate the average steering wheel angle value as the calculated steering wheel angle value for the current calculation period based on the multiple sampled steering wheel angle values; determine the weighting coefficient based on the calculated steering wheel angle value, the preset learning steering angle threshold, and the current zero-drift level; determine the adjustment coefficient based on the actual steering wheel zero-drift value for each target vehicle speed segment, the control input zero-drift value of the lateral function in the target vehicle, and the calculated steering angle value; determine the target weighting coefficient based on the weighting coefficient and the adjustment coefficient; and perform a weighted calculation on the historical steering wheel zero angle and the calculated steering angle value based on the target weighting coefficient to obtain the target steering wheel zero angle for each target vehicle speed segment.

[0134] Optionally, the determining module 140 is further configured to determine the current zero-drift level based on the historical steering wheel zero angle if the lateral function of the target vehicle is activated; and to determine the preset minimum zero-drift level as the current zero-drift level if the lateral function of the target vehicle is deactivated.

[0135] The above-described device is used to execute the method provided in the foregoing embodiments, and its implementation principle and technical effect are similar, so they will not be described again here.

[0136] These modules can be one or more integrated circuits configured to implement the above methods, such as one or more Application Specific Integrated Circuits (ASICs), one or more microprocessors, or one or more Field Programmable Gate Arrays (FPGAs). Alternatively, when a module is implemented using processing element scheduler code, the processing element can be a general-purpose processor, such as a Central Processing Unit (CPU) or other processor capable of calling program code. Furthermore, these modules can be integrated together as a system-on-a-chip (SOC).

[0137] Figure 10 This is a schematic diagram of a control device provided in an embodiment of this application. This control device can be used for zero-position self-learning of vehicle steering system software. Figure 10 As shown, the control device includes: processor 210, storage medium 220, and bus 230.

[0138] Storage medium 220 stores machine-readable instructions executable by processor 210. When the control device is running, processor 210 communicates with storage medium 220 via bus 230, and processor 210 executes the machine-readable instructions to perform the steps of the above method embodiment. The specific implementation and technical effects are similar and will not be described in detail here.

[0139] Optionally, this application also provides a storage medium 220, on which a computer program is stored. When the computer program is run by a processor, it executes the steps of the above-described method embodiments. The specific implementation and technical effects are similar, and will not be repeated here.

[0140] In the several embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0141] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0142] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0143] The integrated units implemented as software functional units described above can be stored in a computer-readable storage medium. These software functional units, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0144] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A zero-position self-learning method for vehicle steering system software, characterized in that, The method includes: Obtain the target vehicle's operating data for a preset time period prior to the current moment, wherein the operating data includes: vehicle speed signal and corresponding steering wheel angle signal; Obtain the historical steering wheel zero angle for multiple preset vehicle speed ranges that are stored in advance for the steering system software in the target vehicle; Based on the operational data, it is determined whether the target vehicle meets the preset self-learning enable conditions at the current moment; If the current time satisfies the preset self-learning enable condition, the vehicle speed signal is segmented according to the vehicle speed range of the multiple preset vehicle speed segments to obtain at least one target vehicle speed segment corresponding to the vehicle speed signal. Determine the angle signal corresponding to the at least one target vehicle speed segment from the steering wheel angle signal; The target steering wheel zero angle for each target speed segment is determined based on the steering angle signal corresponding to each target speed segment and the historical steering wheel zero angle for each target speed segment. Update the historical steering wheel zero angle for each target vehicle speed segment to the target steering wheel zero angle.

2. The method according to claim 1, characterized in that, The operational data also includes: multiple detection signals corresponding to the vehicle speed signal, lane line data of the current driving vehicle identified by the camera perception system, driving control data, and lateral function status information; The step of determining whether the target vehicle meets the preset self-learning enable conditions at the current moment based on the operational data includes: Based on the multiple detection signals and the driving control data, the vehicle status of the target vehicle is detected; If the vehicle status detection result indicates that the target vehicle has no communication failure and is in a steady-state driving state, then the driving scenario detection is performed on the target vehicle based on the lane line data. If the driving scenario detection result indicates that the target vehicle is in a straight driving scenario, then driving behavior detection is performed on the target vehicle based on the driving control data; If the driving behavior detection result indicates that the target vehicle is not in a state of driver over-control, then lane parallelism detection is performed on the target vehicle based on the lane line data; If the lane parallel detection result indicates that the target vehicle is in a lane parallel driving state, then lateral function state detection is performed based on the lateral function state information. If the lateral function status detection result indicates that the lateral function detection has passed, then it is determined that the current moment meets the preset self-learning enable condition.

3. The method according to claim 2, characterized in that, The multiple detection signals also include: longitudinal acceleration signal, lateral acceleration signal and yaw rate signal collected by the inertial navigation sensor; The step of detecting the vehicle status of the target vehicle based on the multiple detection signals and the driving control data includes: Based on the multiple detection signals and the driving control data, the communication status of the target vehicle is detected; If the target vehicle does not have a communication fault, then the target vehicle is determined to be in a steady-state driving state based on the longitudinal acceleration value of the longitudinal acceleration signal, the lateral acceleration value of the lateral acceleration signal, and the yaw rate value of the yaw rate signal.

4. The method according to claim 2, characterized in that, The lane line data includes: the polynomial coefficients of the target vehicle's current driving road, and the confidence level; the driving scene detection of the target vehicle based on the lane line data includes: Based on the polynomial coefficients of the current driving road, the current driving road is determined to be a straight driving road; If the current driving road is a straight driving road, then based on the confidence level of the current driving road, determine whether the current driving road meets the preset lane line clear condition; If the current driving road meets the preset lane line clear condition, then the current driving scenario of the target vehicle is determined to be a straight driving scenario with clear lane lines.

5. The method according to claim 2, characterized in that, The driving control data includes: hand torque, steering wheel speed, and steering wheel angle; the driving behavior detection of the target vehicle based on the driving control data includes: Based on the hand torque and the steering wheel speed, determine whether the target vehicle is under driver control. If the target vehicle is not under driver control, then it is determined that the target vehicle is not under driver over-control. If the target vehicle is under driver control, then the driver determines whether the target vehicle has entered a turning mode based on the hand torque and the steering wheel rotation angle. If the duration for which the target vehicle switches to the steering wheel centering state after entering the turning mode is less than or equal to a preset duration, then the target vehicle is determined to be in a state of driver over-control. If the duration of the target vehicle switching to the steering wheel centering state is greater than the preset duration, then it is determined that the target vehicle is not in a state of driver over-control.

6. The method according to claim 2, characterized in that, The step of performing lane parallelism detection on the target vehicle based on the lane line data includes: Based on the lane line data, determine the heading angle of the target vehicle relative to the lane line and the aiming deviation; Based on the heading angle and aiming deviation of the target vehicle relative to the lane line, determine whether the target vehicle is parallel to the center line of the road; If the target vehicle is parallel to the center line of the road, then the target vehicle is determined to be in a parallel driving state in the lane.

7. The method according to claim 2, characterized in that, The step of detecting the horizontal functional status based on the horizontal functional status information includes: If the lateral function status information indicates that the lateral function of the target vehicle is not activated, then the lateral function detection of the target vehicle is determined to be successful. If the lateral function status information indicates that the lateral function of the target vehicle has been activated and the activation time is greater than a preset duration, and the corresponding performance of the steering function actuator corresponding to the lateral function meets the preset performance constraints, then the lateral function detection of the target vehicle is determined to be successful.

8. The method according to claim 1, characterized in that, The step of determining the target steering wheel zero angle for each target speed segment based on the steering angle signal corresponding to each target speed segment and the historical steering wheel zero angle for each target speed segment includes: Within a preset calculation period, the steering angle signal corresponding to each target vehicle speed segment is sampled to obtain multiple sampled steering wheel angle values ​​within the current calculation period; Based on the multiple sampled steering wheel angle values, the average steering wheel angle value is calculated as the calculated steering wheel angle value for the current calculation cycle; The weighting coefficients are determined based on the calculated steering wheel angle value, the preset learning angle threshold, and the current zero-drift level. Based on the actual steering wheel zero-drift value for each target speed segment, the control input zero-drift value of the lateral function in the target vehicle, and the calculated steering angle value, an adjustment coefficient is determined. The target weight coefficient is determined based on the weighting coefficient and the adjustment coefficient. Based on the target weighting coefficient, the historical steering wheel zero angle and the calculated steering angle value are weighted and calculated to obtain the target steering wheel zero angle for each target speed segment.

9. The method according to claim 8, characterized in that, Before determining the weighting coefficient based on the calculated steering wheel angle value, the preset learning angle threshold, and the current zero-drift level, the method further includes: If the lateral function of the target vehicle is active, the current zero-drift level is determined based on the historical steering wheel zero angle. If the lateral function of the target vehicle is not activated, the preset minimum zero-drift level is determined as the current zero-drift level.

10. A control device, characterized in that, include: The system includes a processor, a storage medium, and a bus. The storage medium stores program instructions executable by the processor. When the control device is running, the processor communicates with the storage medium via the bus. The processor executes the program instructions to perform the steps of the zero-position self-learning method for vehicle steering system software as described in any one of claims 1 to 9.

Images