A by-wire method and system for vehicle steering

Abstract

The application discloses a drive-by-wire method and system for vehicle steering, and belongs to the technical field of steering control. The method comprises the following steps: calculating the vehicle state and the road surface state through an intention recognition model to obtain a driving intention; judging whether the driving intention is to enter a garage or to leave the garage; if yes, judging whether a first condition is met; if yes, marking a current steering wheel steering angle as a first steering angle, and adjusting a current steering ratio to a second steering ratio, wherein the second steering ratio is smaller than the first steering ratio. Under the first condition, the second steering ratio is adjusted to improve the sensitivity of steering, and the flexibility of steering is improved in the case of continuous large-scale steering, which is beneficial to large-angle steering at low speed. Between the third threshold value and the maximum steering angle, the steering ratio is greatly reduced to play a buffering role and avoid steering out of limits. The flexibility of leaving and entering the garage is improved.

Description

Technical Field

[0001] This invention relates to the field of steering control technology, and more specifically to a drive-by-wire method and adjustment system for vehicle steering. Background Technology

[0002] Steer-by-wire (SBW) systems eliminate the mechanical steering column, using an electric motor to directly control wheel steering, offering advantages such as fast response and flexible spatial layout. Because SBW eliminates the rigid mechanical connection between the steering wheel and the steering wheels, the steering ratio can be adjusted via the motor and ECU to adapt to the turning requirements in both low-speed and high-speed scenarios. For example, it can meet the need for steering sensitivity at low speeds or when parking, while simultaneously ensuring stable steering at high speeds.

[0003] Steering ratio is the ratio of the steering wheel angle to the steering wheel angle. With the increasing intelligence of automobiles, the requirements for the steering ratio of steer-by-wire systems are becoming more and more stringent. In addition to meeting safety and emergency handling requirements, the steering ratio must not change abruptly under normal driving conditions.

[0004] CN120135275A discloses a fault-tolerant control method for a vehicle steer-by-wire (SBW) system, which combines the relationship between the vehicle's lateral acceleration and the road surface adhesion coefficient to perform a fusion correction on the dynamic steering ratio. Setting a smaller steering ratio in low-speed scenarios can reduce the steering wheel rotation amplitude and improve agility.

[0005] Currently, a constant steering ratio is typically used at low speeds. However, at large turning angles, frequent and large-angle steering wheel turns are required, resulting in a less than ideal user experience. For example, when parallel parking or driving out of a garage, multiple full turns of the steering wheel in both directions are necessary, leading to a less than ideal steering experience. Summary of the Invention

[0006] In view of the above-mentioned technical problems existing in the prior art, the present invention provides a drive-by-wire method and adjustment system for vehicle steering, which dynamically adjusts the steering ratio under low-speed driving conditions.

[0007] This invention discloses a drive-by-wire method for vehicle steering, comprising the following steps: Collect current vehicle and road conditions; The driving intention is obtained by calculating the vehicle state and road surface state through an intent recognition model; Determine whether the driving intention is to enter or exit the parking space; If so, determine whether the following first condition is met: the current steering wheel angle is greater than the first threshold and less than the third threshold, and the steering rate is greater than the second threshold; If the conditions are met, the current steering wheel angle is marked as the first steering angle, and the current steering ratio is adjusted to the second steering ratio, wherein the second steering ratio is less than the first steering ratio.

[0008] Preferably, the present invention further includes a method for adjusting the steering ratio to a third steering ratio: If the steering wheel angle is between the third threshold and the maximum steering angle, the current steering ratio will be adjusted to the third steering ratio, which is greater than the first steering ratio.

[0009] Preferably, when the steering angle returns to center, the steering wheel angle is between the third threshold and the maximum steering angle, using a third steering ratio; When the steering wheel angle is between the first steering angle and the third threshold range, the second steering ratio is used; The steering wheel angle is between the first steering angle and the return-to-center angle, using the first steering ratio.

[0010] Preferred methods for training the intent recognition model include: Collect training data; An intent recognition model is obtained by training the training set using machine learning methods.

[0011] Preferably, the machine learning method includes BiLSTM, and the specific training method for the intent recognition model includes: Data acquisition, data annotation, and data preprocessing are performed to obtain a dataset; Select modeling features; Based on the modeling features, the dataset is standardized and normalized to obtain the training set; The BiLSTM method is used to train the training set to obtain the intent recognition model.

[0012] Preferably, the characteristic indicators are selected from: vehicle speed, steering wheel angle, steering wheel turning rate, lateral acceleration, trajectory curvature, distance between the vehicle body and the left indicator line, distance between the vehicle body and the right indicator line, angle between the vehicle body and the left indicator line, angle between the vehicle body and the right indicator line, reverse gear, and turn signal.

[0013] Preferably, specific methods for driver intent recognition include: The probability of outbound or inbound is obtained through the intent recognition model; If the second condition is met, the driving intention is to enter or exit the parking space. The second condition includes: the probability is greater than the seventh threshold, the reverse gear value is 1, and the average vehicle speed is less than the eighth threshold.

[0014] Preferably, the method for generating the second steering ratio includes: The first steering angle and the first steering ratio are used as the first node, the preset second steering angle and the second steering ratio are used as the second node, and the third steering angle and the third steering ratio corresponding to the third threshold are used as the third node. Based on the cosine interpolation method, an interpolation curve is generated according to the first node, the second node, and the third node; The second steering ratio is obtained based on the current steering angle and the interpolation curve.

[0015] A second aspect of the present invention provides an adjustment system for implementing the above-described steer-by-wire method for vehicle steering, comprising a data acquisition module, an intent recognition module, a node marking module, a second steering ratio adjustment module, and a third steering ratio adjustment module; The data acquisition module is used to collect the current vehicle status and road surface status; The intent recognition module is used to calculate the driving intent by using the intent recognition model to analyze the vehicle state and road surface state. The node marking module is used to identify the steering wheel angle as the first steering angle when the first condition is met; The second steering ratio adjustment module is used to adjust the current steering ratio to the second steering ratio when the first condition is met; The third steering ratio adjustment module is used to adjust the current steering ratio to the third steering ratio when the steering wheel angle is between the third threshold and the maximum steering angle.

[0016] Preferably, the adjustment system further includes an interpolation module, which is used to: use a first steering angle and a first steering ratio as a first node, a preset second steering angle and a second steering ratio as a second node, and a third steering angle and a third steering ratio corresponding to a third threshold as a third node; Based on the cosine interpolation method, an interpolation curve is generated according to the first node, the second node, and the third node; The second steering ratio is obtained based on the current steering angle and the interpolation curve.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: Under the first condition, adjust to the second steering ratio to improve steering sensitivity and enhance steering flexibility under continuous large-angle steering, which is beneficial for large-angle steering at low speeds. Between the third threshold and the maximum steering angle, the steering ratio is significantly reduced to avoid oversteering and improve the flexibility of outbound and inbound operations. Attached Figure Description

[0018] Figure 1 This is a flowchart of a steer-by-wire method for vehicle steering according to the steer-by-wire system of the present invention; Figure 2 This is a logic block diagram of the adjustment system of the present invention; Figure 3 This is a schematic diagram of the interpolation curve in Example 3. Detailed Implementation

[0019] 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 only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] The present invention will now be described in further detail with reference to the accompanying drawings: Example 1 provides a drive-by-wire method for vehicle steering, such as Figure 1 As shown, it includes the following steps: Step S1: Collect the current vehicle status and road surface status.

[0021] The current vehicle status includes vehicle speed, steering angle, acceleration, and first steering ratio.

[0022] Step S2: Calculate the vehicle state and road surface state using the intent recognition model to obtain the driving intent.

[0023] Step S3: Determine whether the driving intention is to enter or exit the parking space.

[0024] Step S4: If yes, determine whether the following first condition is met: the steering wheel angle is greater than the first threshold and less than the third threshold, and the steering rate is greater than the second threshold, and mark the steering wheel angle as the first steering angle.

[0025] If not, continue collecting vehicle status and road surface status data.

[0026] Step S5: If satisfied, adjust the current steering ratio to the second steering ratio, where the second steering ratio is less than the first steering ratio.

[0027] If the requirements are not met, the first steering ratio will be used.

[0028] Step S6: If the steering wheel angle is between the third threshold and the maximum steering angle, adjust the current steering ratio to the third steering ratio, which is greater than the first steering ratio.

[0029] Step S7: After the steering wheel angle is straightened, proceed to step S1.

[0030] The steering ratio during the steering angle return process corresponds to the steering process itself. Specifically, the third steering ratio is used when the steering wheel angle is between the third threshold and the maximum steering angle; the second steering ratio is used when the steering wheel angle is between the first and third thresholds; and the first steering ratio is used between the first threshold and the return to center.

[0031] The first steering ratio ranges from the point of return to center to the first steering angle; the second steering ratio ranges from the first steering angle to the third threshold; and the third steering ratio ranges from the third threshold to the maximum steering angle.

[0032] Under the first condition, adjust to the second steering ratio to improve steering sensitivity and enhance steering flexibility under continuous large steering angles; between the third threshold and the maximum steering angle, significantly reduce the steering ratio to act as a buffer and prevent steering from exceeding limits; improve the flexibility of outbound and inbound operations.

[0033] In one specific embodiment, the difference between the third threshold and the maximum steering angle can be set between 5 and 100 degrees. The first threshold ranges from 15 to 120 degrees; the second threshold is a change in steering wheel angle greater than 20 degrees within 100 ms.

[0034] In step S2, the intent recognition model can adopt a planning and threshold-based method, such as steering angle threshold, pedal rate threshold, lateral displacement threshold, vehicle side garage line greater than the fifth threshold, and vehicle front and rear obstacle distance less than the sixth threshold, etc.

[0035] Intent recognition models can also employ machine learning-based methods, such as Hidden Markov Models, SVM, Random Forest, and XGBoost; as well as deep learning methods, such as BiLSTM (Bi-directional Long Short-Term Memory), GRU, and GRU-Attention. BiLSTM is composed of a forward LSTM and a backward LSTM.

[0036] Example 2 provides a method for training an intent recognition model based on BiLSTM, including the following steps: Step 201: Data acquisition, data labeling, and data preprocessing to obtain the dataset.

[0037] We randomly collected data on vehicle and road conditions during parking garage entry and exit; vehicle and road conditions during parking space entry and exit; vehicle and road conditions before and after temporary parking; and vehicle and road conditions before and after parking at a red light, totaling 13,631 data points. Each data point was labeled.

[0038] Vehicle status parameters include steering angle, steering rate, vehicle speed, longitudinal acceleration, and trajectory curvature. Road surface status parameters include: distance to side marker lines, angle between side marker lines, obstacles in front of the vehicle, and obstacles behind the vehicle.

[0039] Data preprocessing includes data augmentation and missing data handling.

[0040] The markings for leaving and entering the warehouse are 1, and the markings for temporary parking and parking at red lights are 0.

[0041] Step 202: Select modeling features.

[0042] A time window of T = 3 seconds and a step size of 0.1 seconds are used to construct a 30-step time vector. The feature indicators for each step are selected from: vehicle speed, steering wheel angle, steering wheel rate, lateral acceleration, trajectory curvature, distance between the vehicle body and the left indicator line, distance between the vehicle body and the right indicator line, angle between the vehicle body and the left indicator line, angle between the vehicle body and the right indicator line, reverse gear, and turn signal. The values ​​for reverse gear and turn signal are quantized as 1 or 0.

[0043] The indicator lines can be detected by capturing images of the indicator lines using an onboard camera and then using a vision-based method, which is an existing technology.

[0044] Step 203: Standardize and normalize the data to obtain the training set and validation set.

[0045] Normalize the index values ​​to [0, 1].

[0046] Step 204: Train the intention recognition model on the training set using the BiLSTM method to obtain the model; validate the intention recognition model using the validation set.

[0047] The BiLSTM network consists of an input layer, a BiLSTM layer, a self-attention layer, a fully connected layer, an activation function, a fully connected layer, and a regression layer, connected in sequence.

[0048] The loss function CrossEntropy was used, with the Adam optimizer trained for 100 epochs and a learning rate decay strategy. The average training loss was 0.0646, the validation accuracy was 0.8853, and the final accuracy was 0.8969.

[0049] The intent recognition model outputs the probabilities p0 of outbound and inbound shipments.

[0050] Step 205: Obtain the probability of outbound or inbound through the intent recognition model.

[0051] Step 206: If the second condition is met, the driving intention is to enter or exit the parking space.

[0052] The second condition includes: probability P0 is greater than the seventh threshold, reverse gear value is 1, and average vehicle speed is less than the eighth threshold. Average vehicle speed refers to the average vehicle speed over a certain period of time, such as the average vehicle speed over 2-10 seconds being less than 0-20 km / h, but not limited to this.

[0053] If the second condition is not met, the driving intention is normal driving.

[0054] By combining predicted probabilities and indicator thresholds, driving intentions can be identified, improving the accuracy of identification. At the same time, the interpretability of driving intentions is enhanced.

[0055] Example 3 provides an improved method for generating the second steering ratio, comprising the following steps: Step 301: Take the first steering angle and the first steering ratio as the first node, the preset second steering angle and the second steering ratio as the second node, and the third steering angle and the third steering ratio corresponding to the third threshold as the third node.

[0056] Step 302: Based on the cosine interpolation method, generate an interpolation curve or interpolation table according to the first node, the second node, and the third node.

[0057] Step 303: Obtain the second steering ratio based on the current steering angle and the interpolation curve.

[0058] By using interpolation, a smooth interpolation curve is generated, and a second steering ratio is obtained, thus avoiding driving discomfort caused by sudden changes in steering ratio.

[0059] Figure 3 The specific interpolation curves are shown, with a first steering ratio of 1:10, a minimum second steering ratio of 1:30, and a third steering ratio of 1:7, but not limited to these.

[0060] Example 4 provides an adjustment system for implementing the above-described steer-by-wire method for vehicle steering, such as... Figure 2 As shown, it includes a data acquisition module 1, an intent recognition module 2, a node marking module 3, a second steering ratio adjustment module 4, and a third steering ratio adjustment module 5; The acquisition module 1 is used to acquire the current vehicle status and road surface status.

[0061] The intent recognition module 2 is used to calculate the vehicle state and road surface state through the intent recognition model to obtain the driving intent.

[0062] The node marking module 3 is used to mark the steering wheel angle as the first steering angle when the first condition is met.

[0063] The second steering ratio adjustment module 4 is used to adjust the current steering ratio to the second steering ratio when the first condition is met.

[0064] The third steering ratio adjustment module 5 is used to adjust the current steering ratio to the third steering ratio when the steering wheel angle is between the third threshold and the maximum steering angle.

[0065] The adjustment system also includes an interpolation module 6, which is used to: take the first steering angle and the first steering ratio as the first node, the preset second steering angle and the second steering ratio as the second node, and the third steering angle and the third steering ratio corresponding to the third threshold as the third node; generate an interpolation curve based on the first node, the second node and the third node using a cosine interpolation method; and obtain the second steering ratio based on the current steering angle and the interpolation curve.

[0066] This invention recognizes driving intentions and dynamically and automatically adjusts the steering ratio when entering or exiting a parking space, improving the flexibility of large-angle turns; it also reserves adjustment space between the third threshold and the maximum steering angle to avoid over-limit situations under sharp turns.

[0067] By combining rule-based and intent recognition models, driving intentions can be identified, improving the readability and interpretability of the recognition results.

[0068] The method based on cosine interpolation enables a smooth transition between the second steering ratio and the adjacent first and third steering ratios.

[0069] In some of the processes described in the specification, claims, and accompanying drawings of this invention, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0070] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems and methods according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0071] It should also be noted that, without conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0072] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A drive-by-wire method for vehicle steering, characterized in that, Includes the following steps: Collect current vehicle and road conditions; The driving intention is obtained by calculating the vehicle state and road surface state through an intent recognition model; Determine whether the driving intention is to enter or exit the parking space; If so, determine whether the following first condition is met: the current steering wheel angle is greater than the first threshold and less than the third threshold, and the steering rate is greater than the second threshold; If the conditions are met, the current steering wheel angle is marked as the first steering angle, and the current steering ratio is adjusted to the second steering ratio, wherein the second steering ratio is less than the first steering ratio.

2. The drive-by-wire method for vehicle steering according to claim 1, characterized in that, It also includes a method for adjusting the steering ratio to the third steering ratio: If the steering wheel angle is between the third threshold and the maximum steering angle, the current steering ratio will be adjusted to the third steering ratio, which is greater than the first steering ratio.

3. The drive-by-wire method for vehicle steering according to claim 2, characterized in that, When the steering angle returns to center, the steering wheel angle between the third threshold and the maximum steering angle uses the third steering ratio; When the steering wheel angle is between the first steering angle and the third threshold range, the second steering ratio is used; The steering wheel angle is between the first steering angle and the return-to-center angle, using the first steering ratio.

4. The drive-by-wire method for vehicle steering according to claim 1, characterized in that, Training methods for intent recognition models include: Collect training data; An intent recognition model is obtained by training the training set using machine learning methods.

5. A drive-by-wire method for vehicle steering according to claim 4, characterized in that, The machine learning methods include BiLSTM, and the specific training methods for the intent recognition model include: Data acquisition, data annotation, and data preprocessing are performed to obtain a dataset; Select modeling features; Based on the modeling features, the dataset is standardized and normalized to obtain the training set; The BiLSTM method is used to train the training set to obtain the intent recognition model.

6. A drive-by-wire method for vehicle steering according to claim 5, characterized in that, The characteristic indicators are selected from: vehicle speed, steering wheel angle, steering wheel turning rate, lateral acceleration, trajectory curvature, distance between the vehicle body and the left indicator line, distance between the vehicle body and the right indicator line, angle between the vehicle body and the left indicator line, angle between the vehicle body and the right indicator line, reverse gear, and turn signals.

7. A drive-by-wire method for vehicle steering according to claim 5, characterized in that, Specific methods for driver intent recognition include: The probability of outbound or inbound is obtained through the intent recognition model; If the second condition is met, the driving intention is to enter or exit the parking space. The second condition includes: the probability is greater than the seventh threshold, the reverse gear value is 1, and the average vehicle speed is less than the eighth threshold.

8. A drive-by-wire method for vehicle steering according to claim 1, characterized in that, Methods for generating the second steering ratio include: The first steering angle and the first steering ratio are used as the first node, the preset second steering angle and the second steering ratio are used as the second node, and the third steering angle and the third steering ratio corresponding to the third threshold are used as the third node. Based on the cosine interpolation method, an interpolation curve is generated according to the first node, the second node, and the third node; The second steering ratio is obtained based on the current steering angle and the interpolation curve.

9. A drive-by-wire system for vehicle steering, characterized in that, For implementing a drive-by-wire method for vehicle steering as described in any one of claims 1-8, the adjustment system includes a data acquisition module, an intent recognition module, a node marking module, a second steering ratio adjustment module, and a third steering ratio adjustment module; The data acquisition module is used to collect the current vehicle status and road surface status; The intent recognition module is used to calculate the driving intent by using the intent recognition model to analyze the vehicle state and road surface state. The node marking module is used to identify the steering wheel angle as the first steering angle when the first condition is met; The second steering ratio adjustment module is used to adjust the current steering ratio to the second steering ratio when the first condition is met; The third steering ratio adjustment module is used to adjust the current steering ratio to the third steering ratio when the steering wheel angle is between the third threshold and the maximum steering angle.

10. A drive-by-wire system for vehicle steering according to claim 9, characterized in that, It also includes an interpolation module, which is used to: take the first steering angle and the first steering ratio as the first node, the preset second steering angle and the second steering ratio as the second node, and the third steering angle and the third steering ratio corresponding to the third threshold as the third node; Based on the cosine interpolation method, an interpolation curve is generated according to the first node, the second node, and the third node; The second steering ratio is obtained based on the current steering angle and the interpolation curve.

Images