Lane centering control method and controller, storage medium and vehicle

Abstract

The invention discloses a lane centering control method, a controller, a storage medium and a vehicle. The lane centering control method comprises the following steps: acquiring a vehicle driving road image; acquiring fluctuation state information of the vehicle driving road surface according to the vehicle driving road image; and according to the fluctuation state information of the road surface of the vehicle driving road, vehicle aligning torque is obtained, and the vehicle is controlled to be aligned according to the vehicle aligning torque. The method can control the vehicle to return in time according to the fluctuation state information of the road surface, so that the deviation between the vehicle driving track and the target track is reduced, and the transverse control precision and the centering control reliability are improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicles, and in particular to a lane centering control method, a lane centering controller, a non-volatile readable storage medium and a vehicle. Background Art

[0002] In the related art, a solution based on correction of the steering angle gain proportional factor actively increases the actual steering gear angle when the steering angle is less than a preset value, thereby offsetting the deviation caused by part of the steering clearance. However, the influence of factors such as road roll and tire condition is not taken into account, and there are still certain limitations for some scenarios. There is a certain lag in feedback control, resulting in a large deviation between the actual driving trajectory and the target trajectory, and low lateral control accuracy, which has a negative impact on the reliability of lane centering control. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, one purpose of the present invention is to provide a lane centering control method, which can control the vehicle to return to the center in time according to the undulating state information of the road surface, reduce the deviation between the vehicle's driving trajectory and the target trajectory, and improve the lateral control accuracy and the reliability of the centering control.

[0004] The second object of the present invention is to provide a lane centering controller.

[0005] The third object of the present invention is to provide a non-volatile readable storage medium.

[0006] A fourth object of the present invention is to provide a vehicle.

[0007] In order to solve the above problems, an embodiment of the first aspect of the present invention provides a lane centering control method, including: acquiring a vehicle driving road image; obtaining undulating state information of the vehicle driving road surface based on the vehicle driving road image; obtaining a vehicle centering torque based on the undulating state information of the vehicle driving road surface, so as to control the vehicle to center itself based on the vehicle centering torque.

[0008] According to the lane centering control method of an embodiment of the present invention, after acquiring the vehicle driving road image, the undulation state information of the road surface of the vehicle driving is obtained according to the vehicle driving road image, and finally the vehicle is controlled to return to the center position in time according to the undulation state information of the road surface, so that the deviation between the vehicle driving trajectory and the target trajectory is reduced, and the lateral control accuracy and the reliability of the centering control are improved.

[0009] In some embodiments, obtaining the undulating state information of the road surface of the vehicle driving on the basis of the vehicle driving road image includes: performing two-dimensional lane line recognition, road plane grid estimation and inverse perspective transformation on the vehicle driving road image to obtain the three-dimensional lane line information. Determining the height of the lane line in a direction perpendicular to the road surface according to the three-dimensional lane line information; and determining the undulating state information of the lane line according to the height.

[0010] In some embodiments, the vehicle's righting torque is obtained based on the undulating state information of the road surface on which the vehicle is traveling, including: obtaining the vehicle's righting torque through a neural network model, wherein the input characteristic value of the neural network model is the undulating state information of the road surface on which the vehicle is traveling, and the output characteristic value of the neural network model is the vehicle's righting torque.

[0011] In some embodiments, the lane centering control method further includes: obtaining a cloud-calculated self-aligning requirement torque sent from the cloud, wherein the cloud-calculated self-aligning requirement torque is determined based on a steering angle gain proportional factor; obtaining a vehicle-side calculated self-aligning requirement torque based on the vehicle self-aligning torque; determining a target self-aligning requirement torque based on the cloud-calculated self-aligning requirement torque and the vehicle-side calculated self-aligning requirement torque, so as to control the vehicle self-alignment based on the target self-aligning requirement torque.

[0012] In some embodiments, a target return-to-alignment required torque is determined based on the return-to-alignment required torque calculated on the cloud side and the return-to-alignment required torque calculated on the vehicle side, including: obtaining the torque difference between the return-to-alignment required torque calculated on the cloud side and the return-to-alignment required torque calculated on the vehicle side; when the torque difference is greater than a torque threshold, reminding the driver to take over the vehicle, and performing corresponding operations based on the driver's response to the reminder to take over the vehicle; when the torque difference is less than the torque threshold, the target return-to-alignment required torque is the return-to-alignment required torque calculated on the vehicle side.

[0013] In some embodiments, the torque threshold is obtained by the following formula:

[0014] V = [T1 + (T2 + T') / 2] / 2;

[0015] T' = (T1 + T2) / 2;

[0016] Among them, V is the torque threshold, T1 is the return-to-alignment required torque calculated by the vehicle end, T2 is the return-to-alignment required torque calculated by the cloud end, and T' is the average value of the return-to-alignment required torque calculated by the cloud end and the return-to-alignment required torque calculated by the vehicle end at the previous moment.

[0017] The second aspect of the present invention provides a lane centering controller, comprising: at least one processor; a memory communicatively connected to the at least one processor; the memory stores a computer program executable by the at least one processor, and when the at least one processor executes the computer program, the lane centering control method described in the above embodiment is implemented.

[0018] According to the lane centering controller of the embodiment of the present invention, the corresponding lane centering control program can be stored in the memory. When the lane centering control method is implemented, the processor runs the program in the memory, and controls the vehicle to return to the center position in time according to the undulating state information of the road surface, so as to reduce the deviation between the vehicle's driving trajectory and the target trajectory, thereby improving the lateral control accuracy and the reliability of the centering control.

[0019] A third aspect of the present invention provides a non-volatile readable storage medium having a computer program stored thereon, wherein the computer program, when executed, implements the lane centering control method described in the above embodiment.

[0020] A fourth aspect of the present invention provides a vehicle, comprising: a forward-looking camera for capturing images of the road the vehicle is traveling on; a power steering system for driving the vehicle to return to the center position; and a controller, wherein the controller is respectively connected to the forward-looking camera and the power steering system, and is used to execute the lane centering control method described in the above embodiment.

[0021] According to the vehicle of the embodiment of the present invention, after the front-view camera acquires the image of the road the vehicle is traveling on, the undulation state information of the road the vehicle is traveling on is obtained according to the image of the road the vehicle is traveling on, and finally the controller controls the power steering system to drive the vehicle to return to the center position in time according to the undulation state information of the road surface, so that the deviation between the vehicle's driving trajectory and the target trajectory is reduced, the lateral control accuracy is improved, and the reliability and comfort of the lane centering control are higher.

[0022] In some embodiments, the vehicle further includes: an on-board communication device, connected to the controller, for communicating with the cloud.

[0023] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0025] Figure 1 is a flow chart of a lane centering control method according to an embodiment of the present invention;

[0026] Figure 2is a schematic diagram of an end-to-end system according to an embodiment of the present invention;

[0027] Figure 3 is a schematic diagram of an implementation of a lane centering function according to an embodiment of the present invention;

[0028] Figure 4 is a flow chart of target return required torque determination according to one embodiment of the present invention;

[0029] Figure 5 is a block diagram of a lane centering controller according to an embodiment of the present invention;

[0030] Figure 6 is a structural block diagram of a vehicle according to an embodiment of the present invention.

[0031] Reference numerals:

[0032] Lane centering controller 100; vehicle 200;

[0033] Processor 101 ; memory 102 ; front view camera 201 ; power steering system 202 ; controller 203 ; vehicle-mounted communication device 204 . DETAILED DESCRIPTION

[0034] Embodiments of the present invention are described in detail below. The embodiments described with reference to the accompanying drawings are exemplary. Embodiments of the present invention are described in detail below.

[0035] The first aspect of the present invention provides a lane centering control method, which can control the vehicle to return to the center position in time according to the undulating state information of the road surface, reduce the deviation between the vehicle's driving trajectory and the target trajectory, and improve the lateral control accuracy and the reliability of the centering control.

[0036] Reference below Figure 1 A lane centering control method according to an embodiment of the first aspect of the present invention is described. Figure 1 As shown, the method at least includes step S1 to step S2.

[0037] Step S1, obtaining an image of the road on which the vehicle is traveling.

[0038] Specifically, during the driving process of the vehicle, the forward-looking camera located in front of the vehicle obtains the current image of the road the vehicle is traveling on in real time. The forward-looking camera can capture the real-time image in front of the vehicle, and then analyze the acquired image for road detection, obstacle recognition, and judging whether the road surface is tilted and undulating.

[0039] Step S2, obtaining the undulation state information of the road surface on which the vehicle is traveling according to the image of the road on which the vehicle is traveling.

[0040] Specifically, after the vehicle acquires an image of the road on which the vehicle is traveling, a fusion analysis is performed on the multiple images of the road on which the vehicle is traveling, and features related to the undulations of the road surface are extracted, such as edge detection, texture analysis, height difference calculation, etc. Finally, the undulation state of the road surface is analyzed using the extracted features and undulation state information of the road on which the vehicle is traveling is generated. The undulation state information of the road on which the vehicle is traveling can be used to determine whether the road surface is flat or undulating.

[0041] Step S3, obtaining the vehicle self-centering torque according to the undulating state information of the road surface on which the vehicle is traveling, so as to control the vehicle to be self-centering according to the vehicle self-centering torque.

[0042] Specifically, the ups and downs of the road surface on which the vehicle is traveling will affect the vehicle's driving state and the stress conditions on the tires, ultimately causing the vehicle to deviate from the center of the lane. Therefore, after obtaining the ups and downs of the road surface on which the vehicle is traveling, the righting torque required for the vehicle under the current road surface is analyzed, and the vehicle's righting torque is finally obtained. The vehicle's righting torque is related to the ups and downs of the road surface, the vehicle's mass, the suspension system, etc. The steering angle of the vehicle is adjusted according to the vehicle's righting torque, and the vehicle is controlled to right itself so that the vehicle can return to the desired driving path.

[0043] According to the lane centering control method of an embodiment of the present invention, after acquiring the vehicle driving road image, the undulation state information of the road surface of the vehicle driving is obtained according to the vehicle driving road image, and finally the vehicle is controlled to return to the center position in time according to the undulation state information of the road surface, so that the deviation between the vehicle driving trajectory and the target trajectory is reduced, and the lateral control accuracy and the reliability of the centering control are improved.

[0044] In some embodiments, obtaining the undulating state information of the road surface of the vehicle driving on the basis of the vehicle driving road image includes: performing two-dimensional lane line recognition, road plane grid estimation and inverse perspective transformation on the vehicle driving road image to obtain three-dimensional lane line information. Determining the height of the lane line in a direction perpendicular to the road surface according to the three-dimensional lane line information; and determining the undulating state information of the lane line according to the height.

[0045] Specifically, during the vehicle driving process, two-dimensional lane line recognition, road plane grid estimation and inverse perspective transformation are performed on the obtained vehicle driving road image; two-dimensional lane line recognition can use image processing technology, such as edge detection, color segmentation or deep learning algorithm (such as convolutional neural network), to identify the two-dimensional lane lines in the vehicle driving road image, extract the pixel coordinates of the lane lines, and track them to form a continuous lane line trajectory; road plane grid estimation usually involves perspective transformation or camera calibration of the road image, and the grid size and direction of the road plane are determined by analyzing the texture, markings or other features in the image; three-dimensional lane line information is obtained after two-dimensional lane line recognition, road plane grid estimation and inverse perspective transformation are performed on the vehicle driving road image, and the coordinates of the lane lines in three-dimensional space are analyzed according to the three-dimensional lane line information, especially the components on the vertical axis (Z axis), and then the height of the lane lines in the direction perpendicular to the road surface is calculated. Finally, the undulation state information of the lane lines is determined according to the height of the lane lines.

[0046] In some embodiments, obtaining the vehicle's aligning torque based on the undulating state information of the road surface on which the vehicle is traveling includes: obtaining the vehicle's aligning torque through a neural network model.

[0047] The input characteristic value of the neural network model is the undulating state information of the road surface on which the vehicle is traveling, and the output characteristic value of the neural network model is the vehicle self-aligning torque.

[0048] Specifically, after obtaining the undulating state information of the road surface on which the vehicle is traveling, the undulating state information of the road surface on which the vehicle is traveling is input into the neural network model, and the neural network model selects the feature that best reflects the undulating state of the road surface from the received data as the input of the neural network, and standardizes and calculates the input data to obtain the vehicle's self-aligning torque, and then the neural network model outputs the vehicle's self-aligning torque.

[0049] For example, Figure 2 As shown, the end-to-end system processes and analyzes the input vehicle driving road image, and finally outputs the return demand torque at the output end.

[0050] The end-to-end system includes: 3D lane line detection system, self-aligning torque estimation system, and torque output system; the 3D lane line detection system receives the video input signal and performs pre-processing such as noise reduction and reshape before sending it to the end neural network for perception and control. Perception includes: 2D lane line recognition, road plane grid estimation, IPM (Inverse Perspective Mapping) projection, feature downsampling, and then the self-aligning torque estimation system outputs the vehicle self-aligning torque, converts the vehicle self-aligning torque to obtain the self-aligning required torque, and finally outputs the self-aligning required torque. Because lane lines are non-compact targets that are difficult to determine the center of, the two-dimensional lane lines adopt an anchor-free solution of a fully convolutional network to improve the detection performance for lane narrowing, widening, and large-angle lane lines, and fit the lane lines through line segments in the grid; because it is necessary to estimate the front wheel self-aligning torque, the three-dimensional lane line regressor is cancelled, and the vehicle CAN (Controller Area Network) signal is directly connected to the network as a network feature parameter and the self-aligning required torque is directly output to achieve end-to-end decision-making and control performance. The end-to-end system optimizes the impact of steering clearance on the system and alleviates the phenomenon of lane centering function drawing dragons.

[0051] In some embodiments, the lane centering control method also includes: obtaining a cloud-calculated self-aligning requirement torque sent from the cloud, wherein the cloud-calculated self-aligning requirement torque is determined based on a steering angle gain proportional factor; obtaining a vehicle-side calculated self-aligning requirement torque based on a vehicle self-aligning torque; determining a target self-aligning requirement torque based on the cloud-calculated self-aligning requirement torque and the vehicle-side calculated self-aligning requirement torque, so as to control the vehicle self-alignment based on the target self-aligning requirement torque.

[0052] Specifically, the cloud calculates the required torque for returning to the center position through the steering angle gain proportional factor. Since there is a virtual position in the steering gear, the actual steering gear angle is actively increased when the steering angle is less than the preset value according to the correction scheme of the steering angle gain proportional factor, thereby offsetting the deviation caused by part of the steering clearance.

[0053] The vehicle's return torque is the torque required by the vehicle's wheels to overcome the ups and downs of the road to ensure that the vehicle travels in a straight line. The return demand torque is the torque required to control the steering gear. When the steering gear is turned with the return demand torque, the vehicle's wheels obtain the vehicle's return torque to make the vehicle travel in a straight line. Therefore, the return demand torque can be estimated or directly mapped by the return torque. Finally, the target return demand torque is obtained by comparing and analyzing the return demand torque calculated on the cloud side and the return demand torque calculated on the vehicle side. The target return demand torque is the torque required by the vehicle steering gear, and the steering gear is controlled to steer with the target return demand torque. The cloud redundant system includes a cloud server and a vehicle-side system. The cloud redundant system stores the traditional lane centering control system. By sending the return demand torque control command of the traditional system to the domain controller for comparison, it is determined whether the end-to-end system is stable and prevents risks.

[0054] According to the following Figure 3 Describe the lane centering function, such as Figure 3 As shown in the figure, the lane centering system includes a front-view camera, a domain controller, a TBOX (Telematics BOX), a cloud server, and an electronic hydraulic power steering system. The front-view camera provides the collected video signal, which is supplied to the downstream system to identify the lane line and control it. The domain controller receives the video signal from the front-view camera and compares it with the vehicle position. It calculates the required return torque through dynamic parameters such as vehicle speed and steering gear ratio and sends it to the vehicle CAN line. The electronic hydraulic power steering system receives the required return torque through the vehicle CAN line and executes it. A traditional lane centering control system is redundant in the cloud server and runs simultaneously with the vehicle-side system. The cloud server sends the calculated return torque to the TBOX, forwards it to the vehicle-side system, and compares the return torques of the two in the domain controller to determine whether the vehicle-side system is stable. TBOX can transmit communication data between the vehicle and the cloud.

[0055] In some embodiments, a target return-to-alignment required torque is determined based on the return-to-alignment required torque calculated on the cloud side and the return-to-alignment required torque calculated on the vehicle side, including: obtaining the torque difference between the return-to-alignment required torque calculated on the cloud side and the return-to-alignment required torque calculated on the vehicle side; when the torque difference is greater than a torque threshold, reminding the driver to take over the vehicle, and performing corresponding operations based on the driver's response to the reminder to take over the vehicle; when the torque difference is less than the torque threshold, the target return-to-alignment required torque is the return-to-alignment required torque calculated on the vehicle side.

[0056] Specifically, the target return torque is obtained by comparing and analyzing the return torque required by the cloud and the return torque required by the vehicle. When the steering gear is controlled by the target return torque required, the torque difference between the return torque required by the cloud and the return torque required by the vehicle is first obtained, and the torque difference is compared with the torque threshold. Since the return torque required by the cloud is determined according to the steering angle gain proportional factor without considering the undulation of the road surface, it deviates from the return torque required by the vehicle, but the deviation value should not be too large. If the torque difference is greater than the torque threshold, at this time, there is a calculation error between the return torque required by the cloud and the return torque required by the vehicle. In order to avoid accidents caused by excessive steering angles of the vehicle, the driver is reminded to take over the vehicle. After the driver takes over, the lane centering control system is exited. If the driver does not take over, the vehicle slows down and stops in a safe area. If the torque difference is less than the torque threshold, the deviation between the return torque required by the cloud and the return torque required by the vehicle is within a reasonable range, and the target return torque required is the return torque required by the vehicle. Dynamic judgment is performed by an algorithm calculated by a non-fixed torque difference threshold, thereby reducing system risks.

[0057] For example, the present invention uses an end-to-end solution to integrate perception and control to perform three-dimensional lane line detection, and includes factors such as lane roll and undulation in the model to reduce their impact; it also estimates the vehicle tire self-aligning torque and performs feedforward control by predicting the front wheel turning angle, thereby improving the problem of rapidity of traditional feedback control; a cloud-based redundant system is designed for the stability of the end-to-end lane centering system, and when there is a large gap between the control output of the end-to-end system and the cloud-based redundant system, an alarm is issued to require the driver to take over the vehicle and check for potential faults in the vehicle system, thereby improving the stability of the overall system.

[0058] Reference below Figure 4 The target return torque requirement determination process is described, including steps S4 to S11.

[0059] Step S4, cloud computing returns the required torque.

[0060] Step S5, sending to the domain controller via TBOX.

[0061] Step S6, judging the torque difference, if the torque difference is greater than a threshold, executing the risk minimization strategy, if the torque difference is less than the threshold, executing step S7.

[0062] Step S7, continue to execute vehicle-side control.

[0063] Step S8, reminding the driver to take over the vehicle.

[0064] Step S9, determine whether to take over, if so, execute S10, otherwise execute S11.

[0065] Step S10, the system exits.

[0066] Step S11, pull over to the side of the road.

[0067] In some embodiments, the torque threshold is obtained by the following formula:

[0068] V = [T1 + (T2 + T') / 2] / 2;

[0069] T' = (T1 + T2) / 2;

[0070] Among them, V is the torque threshold, T1 is the return-to-alignment required torque calculated by the vehicle end, T2 is the return-to-alignment required torque calculated by the cloud end, and T' is the average value of the return-to-alignment required torque calculated by the cloud end and the return-to-alignment required torque calculated by the vehicle end at the previous moment.

[0071] Specifically, a torque threshold is obtained by calculation, and then the torque difference between the return-to-alignment required torque calculated on the cloud side and the return-to-alignment required torque calculated on the vehicle side is compared with the torque threshold to obtain a target return-to-alignment required torque, and the steering gear is controlled to steer with the target return-to-alignment required torque.

[0072] A second aspect of the present invention provides a lane centering controller, such as Figure 5 As shown, the lane centering controller 100 includes: at least one processor 101 and a memory 102 .

[0073] The memory 102 stores a computer program that can be executed by at least one processor 101, and the lane centering control method is implemented when the at least one processor 101 executes the computer program.

[0074] According to the lane centering controller of the embodiment of the present invention, the corresponding lane centering control program can be stored in the memory. When the lane centering control method is implemented, the processor runs the program in the memory, and controls the vehicle to return to the center position in time according to the undulating state information of the road surface, so as to reduce the deviation between the vehicle's driving trajectory and the target trajectory, thereby improving the lateral control accuracy and the reliability of the centering control.

[0075] The third aspect of the present invention provides a non-volatile readable storage medium having a computer program stored thereon, wherein the lane centering control method of the above embodiment is implemented when the computer program is executed.

[0076] A fourth aspect of the present invention provides a vehicle, such as Figure 6 As shown, the vehicle 200 includes: a forward-looking camera 201 , a power steering system 202 , and a controller 203 .

[0077] Among them, the front-view camera 201 is used to collect images of the road the vehicle is traveling on; the power steering system 202 is used to drive the vehicle to return to the center and turn; the controller 203 is connected to the front-view camera 201 and the power steering system 202 respectively, and is used to execute the lane centering control method.

[0078] According to the vehicle of the embodiment of the present invention, after the front-view camera acquires the image of the road the vehicle is traveling on, the undulation state information of the road the vehicle is traveling on is obtained according to the image of the road the vehicle is traveling on, and finally the controller controls the power steering system to drive the vehicle to return to the center position in time according to the undulation state information of the road surface, so that the deviation between the vehicle's driving trajectory and the target trajectory is reduced, the lateral control accuracy is improved, and the reliability and comfort of the lane centering control are higher.

[0079] In some embodiments, Figure 6 As shown, the vehicle 200 includes: an on-board communication device 204 .

[0080] The vehicle-mounted communication device 204 is connected to the controller 203 for communicating with the cloud.

[0081] Specifically, the vehicle-mounted communication device 204 communicates with the cloud to obtain the self-aligning required torque calculated by the cloud, and sends the obtained self-aligning required torque calculated by the cloud to the controller 203, so that the self-aligning required torque calculated by the cloud is subtracted from the self-aligning required torque calculated by the vehicle end to obtain the torque difference, and the torque difference is compared with the torque threshold to obtain the target self-aligning required torque.

[0082] In the description of this specification, any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code including one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present invention belong.

[0083] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0084] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0085] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0086] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0087] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present invention. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

[0088] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, substrates, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example.

[0089] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A lane centering control method, characterized in that: include: Acquire a vehicle driving road image; Obtaining undulating state information of the road surface on which the vehicle is traveling according to the vehicle traveling road image; A vehicle self-centering torque is obtained according to the undulating state information of the road surface on which the vehicle is traveling, so as to control the vehicle to be self-centering according to the vehicle self-centering torque.

2. The lane centering control method according to claim 1, characterized in that: Obtaining the undulating state information of the road surface of the vehicle driving according to the vehicle driving road image, including: The vehicle driving road image is subjected to two-dimensional lane line recognition, road plane grid estimation and inverse perspective transformation to obtain the three-dimensional lane line information. Determine the height of the lane line in a direction perpendicular to the road surface according to the three-dimensional lane line information; The undulation state information of the lane line is determined according to the height.

3. The lane centering control method according to claim 1, characterized in that: Obtaining the vehicle aligning torque according to the undulating state information of the road surface on which the vehicle is traveling, including: The vehicle self-aligning torque is obtained through a neural network model, wherein the input characteristic value of the neural network model is the undulating state information of the road surface on which the vehicle is traveling, and the output characteristic value of the neural network model is the vehicle self-aligning torque.

4. The lane centering control method according to claim 1, further comprising: Obtaining a cloud-calculated return-to-alignment demand torque sent by the cloud, wherein the cloud-calculated return-to-alignment demand torque is determined according to a steering angle gain proportional factor; Obtaining a required torque for calculating the vehicle-side self-alignment according to the vehicle self-alignment torque; A target return-to-centering required torque is determined based on the return-to-centering required torque calculated on the cloud side and the return-to-centering required torque calculated on the vehicle side, so as to control the vehicle return-to-centering according to the target return-to-centering required torque.

5. The lane centering control method according to claim 4, characterized in that: Determining a target return-to-alignment required torque according to the return-to-alignment required torque calculated by the cloud and the return-to-alignment required torque calculated by the vehicle end includes: Obtaining a torque difference between the return-to-alignment required torque calculated by the cloud and the return-to-alignment required torque calculated by the vehicle end; When the torque difference is greater than a torque threshold, prompting the driver to take over the vehicle, and performing corresponding operations according to the driver's response to the prompt to take over the vehicle; When the torque difference is less than a torque threshold, the target return-to-alignment required torque is the return-to-alignment required torque calculated by the vehicle end.

6. The lane centering control method according to claim 5, characterized in that: The torque threshold is obtained by the following formula: V = [T1 + (T2 + T') / 2] / 2; T' = (T1 + T2) / 2; Among them, V is the torque threshold, T1 is the return-to-alignment required torque calculated by the vehicle end, T2 is the return-to-alignment required torque calculated by the cloud end, and T' is the average value of the return-to-alignment required torque calculated by the cloud end and the return-to-alignment required torque calculated by the vehicle end at the previous moment.

7. A lane centering controller, characterized in that: include: at least one processor; a memory communicatively coupled to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, and when the at least one processor executes the computer program, the lane centering control method according to any one of claims 1 to 6 is implemented.

8. A non-volatile readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the lane centering control method according to any one of claims 1 to 6 is implemented.

9. A vehicle, characterized in that: include: A front-view camera, used to collect images of the road on which the vehicle is traveling; Power steering system, used to drive the vehicle to return to the center position; A controller, wherein the controller is connected to the forward-looking camera and the power-assisted steering system, respectively, and is used to execute the lane centering control method according to any one of claims 1 to 6.

10. The vehicle according to claim 9, characterized in that The vehicle further comprises: The vehicle-mounted communication device is connected to the controller and is used for communicating with the cloud.

Images