Vehicle lane changing method, device and vehicle

Abstract

The application discloses a vehicle lane changing method and device and a vehicle. The method comprises the following steps: obtaining driving information of the vehicle; converting the driving information to obtain a plurality of first displacements, wherein the first displacement is used to represent the displacement of the vehicle in a preset direction in different lane changing stages, and the lane changing stage is obtained by dividing the lane changing process of the vehicle; generating a vehicle lane changing curve based on the plurality of first displacements, wherein the vehicle lane changing curve is used to represent the change curve of the steering wheel angle in the process of lane changing; and inputting the vehicle lane changing curve into a steering controller to control the vehicle lane changing. The application solves the technical problem of poor flexibility of the strategy for assisting the driver to change lanes in the related art.

Description

Technical Field

[0001] This invention relates to the field of vehicle control, and more specifically, to a vehicle lane-changing method, apparatus, and vehicle. Background Technology

[0002] In recent years, more and more car manufacturers have chosen lateral driving assistance systems to assist drivers in changing lanes and reduce traffic accidents caused by lane changes. However, since each driver has different lane-changing habits, various factors such as longitudinal and lateral control of the vehicle and speed adjustment need to be considered when changing lanes. This makes it possible that using a single lane-changing assistance strategy may not be able to effectively assist drivers in changing lanes, resulting in poor flexibility in the lane-changing process when using lateral driving assistance systems to assist drivers.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This invention provides a vehicle lane-changing method, apparatus, and vehicle to at least address the technical problem of poor flexibility in strategies for assisting drivers in lane-changing in related technologies.

[0005] According to one aspect of the present invention, a vehicle lane-changing method is provided. The method includes: acquiring vehicle driving information; converting the driving information to obtain a plurality of first displacements, wherein the first displacements are used to characterize the displacement traveled by the vehicle in a preset direction during different lane-changing stages, and the lane-changing stage is obtained by dividing the vehicle's lane-changing process; generating a vehicle lane-changing curve based on the plurality of first displacements, wherein the vehicle lane-changing curve is used to characterize the change curve of the steering wheel's torsional angle during the lane-changing process; and inputting the vehicle lane-changing curve into a steering controller to control the vehicle lane-changing.

[0006] Optionally, generating a vehicle lane-changing curve based on multiple first displacements includes: obtaining the vehicle's current position; determining the lane-changing position corresponding to different lane-changing stages based on the current position and the first displacements; determining the angle determination equation corresponding to the lane-changing position based on the lane-changing position and a preset database, wherein the angle determination equation is used to characterize the change equation of the torsion angle of the vehicle in different lane-changing stages, and the preset database is used to characterize the mapping relationship between the lane-changing position and the default torsion angle of the steering wheel; and concatenating the angle determination equation based on the lane-changing positions corresponding to different lane-changing stages to obtain the vehicle lane-changing curve.

[0007] Optionally, based on the current position and the first displacement, the lane-changing position corresponding to the vehicle in different lane-changing stages is determined, including: based on the current position and the first displacement, the starting position and ending position corresponding to the vehicle in different lane-changing stages are determined; based on the starting position and the ending position, the driving path corresponding to the vehicle in different lane-changing stages is simulated; and the driving path is divided according to a preset division rule to obtain the lane-changing position.

[0008] Optionally, based on the lane change position and a preset database, the angle determination equation corresponding to the lane change position is determined, including: determining the default torsion angle corresponding to the lane change position from the preset database based on the lane change position; and determining the angle determination equation based on the first displacement, the lane change position, and the default torsion angle.

[0009] Optionally, based on the first displacement, lane-changing position, and default torsion angle, an angle determination equation is determined, including: obtaining a preset turning angle equation corresponding to the lane-changing stage, wherein the preset turning angle equation is used to characterize the mapping relationship between the torsion angle and the second displacement at different lane-changing positions, and the second displacement is used to characterize the displacement of the vehicle position relative to the current position in a preset direction; constructing an angle coefficient equation using the mapping relationship between the torsion angle and the default torsion angle, wherein the angle coefficient equation is used to characterize the mapping relationship between the first displacement and the second displacement at different lane-changing positions; and adjusting the preset turning angle equation using the angle coefficient equation to obtain the angle determination equation.

[0010] Optionally, the driving information is transformed to obtain multiple first displacements, including: obtaining the driving speed and road curvature from the driving information; and using a displacement determination model to transform the driving speed and road curvature to obtain multiple first displacements.

[0011] Optionally, the displacement determination model includes an input layer, a hidden layer, and an output layer. The method further includes: obtaining an initial weight matrix and an initial vector threshold, wherein the weight matrix is ​​used to represent the matrix corresponding to the weights between multiple nodes in the input layer, hidden layer, and output layer, and the vector threshold is used to represent the number threshold of output vectors in the hidden layer and output layer; initializing the initial weight matrix and the initial vector threshold to obtain a first weight matrix and a first vector threshold; constructing an initial determination model based on the first weight matrix, the first vector threshold, and a preset number of nodes, wherein the preset number of nodes includes at least: the number of nodes in the input layer, the number of nodes in the hidden layer, and the number of nodes in the output layer; and training the initial determination model using multiple training samples to obtain a displacement determination model.

[0012] Optionally, the training samples include at least labeled driving information and labeled first displacement. The initial determination model is trained using multiple training samples to obtain a displacement determination model, including: inputting the labeled driving information into the initial determination model to determine the first output result of the output layer; determining the output error based on the labeled first displacement, the first output result, and the number of training samples; and determining the trained initial determination model as the displacement determination model in response to the output error being less than a preset error threshold.

[0013] Optionally, the method further includes: in response to an output error not being less than a preset error threshold, obtaining a second output result of the hidden layer; adjusting the first weight matrix and the first vector threshold based on the first output result and the second output result to obtain a second weight matrix and a second vector threshold; adjusting the initial determined model based on the second weight matrix and the second vector threshold to obtain an adjusted determined model; and retraining the adjusted determined model using training samples to obtain a displacement determined model.

[0014] According to another aspect of the present invention, a vehicle lane-changing device is also provided. The device includes: an acquisition module for acquiring vehicle driving information; a conversion module for converting the driving information to obtain a plurality of first displacements, wherein the first displacements characterize the displacements traveled by the vehicle in a preset direction during different lane-changing stages, and the lane-changing stages are obtained by dividing the vehicle's lane-changing process; a generation module for generating a vehicle lane-changing curve based on the plurality of first displacements, wherein the vehicle lane-changing curve characterizes the change curve of the steering wheel's torsion angle during the lane-changing process; and a control module for inputting the vehicle lane-changing curve into a steering controller to control the vehicle's lane-changing.

[0015] Optionally, the generation module includes: a first acquisition unit for acquiring the current position of the vehicle; a first determination unit for determining the lane-changing position of the vehicle at different lane-changing stages based on the current position and a first displacement; a second determination unit for determining the angle determination equation corresponding to the lane-changing position based on the lane-changing position and a preset database, wherein the angle determination equation is used to characterize the change equation of the torsion angle of the vehicle at different lane-changing stages, and the preset database is used to characterize the mapping relationship between the lane-changing position and the default torsion angle of the steering wheel; and a splicing unit for splicing the angle determination equation based on the lane-changing position corresponding to different lane-changing stages to obtain the vehicle lane-changing curve.

[0016] Optionally, the first determining unit includes: a first determining subunit, used to determine the starting position and ending position of the vehicle in different lane-changing stages based on the current position and the first displacement; a simulation subunit, used to simulate the driving path of the vehicle in different lane-changing stages based on the starting position and the ending position; and a dividing subunit, used to divide the driving path according to a preset dividing rule to obtain the lane-changing position.

[0017] Optionally, the second determining unit includes: a second determining subunit, used to determine the default torsion angle corresponding to the lane change position from a preset database based on the lane change position; and an equation determining subunit, used to determine the angle determining equation based on the first displacement, the lane change position, and the default torsion angle.

[0018] Optionally, the equation determination subunit is also used to: obtain the preset turning angle equation corresponding to the lane-changing stage, wherein the preset turning angle equation is used to characterize the mapping relationship equation between the torsion angle and the second displacement at different lane-changing positions, and the second displacement is used to characterize the displacement of the vehicle position and the current position in a preset direction; construct the angle coefficient equation using the mapping relationship between the torsion angle and the default torsion angle, wherein the angle coefficient equation is used to characterize the mapping relationship equation between the first displacement and the second displacement at different lane-changing positions; and adjust the preset turning angle equation using the angle coefficient equation to obtain the angle determination equation.

[0019] Optionally, the conversion module includes: a second acquisition unit for acquiring driving speed and road curvature from driving information; and a conversion unit for converting driving speed and road curvature using a displacement determination model to obtain multiple first displacements.

[0020] Optionally, the displacement determination model includes an input layer, a hidden layer, and an output layer. The device further includes: a parameter acquisition module for acquiring an initial weight matrix and an initial vector threshold, wherein the weight matrix is ​​used to represent the matrix corresponding to the weights between multiple nodes in the input layer, hidden layer, and output layer, and the vector threshold is used to represent the number threshold of output vectors in the hidden layer and output layer; an initialization module for initializing the initial weight matrix and the initial vector threshold to obtain a first weight matrix and a first vector threshold; a model construction module for constructing an initial determination model based on the first weight matrix, the first vector threshold, and a preset number of nodes, wherein the preset number of nodes includes at least the number of nodes in the input layer, the number of nodes in the hidden layer, and the number of nodes in the output layer; and a model training module for training the initial determination model using multiple training samples to obtain the displacement determination model.

[0021] Optionally, the training samples include at least labeled driving information and labeled first displacement. The model training module includes: a first result determination unit, used to input the labeled driving information into the initial determination model and determine the first output result of the output layer; an error determination unit, used to determine the output error based on the labeled first displacement, the first output result, and the number of training samples; and a model determination unit, used to determine the trained initial determination model as the displacement determination model in response to the output error being less than a preset error threshold.

[0022] Optionally, the device further includes: a second result acquisition module, used to acquire a second output result of the hidden layer in response to an output error not being less than a preset error threshold; a parameter adjustment module, used to adjust a first weight matrix and a first vector threshold based on the first output result and the second output result to obtain a second weight matrix and a second vector threshold; a model adjustment module, used to adjust an initially determined model based on the second weight matrix and the second vector threshold to obtain an adjusted determined model; and a retraining module, used to retrain the adjusted determined model using training samples to obtain a displacement determined model.

[0023] According to another aspect of the present invention, a computer-readable storage medium is also provided, including a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the vehicle lane-changing method described above.

[0024] According to another aspect of the present invention, a processor is also provided, wherein the processor's program executes any of the above-described vehicle lane-changing methods.

[0025] According to another aspect of the present invention, a target vehicle is also provided, comprising: one or more processors; a storage device for storing one or more programs; and, when the one or more programs are executed by the one or more processors, causing the one or more processors to perform the vehicle lane-changing method described above.

[0026] In this embodiment of the invention, the method involves acquiring vehicle driving information; converting the driving information to obtain multiple first displacements; generating a vehicle lane-changing curve based on the multiple first displacements; and inputting the vehicle lane-changing curve into a steering controller to control the vehicle lane-changing. By converting the vehicle driving information into multiple first displacements that the vehicle needs to pass through when changing lanes, and generating the vehicle lane-changing curve in real time based on the multiple first displacements, the timeliness of the vehicle lane-changing curve is ensured. Then, the vehicle lane-changing curve is used to control the vehicle or assist the driver in performing the vehicle lane-changing operation, which further improves the flexibility of vehicle lane-changing and solves the technical problem of poor flexibility in the strategy of assisting the driver in performing vehicle lane-changing in related technologies. Attached Figure Description

[0027] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0028] Figure 1 This is a flowchart illustrating a vehicle lane-changing method according to an embodiment of the present invention;

[0029] Figure 2This is a schematic diagram illustrating a vehicle lane-changing displacement according to an embodiment of the present invention;

[0030] Figure 3 This is a structural block diagram of an information processing device according to an embodiment of the present invention. Detailed Implementation

[0031] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0032] It should be noted that the terms "first," "second," etc., 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 the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, 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.

[0033] Example 1

[0034] According to an embodiment of the present invention, a method embodiment of a vehicle lane changing method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0035] Figure 1 This is a flowchart illustrating a vehicle lane-changing method according to an embodiment of the present invention, such as... Figure 1 As shown, the method includes the following steps:

[0036] Step S102: Obtain vehicle driving information.

[0037] The aforementioned driving information can refer to the current vehicle and road information obtained when the vehicle lane-changing system determines that a lane change is required or detects that the driver is currently performing a lane-changing operation. The driving information may include, but is not limited to, vehicle speed, road curvature, vehicle current position and distance from lane lines, etc.

[0038] In one optional embodiment, in order to improve the safety and accuracy of lane changing results, the lane changing system can first obtain the vehicle's driving information, such as the vehicle's current speed, the curvature of the road the vehicle is currently traveling on, and the distance from the vehicle's current position to the lane it needs to change to.

[0039] In one optional embodiment, the lane-changing system can determine whether a lane-changing operation is needed or exists by detecting the road surface condition of the road the vehicle is currently traveling on, or the driver's control information of the vehicle. For example, the lane-changing system can monitor the road surface condition in real time, such as whether there are obstacles within a preset distance in front of the vehicle. If there are, it can determine that the vehicle needs to change lanes and remind the driver to perform the lane-changing operation through preset prompts, such as voice broadcasts or flashing indicator lights, while simultaneously acquiring the aforementioned driving information. The lane-changing system can also monitor the control information received by the vehicle in real time, such as steering wheel angle and vehicle acceleration. If the steering wheel angle is greater than a certain angle threshold and the vehicle acceleration is negative, it indicates that the driver may be performing a lane-changing operation, and the lane-changing system can acquire the aforementioned driving information at this time.

[0040] Step S104: Convert the driving information to obtain multiple first displacements;

[0041] The first displacement is used to characterize the displacement of the vehicle in a preset direction during different lane-changing stages. The lane-changing stage is obtained by dividing the vehicle's lane-changing process.

[0042] The aforementioned lane-changing phase can refer to the division of the entire lane-changing process of the vehicle. In one optional embodiment, the lane-changing process of the vehicle can be divided into multiple phases, which may include, but are not limited to: the lane-changing action generation phase, the vehicle crossing the lane phase, the lane-changing corner recovery phase, and the lane-changing position adjustment phase.

[0043] The aforementioned preset direction can refer to the vehicle's driving direction or a direction perpendicular to the vehicle's driving direction; the aforementioned first displacement can refer to the displacement generated by the vehicle in different preset directions during different lane-changing phases.

[0044] Figure 2 This is a schematic diagram illustrating a vehicle lane-changing displacement according to an embodiment of the present invention, such as... Figure 2 As shown, region 1 represents the lane-changing action generation stage, region 2 represents the vehicle crossing the lane stage, region 3 represents the lane-changing corner recovery stage, and region 4 represents the lane-changing position adjustment stage. Point A represents the initial position of the vehicle before lane changing, point B is the point on the lane line perpendicularly mapped from point A, point D represents the target position of the vehicle after lane changing, point C is the point on the lane line perpendicularly mapped from point D, point E represents the position of the vehicle at the end of the lane-changing action generation stage, and point F is the point on the lane line perpendicularly mapped from point E.

[0045] and Figure 2 Corresponding to the lane-changing phase, the aforementioned first displacement can refer to the displacement of the vehicle in different areas in a preset direction. Taking the preset direction as the direction perpendicular to the driving direction as an example, the aforementioned first displacement can at least include the difference between the distance from point A to point B and the distance from point E to point F in area 1.

[0046] In one optional embodiment, considering that the displacement generated by the driver controlling the vehicle to change lanes varies under different driving conditions, for example, if the vehicle's current speed is high and the road curvature is large, the final displacement generated by the driver controlling the vehicle to change lanes, such as the distance from point C to point D, may be large. In order to ensure driving safety, the driver may adjust the vehicle's lane-changing position to the target position, that is, reduce the distance from point C to point D. If the driving speed is slow and the road curvature is small, the final displacement generated, such as the distance from point C to point D, may be small. In order to ensure driving safety, the driver may adjust the vehicle's lane-changing position to the target position, that is, increase the distance from point C to point D. Therefore, in order to ensure that the lane change result is more in line with the driver's expectations and to ensure the safety of the vehicle when changing lanes, the vehicle lane change system can first simulate the multiple displacements that the driver may produce in different lane change stages under the current situation, based on the acquired driving information, as well as the displacements that need to be adjusted to ensure vehicle safety. In other words, the vehicle lane change system can convert the acquired driving information into the aforementioned multiple first displacements.

[0047] Step S106: Generate vehicle lane-changing curves based on multiple first displacements.

[0048] Among them, the vehicle lane change curve is used to characterize the change in steering wheel torsion angle during the lane change process.

[0049] In one optional embodiment, in order to ensure the smoothness of lane changing and avoid driver discomfort caused by abnormal lane changing, such as excessive steering angle during lane changing, the vehicle lane changing system can generate one or more vehicle lane changing curves based on the multiple first displacements obtained by conversion. These curves can be used to represent the curves that make the steering wheel torsion angle change smoothly during lane changing.

[0050] In one optional embodiment, if multiple lane-changing curves are generated, the lane-changing system can select a target lane-changing curve from the multiple lane-changing curves based on driving information, thereby improving the efficiency of lane-changing while ensuring lane-changing safety.

[0051] For example, considering that a vehicle lane change curve can be used to represent the steering wheel torsion angle, a range of steering wheel torsion angles can be determined first based on the vehicle's current speed and the distance from the current position to the obstacle in front of the vehicle. Then, from multiple vehicle lane change curves, the curve that fits the range of torsion angles and has a moderate trend of torsion angle change can be selected as the target vehicle lane change curve.

[0052] Step S108: Input the vehicle lane change curve into the steering controller to control the vehicle to change lanes.

[0053] After obtaining the vehicle lane change curve, the vehicle lane change system can input the curve into the vehicle's steering controller to control the vehicle's lane change.

[0054] In one optional embodiment, the vehicle lane-changing system can also control or assist the driver in performing lane-changing operations based on the vehicle's driving status.

[0055] For example, if the vehicle is currently controlled by the driver, the lane-changing system can monitor the difference between the received steering wheel torque and the vehicle's steering angle in the lane-changing curve in real time. If the difference is large, the lane-changing system can appropriately increase the steering wheel torque to prevent the driver from turning the steering wheel too far, which would affect the driving experience. If the difference is small, the lane-changing system can reduce the steering wheel torque to prevent the vehicle from changing direction too little during lane changes, which could lead to a collision with obstacles in front of the vehicle.

[0056] If the vehicle is currently in autonomous driving mode, the lane-changing system can directly control the vehicle to steer and change lanes according to the lane-changing curve, without the need for further steering adjustments, such as changing the steering wheel torque, thereby improving the efficiency of lane changing.

[0057] In this embodiment of the invention, the method involves acquiring vehicle driving information; converting the driving information to obtain multiple first displacements; generating a vehicle lane-changing curve based on the multiple first displacements; and inputting the vehicle lane-changing curve into a steering controller to control the vehicle lane-changing. By converting the vehicle driving information into multiple first displacements that the vehicle needs to pass through when changing lanes, and generating the vehicle lane-changing curve in real time based on the multiple first displacements, the timeliness of the vehicle lane-changing curve is ensured. Then, the vehicle lane-changing curve is used to control the vehicle or assist the driver in performing the vehicle lane-changing operation, which further improves the flexibility of vehicle lane-changing and solves the technical problem of poor flexibility in the strategy of assisting the driver in performing vehicle lane-changing in related technologies.

[0058] Optionally, generating a vehicle lane-changing curve based on multiple first displacements includes: obtaining the vehicle's current position; determining the lane-changing position corresponding to different lane-changing stages based on the current position and the first displacements; determining the angle determination equation corresponding to the lane-changing position based on the lane-changing position and a preset database, wherein the angle determination equation is used to characterize the change equation of the torsion angle of the vehicle in different lane-changing stages, and the preset database is used to characterize the mapping relationship between the lane-changing position and the default torsion angle of the steering wheel; and concatenating the angle determination equation based on the lane-changing positions corresponding to different lane-changing stages to obtain the vehicle lane-changing curve.

[0059] The aforementioned angle determination equation can refer to the equation for the change of the torsional angle as the vehicle travels from the previous lane-changing position, such as the current position, to the next lane-changing position. For example, if there are three lane-changing positions, position 1, position 2, and position 3, then the angle determination equation can refer to the equation for the change of the torsional angle as the vehicle travels from position 1 to position 2, and the equation for the change of the torsional angle as the vehicle travels from position 2 to position 3.

[0060] The aforementioned preset database may refer to a database obtained using big data, which consists of different lane-changing positions and corresponding steering wheel angles when the driver controls the vehicle to change lanes. The corresponding steering wheel angle may refer to the aforementioned default torsion angle.

[0061] In one optional embodiment, considering that the steering wheel angle of the vehicle changes differently in different lane-changing phases, for example, in the aforementioned lane-crossing phase, the vehicle steering wheel may be turned clockwise with a larger turning angle, while in the lane-changing angle recovery phase, the vehicle steering wheel may be turned counterclockwise with a smaller turning angle, and in the lane-changing position adjustment phase, the vehicle steering wheel may be turned clockwise first and then counterclockwise with a smaller turning angle.

[0062] Therefore, to ensure the stability and smoothness of the lane-changing process as much as possible, several relatively important lane-changing positions can be extracted from different lane-changing stages, such as the starting or ending points of the lane-changing stage. Using these lane-changing positions and a preset database, the aforementioned angle determination equation is determined. Then, these angle determination equations are concatenated to obtain the vehicle lane-changing curve. This ensures that the vehicle will not experience significant lane-changing problems at any time or position during the lane-changing process, thus preventing disruption to the driver's normal driving. The process of determining the angle determination equation using lane-changing positions and a preset database is shown below.

[0063] For example, the steering torque set by the vehicle's lane-changing system may differ in different lane-changing phases. For instance, the steering torque during the lane-crossing phase may be greater than that during the lane-changing action initiation phase. If a uniform angle determination equation is currently used to control the vehicle's lane-changing, the driver may be unable to adapt to the sudden change in angle torque when transitioning from the lane-changing action initiation phase to the lane-crossing phase, resulting in vehicle jolting. Therefore, it is possible to determine the angle determination equations corresponding to different lane-changing positions in different lane-changing phases through the above process, and then concatenate these equations to obtain the vehicle's lane-changing curve, thereby avoiding the problem of poor applicability of the vehicle's lane-changing process due to a single lane-changing strategy.

[0064] Optionally, based on the current position and the first displacement, the lane-changing position corresponding to the vehicle in different lane-changing stages is determined, including: based on the current position and the first displacement, the starting position and ending position corresponding to the vehicle in different lane-changing stages are determined; based on the starting position and the ending position, the driving path corresponding to the vehicle in different lane-changing stages is simulated; and the driving path is divided according to a preset division rule to obtain the lane-changing position.

[0065] In one optional embodiment, the vehicle control system can first acquire a plan view of the road surface corresponding to the current driving road of the vehicle. Then, based on the first displacement obtained above, such as the displacement in the driving direction and the displacement in the direction perpendicular to the driving direction, the system determines the starting position and ending position of the vehicle in different lane-changing stages. The starting position and ending position are then projected onto the plan view. Simultaneously, based on the starting position and ending position of different lane-changing stages, the system simulates a relatively flat driving path for the vehicle in that lane-changing stage. For example, the minimum curvature of the driving path is greater than a preset curvature threshold. Then, according to preset division rules, such as selecting the point with the largest curvature in the driving path or selecting a preset number of equally divided points in the driving path, the system divides the driving path to obtain the lane-changing position in different lane-changing stages.

[0066] It should be noted that the above preset conditions are merely illustrative and are not intended to impose any specific limitations.

[0067] Optionally, based on the lane change position and a preset database, the angle determination equation corresponding to the lane change position is determined, including: determining the default torsion angle corresponding to the lane change position from the preset database based on the lane change position; and determining the angle determination equation based on the first displacement, the lane change position, and the default torsion angle.

[0068] In one optional embodiment, in order to improve the accuracy of obtaining the angle determination equation, the steering wheel angle corresponding to different lane change positions can be determined from the preset database first, i.e., the default torsion angle, and then the first displacement, lane change position and default torsion angle can be used to determine the angle determination equation.

[0069] Specifically, taking the equation for determining the angle corresponding to different lane-changing positions during the lane-changing action phase as an example, if the determined lane-changing position is the starting position, the ending position, and two preset displacement division points of this lane-changing phase, then the steering wheel angle determined according to the preset database, i.e., the default torsion angle, can be as shown in Table 1 below:

[0070] Table 1

[0071]

[0072] Table 1 is a table showing the relationship between lane-changing position and steering wheel angle according to an embodiment of this application. Taking the direction perpendicular to the driving direction as the aforementioned preset direction as an example, the aforementioned first displacement can refer to the lateral displacement of the vehicle relative to the starting position in the lane-changing stage. The corresponding two displacement division points can refer to the two division points generated when the lateral displacement is divided into three equal parts. By mapping these two division points to the aforementioned driving path, the two lane-changing positions in the lane-changing stage can be obtained.

[0073] As shown in Table 1, using a preset database, the first displacement corresponding to lane change position 1 can be determined as Y. LC00 The corresponding steering wheel angle is θ S00 The first displacement corresponding to lane change position 2 is The corresponding steering wheel angle is θ S0a The first displacement corresponding to lane change position 3 is The corresponding steering wheel angle is θ S0b The first displacement corresponding to lane change position 4 is Y. LC01 The corresponding steering wheel angle is θ S01 Among them, the steering wheel angle θ S00 θ S0a θ S0b and θ S01These are all default torsional angles determined through big data statistics. After determining the steering wheel angle corresponding to the lane change position using a preset database, the first displacement, steering wheel angle, and lane change position mentioned above can be used to determine the corresponding angle determination equation.

[0074] Optionally, based on the first displacement, lane-changing position, and default torsion angle, an angle determination equation is determined, including: obtaining a preset turning angle equation corresponding to the lane-changing stage, wherein the preset turning angle equation is used to characterize the mapping relationship between the torsion angle and the second displacement at different lane-changing positions, and the second displacement is used to characterize the displacement of the vehicle position relative to the current position in a preset direction; constructing an angle coefficient equation using the mapping relationship between the torsion angle and the default torsion angle, wherein the angle coefficient equation is used to characterize the mapping relationship between the first displacement and the second displacement at different lane-changing positions; and adjusting the preset turning angle equation using the angle coefficient equation to obtain the angle determination equation.

[0075] The aforementioned second displacement can refer to the displacement that the vehicle actually travels in the same direction as the first displacement.

[0076] In one optional embodiment, in order to improve the efficiency of obtaining the angle determination equation, different preset steering angle equations can be configured for different lane-changing stages to represent the relationship between the steering wheel torsion angle and the second displacement and the default torsion angle.

[0077] Continuing with the example of four lane-changing positions corresponding to the lane-changing action generation stage in Table 1 above, let's assume that the preset turning angle equation θ(Y) is:

[0078]

[0079] in, This represents the equation for the angle coefficients, θ k This represents the default torsion angle. Considering that when the vehicle travels to the lane-changing position k, i.e., the second displacement equals the first displacement, the following needs to be satisfied:

[0080] θ(Y k )=θ k (k = 0, 1, ..., 3),

[0081] Then the corresponding This requires satisfying:

[0082]

[0083] Therefore, we can assume at this point:

[0084]

[0085] Then, when i = k, It can be deduced that:

[0086]

[0087] Correspondingly, Z k Resubstitute to From the assumed formula, we can obtain:

[0088]

[0089] Furthermore, by adjusting the preset angle equation, i.e., θ(Y) mentioned above, using the angle coefficient equation, the angle determination equation corresponding to the lane-changing action phase can be obtained as F. 01 (Y):

[0090]

[0091] In one optional embodiment, after determining the angle determination equations corresponding to different lane-changing stages according to the above process, multiple torsion angle change equations can be further spliced ​​together according to the order of different lane-changing stages to obtain the above-mentioned vehicle lane-changing curve.

[0092] Optionally, the driving information is transformed to obtain multiple first displacements, including: obtaining the driving speed and road curvature from the driving information; and using a displacement determination model to transform the driving speed and road curvature to obtain multiple first displacements.

[0093] In one optional embodiment, considering that the vehicle's speed and the road curvature of the current road have a significant impact on the vehicle's lane-changing process, in the process of converting driving information to obtain multiple first displacements, the driving speed and road curvature in the driving information can be obtained first, and then the driving speed and road curvature can be converted using a preset displacement determination model to obtain multiple first displacements.

[0094] Optionally, the displacement determination model includes an input layer, a hidden layer, and an output layer. The method further includes: obtaining an initial weight matrix and an initial vector threshold, wherein the weight matrix is ​​used to represent the matrix corresponding to the weights between multiple nodes in the input layer, hidden layer, and output layer, and the vector threshold is used to represent the number threshold of output vectors in the hidden layer and output layer; initializing the initial weight matrix and the initial vector threshold to obtain a first weight matrix and a first vector threshold; constructing an initial determination model based on the first weight matrix, the first vector threshold, and a preset number of nodes, wherein the preset number of nodes includes at least: the number of nodes in the input layer, the number of nodes in the hidden layer, and the number of nodes in the output layer; and training the initial determination model using multiple training samples to obtain a displacement determination model.

[0095] In one alternative embodiment, the displacement determination model described above can be obtained by training an initial determination model using multiple training samples.

[0096] Specifically, the displacement determination model can include an input layer, a hidden layer, and an output layer. When constructing the displacement determination model, we can first obtain the weight matrices representing the weight values ​​between multiple nodes in the input, hidden, and output layers, as well as the vector thresholds used to limit the number of output vectors in the hidden and output layers. That is, we can first obtain two initial weight matrices ω. ik and ω kj , where ω ik This represents the matrix corresponding to the weight values ​​between the i-th node in the input layer and the k-th node in the hidden layer, ω. kj This represents the matrix corresponding to the weight values ​​between the k-th node of the hidden layer and the j-th node of the output layer; and two initial vector thresholds a are obtained. j and b k , where a j This represents the threshold corresponding to the j-th node in the hidden layer, b k This corresponds to the threshold of the kth node in the output layer.

[0097] It should be noted that the matrix elements in the initial weight matrix mentioned above can be any value, and there are no restrictions here.

[0098] In one optional embodiment, in order to improve the efficiency of obtaining the displacement determination model, the initial weight matrix and the initial vector threshold can be initialized in a normal distribution manner to obtain the first weight matrix and the first vector threshold, so that the mean of the first weight matrix and the first vector threshold is a first preset value, such as 0, and the variance is a second preset value, such as 0.01.

[0099] Then, an initial model is constructed according to the preset number of nodes, the first weight matrix, and the first vector threshold. The preset number of nodes includes at least the number of nodes in the input layer, the number of nodes in the hidden layer, and the number of nodes in the output layer, and the preset number of nodes corresponds to the first weight matrix.

[0100] To facilitate training, the number of nodes in the input layer can be set to n=3, the number of nodes in the hidden layer to m=20, and the number of nodes in the output layer to p=3.

[0101] After constructing the initial deterministic model described above, multiple training samples can be input into the initial deterministic model to train the model and obtain the displacement deterministic model.

[0102] Optionally, the training samples include at least labeled driving information and labeled first displacement. The initial determination model is trained using multiple training samples to obtain a displacement determination model, including: inputting the labeled driving information into the initial determination model to determine the first output result of the output layer; determining the output error based on the labeled first displacement, the first output result, and the number of training samples; and determining the trained initial determination model as the displacement determination model in response to the output error being less than a preset error threshold.

[0103] In one optional embodiment, the training samples may include, but are not limited to, labeled driving information and labeled first displacement. When training the initial determination model using the training samples, the labeled driving information can be input into the initial determination model to determine the first output result of the output layer. The formula for the first output result may be:

[0104]

[0105] Among them, O j This represents the first output result, U. k This represents the second output result of the hidden layer.

[0106] After obtaining the first output result, the output error of the current first output result can be determined by further utilizing the labeled first displacement, the first output result, and the number of training samples. The formula for the output error can be:

[0107]

[0108] Where ρ represents the output error, g represents the number of training samples, and Y... j It can refer to the first displacement.

[0109] In one optional embodiment, if the output error is less than a preset error threshold, it can be determined that the current model has high accuracy, and the trained model can be determined as the displacement determination model described above.

[0110] In one alternative embodiment, the aforementioned preset threshold may refer to the aforementioned variance.

[0111] Optionally, the method further includes: in response to the output error being not less than a preset error threshold, obtaining a second output result of the hidden layer; adjusting the first weight matrix and the first vector threshold based on the first output result and the second output result to obtain a second weight matrix and a second vector threshold; adjusting the initial determined model based on the second weight matrix and the second vector threshold to obtain an adjusted determined model; and retraining the adjusted determined model using training samples to obtain a displacement determined model.

[0112] In one optional scheme of this embodiment, if the output error is not less than a preset error threshold, the aforementioned first weight matrix and first vector threshold can be adjusted to obtain a second weight matrix and a second vector threshold. The initial determined model is then adjusted using the second weight matrix and the second vector threshold to obtain an adjusted determined model. The adjusted determined model is then retrained using training samples to finally obtain a displacement determined model with an output error less than the preset error threshold.

[0113] In one optional embodiment, the first weight matrix and the first vector threshold can be adjusted using the first output result from the output layer and the second output result from the hidden layer to obtain the second weight matrix and the second vector threshold.

[0114] Specifically, the formula for adjusting the first weight matrix can be:

[0115]

[0116] Where, x i It represents

[0117] The formula for adjusting the first vector threshold can be:

[0118]

[0119] Optionally, the method further includes: obtaining the driver's identity information; determining lane-changing features based on the identity information, wherein the lane-changing features are used to characterize the features generated when the driver controls the vehicle to change lanes; and dividing the vehicle's lane-changing process based on the lane-changing features to obtain lane-changing stages.

[0120] In one optional solution of this embodiment, considering that different drivers have different driving habits, if the lane changing process is divided into lane changing stages according to a uniform division method, the divided lane changing stages may not meet the driver's expectations and affect the driving experience. Therefore, when dividing the lane changing process to obtain lane changing stages, the driver's identity information can be obtained first, and the lane changing features corresponding to the driver's driving habits can be determined from a preset feature library using the identity information.

[0121] Example 2

[0122] According to another aspect of the embodiments of the present invention, corresponding to the above-described vehicle lane-changing method embodiments, this specification also provides a vehicle lane-changing device, please refer to... Figure 3 , Figure 3This is a structural block diagram of an information processing device according to an embodiment of the present invention. The device includes: an acquisition module 302 for acquiring vehicle driving information; a conversion module 304 for converting the driving information to obtain multiple first displacements, wherein the first displacements are used to characterize the displacements traveled by the vehicle in a preset direction during different lane-changing stages, and the lane-changing stages are obtained by dividing the vehicle's lane-changing process; a generation module 306 for generating a vehicle lane-changing curve based on the multiple first displacements, wherein the vehicle lane-changing curve is used to characterize the change curve of the steering wheel's torsion angle during the lane-changing process; and a control module 308 for inputting the vehicle lane-changing curve into a steering controller to control the vehicle's lane-changing.

[0123] Optionally, the generation module 306 includes: a first acquisition unit for acquiring the current position of the vehicle; a first determination unit for determining the lane-changing position of the vehicle at different lane-changing stages based on the current position and a first displacement; a second determination unit for determining the angle determination equation corresponding to the lane-changing position based on the lane-changing position and a preset database, wherein the angle determination equation is used to characterize the change equation of the torsion angle of the vehicle at different lane-changing stages, and the preset database is used to characterize the mapping relationship between the lane-changing position and the default torsion angle of the steering wheel; and a splicing unit for splicing the angle determination equation based on the lane-changing position corresponding to different lane-changing stages to obtain the vehicle lane-changing curve.

[0124] Optionally, the first determining unit includes: a first determining subunit, used to determine the starting position and ending position of the vehicle in different lane-changing stages based on the current position and the first displacement; a simulation subunit, used to simulate the driving path of the vehicle in different lane-changing stages based on the starting position and the ending position; and a dividing subunit, used to divide the driving path according to a preset dividing rule to obtain the lane-changing position.

[0125] Optionally, the second determining unit includes: a second determining subunit, used to determine the default torsion angle corresponding to the lane change position from a preset database based on the lane change position; and an equation determining subunit, used to determine the angle determining equation based on the first displacement, the lane change position, and the default torsion angle.

[0126] Optionally, the equation determination subunit is further configured to: obtain the position determination equation corresponding to the lane-changing phase, wherein the position determination equation is used to characterize the relationship equation between the second displacement and the vehicle position during the lane-changing phase, and the second displacement is used to characterize the displacement of the vehicle position relative to the current position in a preset direction; construct an angle simulation equation using the first displacement and the second displacement, wherein the angle simulation equation is used to characterize the relationship equation between the torsion angle and the target distance simulated according to the training samples, and the training samples are used to train the displacement determination model; and adjust the position determination equation using the angle simulation equation, the lane-changing position, and the default torsion angle to obtain the angle determination equation.

[0127] Optionally, the conversion module 304 includes: a second acquisition unit for acquiring driving speed and road curvature from driving information; and a conversion unit for converting driving speed and road curvature using a displacement determination model to obtain multiple first displacements.

[0128] Optionally, the displacement determination model includes an input layer, a hidden layer, and an output layer. The device further includes: a parameter acquisition module for acquiring an initial weight matrix and an initial vector threshold, wherein the weight matrix is ​​used to represent the matrix corresponding to the weights between multiple nodes in the input layer, hidden layer, and output layer, and the vector threshold is used to represent the number threshold of output vectors in the hidden layer and output layer; an initialization module for initializing the initial weight matrix and the initial vector threshold to obtain a first weight matrix and a first vector threshold; a model construction module for constructing an initial determination model based on the first weight matrix, the first vector threshold, and a preset number of nodes, wherein the preset number of nodes includes at least the number of nodes in the input layer, the number of nodes in the hidden layer, and the number of nodes in the output layer; and a model training module for training the initial determination model using multiple training samples to obtain the displacement determination model.

[0129] Optionally, the training samples include at least labeled driving information and labeled first displacement. The model training module includes: a first result determination unit, used to input the labeled driving information into the initial determination model and determine the first output result of the output layer; an error determination unit, used to determine the output error based on the labeled first displacement, the first output result, and the number of training samples; and a model determination unit, used to determine the trained initial determination model as the displacement determination model in response to the output error being less than a preset error threshold.

[0130] Optionally, the device further includes: a second result acquisition module, used to acquire a second output result of the hidden layer in response to an output error not being less than a preset error threshold; a parameter adjustment module, used to adjust a first weight matrix and a first vector threshold based on the first output result and the second output result to obtain a second weight matrix and a second vector threshold; a model adjustment module, used to adjust an initially determined model based on the second weight matrix and the second vector threshold to obtain an adjusted determined model; and a retraining module, used to retrain the adjusted determined model using training samples to obtain a displacement determined model.

[0131] Example 3

[0132] According to another aspect of the present invention, a computer-readable storage medium is also provided, including a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform any of the above-described vehicle lane-changing methods.

[0133] Example 4

[0134] According to another aspect of the present invention, a processor is also provided, wherein the processor's program executes any of the above-described vehicle lane-changing methods.

[0135] Example 5

[0136] According to another aspect of the present invention, a target vehicle is also provided, comprising: one or more processors; a storage device for storing one or more programs; and, when the one or more programs are executed by the one or more processors, causing the one or more processors to perform the vehicle lane-changing method described above.

[0137] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0138] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0139] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be 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 displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0140] 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 units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0141] 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 as a software functional unit.

[0142] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0143] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A vehicle lane-changing method, characterized in that, The method includes: Obtain vehicle driving information; The driving information is converted to obtain multiple first displacements, wherein the first displacement is used to characterize the displacement traveled by the vehicle in a preset direction during different lane-changing stages, and the lane-changing stage is obtained by dividing the lane-changing process of the vehicle. A vehicle lane-changing curve is generated based on the plurality of first displacements, wherein the vehicle lane-changing curve is used to characterize the change curve of the steering wheel torsion angle during the lane-changing process. The vehicle lane-changing curve is input into the steering controller to control the vehicle to change lanes; The method of generating a vehicle lane-changing curve based on the plurality of first displacements includes: obtaining the current position of the vehicle; determining the lane-changing position of the vehicle corresponding to different lane-changing stages based on the current position and the first displacements; determining an angle determination equation corresponding to the lane-changing position based on the lane-changing position and a preset database, wherein the angle determination equation is used to characterize the change equation of the torsion angle of the vehicle in different lane-changing stages, and the preset database is used to characterize the mapping relationship between the lane-changing position and the default torsion angle of the steering wheel; and concatenating the angle determination equation based on the lane-changing position corresponding to the different lane-changing stages to obtain the vehicle lane-changing curve. Based on the lane change position and a preset database, the angle determination equation corresponding to the lane change position is determined, including: determining the default torsion angle corresponding to the lane change position from the preset database based on the lane change position; and determining the angle determination equation based on the first displacement, the lane change position, and the default torsion angle. Based on the first displacement, the lane-changing position, and the default torsion angle, the angle determination equation is determined, including: obtaining a preset turning angle equation corresponding to the lane-changing stage, wherein the preset turning angle equation is used to characterize the mapping relationship between the torsion angle and the second displacement at different lane-changing positions, and the second displacement is used to characterize the displacement of the vehicle position relative to the current position in the preset direction; constructing an angle coefficient equation using the mapping relationship between the torsion angle and the default torsion angle, wherein the angle coefficient equation is used to characterize the mapping relationship between the first displacement and the second displacement at different lane-changing positions; and adjusting the preset turning angle equation using the angle coefficient equation to obtain the angle determination equation.

2. The method according to claim 1, characterized in that, Based on the current position and the first displacement, determining the lane-changing position of the vehicle corresponding to the different lane-changing stages includes: Based on the current position and the first displacement, the starting position and ending position of the vehicle in the different lane-changing stages are determined; Based on the starting position and the ending position, simulate the driving path of the vehicle at different lane-changing stages; The driving path is divided according to a preset division rule to obtain the lane change position.

3. The method according to claim 1, characterized in that, The driving information is converted to obtain multiple first displacements, including: Obtain the driving speed and road curvature from the driving information; The driving speed and road curvature are converted using a displacement determination model to obtain the multiple first displacements.

4. The method according to claim 3, characterized in that, The displacement determination model includes an input layer, a hidden layer, and an output layer, and the method further includes: Obtain an initial weight matrix and an initial vector threshold, wherein the weight matrix is ​​used to represent the matrix corresponding to the weights between multiple nodes in the input layer, the hidden layer and the output layer, and the vector threshold is used to represent the threshold number of output vectors in the hidden layer and the output layer; The initial weight matrix and the initial vector threshold are initialized to obtain the first weight matrix and the first vector threshold; Based on the first weight matrix, the first vector threshold, and the preset number of nodes, an initial determination model is constructed, wherein the preset number of nodes includes at least: the number of nodes in the input layer, the number of nodes in the hidden layer, and the number of nodes in the output layer; The initial determination model is trained using multiple training samples to obtain the displacement determination model.

5. The method according to claim 4, characterized in that, The training samples at least contain labeled driving information and labeled first displacement. The initial determination model is trained using multiple training samples to obtain the displacement determination model, including: The labeled driving information is input into the initial determination model to determine the first output result of the output layer; The output error is determined based on the labeled first displacement, the first output result, and the number of training samples. In response to the output error being less than a preset error threshold, the trained initial determination model is determined as the displacement determination model.

6. The method according to claim 5, characterized in that, The method further includes: In response to the output error being no less than the preset error threshold, the second output result of the hidden layer is obtained; Based on the first output result and the second output result, the first weight matrix and the first vector threshold are adjusted to obtain the second weight matrix and the second vector threshold; Based on the second weight matrix and the second vector threshold, the initial determined model is adjusted to obtain the adjusted determined model; The adjusted determination model is retrained using the training samples to obtain the displacement determination model.

7. A vehicle comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the vehicle lane-changing method according to any one of claims 1-6.

Images