Vehicle control method and device, electronic equipment and automatic driving vehicle

Abstract

The invention provides a vehicle control method and device, electronic equipment and an automatic driving vehicle, and relates to the technical field of artificial intelligence, in particular to the technical field of automatic driving. According to the specific implementation scheme, the vehicle control method comprises the steps that a first initial control sequence is generated by correcting a historical control sequence generated at historical moments on the basis of the deviation between an expected driving state and an actual driving state of a target vehicle at the current moment; wherein the expected driving state at the current moment is determined according to a historical control sequence; and under the dynamic constraint and the driving cost constraint, iteration processing is carried out on the first initial control sequence to obtain a target control sequence, and the target control sequence is used for controlling the driving of the target vehicle.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, in particular to the field of autonomous driving technology, and specifically to vehicle control methods, devices, electronic equipment and autonomous driving vehicles. Background Art

[0002] With the widespread application of artificial intelligence technology in the field of autonomous driving, controlling the vehicle to complete lane keeping, lane changing, obstacle avoidance and other driving operations according to changes in the vehicle's driving environment needs to meet both the safety and comfort requirements of autonomous driving and the real-time requirements of autonomous driving. Summary of the invention

[0003] The present disclosure provides a vehicle control method, device, electronic equipment and an autonomous driving vehicle.

[0004] According to one aspect of the present disclosure, a vehicle control method is provided, comprising: generating a first initial control sequence by correcting a historical control sequence generated at a historical moment based on a deviation between an expected driving state and an actual driving state of a target vehicle at a current moment; wherein the expected driving state at the current moment is determined based on the historical control sequence; and iteratively processing the first initial control sequence under dynamic constraints and driving cost constraints to obtain a target control sequence, wherein the target control sequence is used to control the driving of the target vehicle.

[0005] According to another aspect of the present disclosure, a vehicle control device is provided, including: a correction module and an optimization module.

[0006] The correction module is used to generate a first initial control sequence by correcting a historical control sequence generated at a historical moment based on the deviation between the expected driving state of the target vehicle at the current moment and the actual driving state; wherein the expected driving state at the current moment is determined according to the historical control sequence.

[0007] The optimization module is used to iteratively process the first initial control sequence under the dynamic constraint and the driving cost constraint to obtain a target control sequence, and the target control sequence is used to control the driving of the target vehicle.

[0008] According to another aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described above.

[0009] According to another aspect of the present disclosure, an autonomous driving vehicle including the above-mentioned electronic device is provided.

[0010] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method described above.

[0011] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the method described above is implemented.

[0012] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0014] Figure 1 An exemplary system architecture to which the vehicle control method and device according to an embodiment of the present disclosure can be applied is schematically shown;

[0015] Figure 2 A flow chart schematically shows a vehicle control method according to an embodiment of the present disclosure;

[0016] Figure 3 A schematic diagram schematically shows a method of correcting a historical control sequence according to an embodiment of the present disclosure;

[0017] Figure 4 A schematic diagram schematically shows a control of a vehicle to completely track a reference lane line according to an embodiment of the present disclosure;

[0018] Figure 5 A schematic diagram schematically shows a method of controlling a vehicle to track a reference lane line according to an actual driving state of the vehicle according to an embodiment of the present disclosure;

[0019] Figure 6 A schematic diagram schematically shows a vehicle tracking reference line constraint according to an embodiment of the present disclosure;

[0020] Figure 7 A schematic diagram schematically shows a safety corridor constraint of a vehicle according to an embodiment of the present disclosure;

[0021] Figure 8 A schematic diagram of obstacle constraints according to an embodiment of the present disclosure is schematically shown;

[0022] Fig. 9 An exemplary framework diagram of an application vehicle control method according to an embodiment of the present disclosure is schematically shown;

[0023] Fig.10Schematically shows an exemplary framework diagram of an application vehicle control method according to another embodiment of the present disclosure;

[0024] Fig.11 A block diagram schematically shows a vehicle control device according to an embodiment of the present disclosure; and

[0025] Fig.12 A block diagram of an electronic device suitable for implementing a vehicle control method according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0026] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0027] The iLQR (Iterative Linear Quadratic Regulator) algorithm can improve the accuracy of vehicle control by iteratively linearizing the vehicle's kinematic model and quadraticizing the cost function. However, since the driving environment of the vehicle is complex and changeable during driving, the convergence speed is slow when the iLQR algorithm is used for iterative processing based on the randomly generated initial control sequence, resulting in poor timeliness in controlling the vehicle.

[0028] In view of this, the embodiment of the present disclosure corrects the historical control sequence generated at the historical moment based on the deviation between the actual driving state and the expected driving state of the target vehicle, and iterates the corrected control sequence as the initial control sequence, thereby further improving the convergence speed. Under the premise of meeting the control accuracy, the efficiency of real-time control of the vehicle is improved.

[0029] Figure 1 An exemplary system architecture to which the vehicle control method and apparatus according to an embodiment of the present disclosure can be applied is schematically shown.

[0030] It should be noted that Figure 1 The examples shown are only examples of system architectures to which the embodiments of the present disclosure can be applied, in order to help those skilled in the art understand the technical content of the present disclosure, but do not mean that the embodiments of the present disclosure cannot be used in other devices, systems, environments or scenarios. For example, in another embodiment, an exemplary system architecture to which the vehicle control method and apparatus can be applied may include a terminal device, but the terminal device may implement the vehicle control method and apparatus provided by the embodiments of the present disclosure without interacting with a server.

[0031] like Figure 1 As shown, the system architecture 100 according to this embodiment may include an autonomous driving vehicle 101, a network 102, and a server 103. The network 102 is used to provide a medium for a communication link between the autonomous driving vehicle 101 and the server 103. The connection type of the network 102 may be a wireless communication link.

[0032] A variety of sensors may be configured on the autonomous driving vehicle 101 to collect driving environment information. Then, the collected driving environment information is sent to the server 103 via the network 102. The server 103 may generate a target control sequence by executing the vehicle control method of the embodiment of the present disclosure, and send the target control sequence to the autonomous driving vehicle 101 to control the driving of the autonomous driving vehicle.

[0033] The vehicle control method provided in the embodiment of the present disclosure may also be generally executed by an electronic device configured on the autonomous driving vehicle 101. Accordingly, the vehicle control device provided in the embodiment of the present disclosure may also be provided in the autonomous driving vehicle 101.

[0034] It should be understood that Figure 1 The number of autonomous driving vehicles, networks, and servers in the embodiment is only for illustration purposes. Any number of autonomous driving vehicles, networks, and servers may be provided as required.

[0035] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision, disclosure and application of user personal information involved comply with the provisions of relevant laws and regulations, take necessary confidentiality measures, and do not violate public order and good morals.

[0036] In the technical solution of the present disclosure, the user's authorization or consent is obtained before obtaining or collecting the user's personal information.

[0037] Figure 2 The flowchart of the vehicle control method according to the embodiment of the present disclosure is schematically shown.

[0038] like Figure 2 As shown, the method 200 includes operations S210 to S220.

[0039] In operation S210, a first initial control sequence is generated by correcting a historical control sequence generated at a historical moment based on a deviation between an expected driving state and an actual driving state of the target vehicle at a current moment.

[0040] In operation S220, the first initial control sequence is iteratively processed under the dynamic constraint and the driving cost constraint to obtain a target control sequence, and the target control sequence is used to control the driving of the target vehicle.

[0041] According to an embodiment of the present disclosure, the expected driving state at the current moment is determined based on a historical control sequence. The control sequence represents a time sequence of control information. The historical control sequence may be the result of iterative processing of the initial control sequence at the previous moment using the iLQR algorithm at the previous sampling moment.

[0042] For example, a historical control sequence may include control information at T moments, and the control information may include information such as the steering angle and speed of the steering wheel at each moment. The time interval between moments may be preconfigured or dynamically adjusted according to actual needs.

[0043] For example, the planning duration may be T=6s, the number of planning steps may be N=30, and the time interval between each moment may be △t=T / N=0.5s.

[0044] According to an embodiment of the present disclosure, the driving status may include the driving position, the front direction, the driving speed, the turning angle of the steering wheel, etc. of the target vehicle.

[0045] For example: the historical control sequence can be [u0,u1,u2,…,u 29 ], according to the historical control sequence, the expected driving state of the vehicle at each moment can be determined as [x0, x1, x2,…, x 29 ]. Since the time interval between each moment is △t=0.5s, the expected driving state of the vehicle should be x1. However, when t0+0.25s, the historical control sequence can be corrected according to the deviation between the current driving state x* of the vehicle and the expected driving state to reduce the direct use of the historical control sequence as the initial control sequence of the next moment. Since the timestamps of the initial moment are not aligned, the obtained control sequence has delays or deviations.

[0046] In some embodiments, dynamic constraints may be used to constrain the relationship between the driving position, driving speed, heading angle, and curvature in the driving state of the vehicle.

[0047] In some embodiments, the driving cost constraint can be used to constrain the vehicle's driving trajectory in terms of driving safety, driving comfort, rationality of the driving trajectory, and the matching degree between the driving operation and the driving trajectory.

[0048] For example, the first initial control sequence can be used as the initial solution of the iLQR algorithm to start the iterative processing operation. Under the dual constraints of the dynamic constraints and the driving cost constraints, the iteration is returned until the output result converges or the maximum number of iterations is reached, and the target control sequence is obtained to control the vehicle driving at the next moment.

[0049] According to an embodiment of the present disclosure, based on the deviation between the actual driving state and the expected driving state of the target vehicle, the historical control sequence generated at the historical moment is corrected, and the corrected control sequence is used as the initial control sequence for iterative processing, thereby further improving the convergence speed. Under the premise of satisfying the control accuracy, the efficiency of real-time control of the vehicle is improved.

[0050] In some embodiments, the control sequence generated at the previous moment may be corrected based on the actual driving state at the current moment to align the timestamp of the initial control sequence at the current moment.

[0051] For example: the historical control sequence includes control information of T moments, where T is an integer greater than 1; based on the deviation between the expected driving state and the actual driving state of the target vehicle at the current moment, the first initial control sequence is generated by correcting the historical control sequence generated at the historical moment, which may include the following operations: according to the control information at the t-1th moment, determine the expected driving state at the tth moment, where t=2, 3, ..., T-1; based on the deviation between the expected driving state at the tth moment and the actual driving state at the tth moment, interpolate between the control information at each adjacent moment in the historical control sequence to generate the first initial control sequence.

[0052] Figure 3 A schematic diagram of correcting a historical control sequence according to an embodiment of the present disclosure is schematically shown.

[0053] like Figure 3 As shown, the t-1th moment may be the previous sampling moment adjacent to the current sampling moment (the tth moment) in the sampling period.

[0054] For the historical control sequence [u0,u1,u2, u3,u4] output at time t-1, the historical control sequence determines the expected driving state at time t. In actual application scenarios, the time interval between adjacent moments in the historical control sequence is greater than the time interval between time t-1 and time t.

[0055] For example, the time interval between each moment in the historical control sequence may be 0.5s. The time interval between the t-1th moment and the tth moment may be 0.25s. Therefore, if the historical control sequence output at the tth moment is directly used as the first initial control sequence for iteratively generating the target control sequence for the tth moment, there is obviously a large delay deviation.

[0056] For example, the historical control sequence [u0,u1,u2, u3,u4] output at time t-1 can represent the control information u0 at 0.5s, the control information u1 at 1.0s, the control information u3 at 1.5s, and the control information u4 at 2.0s. The expected driving state x0 from 0.5s to 1s can be determined based on the control information u0 at 0.5s (t-1s), and the expected driving state x1 from 1s to 1.5s can be determined based on the control information u1 at 1.0s. In the actual driving process of the vehicle, the vehicle may start to iteratively generate the target control sequence for 0.75s when it reaches 0.75s (ts).

[0057] Since the historical control sequence lacks the control information of the 0.75th second, and only includes the control information of the 0.5th second and the 1st second adjacent to the 0.75th second, if the historical control sequence is directly used as the first initial control sequence of this round of iteration, there will obviously be a large delay deviation. Therefore, based on the deviation between the actual driving state x* of the vehicle at the 0.75th second and the expected driving state x0 of the vehicle at the 0.75th second, for example, the difference between the actual driving state at the 0.75th second and the expected driving state at the 0.75th second can be used as an interpolation coefficient, and then, based on the interpolation coefficient, the value used to insert between the historical control sequence of the 0.5th second and the historical control sequence of the 1st second is obtained, and the historical control sequence is corrected, thereby reducing the delay deviation.

[0058] In actual application scenarios, the historical control sequence output at time t-1 can be a control sequence for controlling the vehicle to travel at 0.5s, 1s, 1.5s, etc. The moment when the vehicle generates the control sequence is within the time period corresponding to the historical control sequence. For example, at 1s, the control sequence for traveling at 1.5s, 2s, 2.5s, etc. will be regenerated. By analogy, therefore, when interpolating, the historical control sequence can be interpolated based on the deviation between the actual driving state at the moment that has occurred and the expected driving state. The moment that has occurred must be earlier than the last moment in the historical control sequence.

[0059] Therefore, the historical control sequence can be interpolated based on the actual driving state of the vehicle at the tth moment, and the interpolated control sequence can be used as the first initial control sequence to reduce the direct use of the historical control sequence as the initial control sequence at the next moment. Due to the misalignment of the timestamps at the initial moment, the obtained control sequence will have delays or deviations.

[0060] In some embodiments, the delay deviation between the control sequence output at the previous moment and the initial control sequence at the current moment can be reduced by interpolating the control information between adjacent moments.

[0061] For example: based on the deviation between the expected driving state at the tth moment and the actual driving state at the tth moment, interpolation is performed between the control information at each adjacent moment in the historical control sequence to generate a first initial control sequence, which may include the following operations: determining the expected driving state at the t+1th moment according to the control information at the tth moment; obtaining an interpolation coefficient according to the ratio of the deviation to the difference between the deviation and the expected driving state; and interpolating between the control information at each adjacent moment in the historical control sequence based on the interpolation coefficient and the control information at each moment to obtain the first initial control sequence.

[0062] According to an embodiment of the present disclosure, the expected driving state difference indicates a difference between the expected driving state at the t-th time and the expected driving state at the t+1-th time.

[0063] For example, in the historical control sequence [u0,u1,u2, u3,u4] output at time t-1, the expected driving state [x0,x1,x2, x3,x4] at each moment in the future period can be determined. Interpolation can be performed between the control information u0 and the control information u1 according to formula (1).

[0064] (1)

[0065] Among them, r represents the interpolation coefficient, x* represents the actual driving state at the tth moment, x0 represents the expected driving state at the tth moment, and x1 represents the expected driving state at the t+1th moment; u0 represents the control information at the tth moment, u1 represents the control information at the t+1th moment; u* represents the interpolation point, that is, the control information at the tth moment after correction in the first initial control sequence.

[0066] Similarly, for [u1,u2, u3,u4] in the historical control sequence, the interpolation coefficient r calculated based on formula (1) can be used to interpolate between control information u1 and control information u2, between control information u2 and control information u3, and between control information u3 and control information u4 to obtain the first initial control sequence.

[0067] According to an embodiment of the present disclosure, based on the deviation between the actual driving state at the current moment and the expected driving state determined based on the historical control sequence, interpolation is performed between the control information at adjacent moments in the historical control sequence, thereby reducing the error between the initial control series at the current moment and the actual driving state of the vehicle, and further improving the convergence speed of iteratively generating the target control sequence.

[0068] In some embodiments, in addition to correcting the historical control sequence, an initial control sequence for performing iterative operations may be generated in combination with a reference lane line in the vehicle driving environment. For example: based on a reference lane line associated with the target vehicle's driving intention, a second initial control sequence for controlling the target vehicle to travel along the reference lane line is generated; based on the objective function, an initial control sequence for performing iterative operations is determined from the first initial control sequence and the second initial control sequence.

[0069] According to an embodiment of the present disclosure, the reference lane line associated with the target vehicle's driving intention may represent a reference lane line associated with the driving environment in which the target vehicle is currently located.

[0070] For example, when the target vehicle is traveling in a straight area, the reference lane line may be the center line of the lane where the vehicle is located.

[0071] For example, in a driving scenario where the target vehicle is changing lanes or turning, the reference lane line may be the center line of the opposite lane opposite to the vehicle's driving direction.

[0072] The second initial control sequence may be a control sequence for setting the vehicle to completely track the reference lane line, or may be a control sequence for controlling the vehicle to track the reference lane line based on the actual driving state of the vehicle.

[0073] In some embodiments, a target control sequence for iterative operation may be determined from the first initial control sequence and the second initial control sequence based on a target function.

[0074] According to the embodiments of the present disclosure, the objective function can be constructed based on the needs of the actual application scenario, taking into account aspects such as trajectory tracking accuracy, driving safety, and comfort.

[0075] For example, the objective function may include a reference line tracking cost function, a target state cost function, a driving comfort cost function, and a safety cost function. The specific form of the objective function can be set based on actual application requirements, as long as it can constrain trajectory tracking accuracy, driving safety, and comfort.

[0076] In some embodiments, based on the objective function, determining an initial control sequence for performing an iterative operation from a first initial control sequence and a second initial control sequence may include the following operations: processing the first initial control sequence based on the objective function to generate a first generation value; processing the second initial control sequence based on the objective function to generate a second generation value; in response to determining that the first generation value is less than the second generation value, determining the first initial control sequence as the initial control sequence; in response to determining that the first generation value is greater than or equal to the second generation value, determining the second initial control sequence as the initial control sequence.

[0077] For example, the objective function is used to calculate the function value of the first initial control sequence and the function value of the second initial control sequence respectively, and the initial control sequence with the smaller function value is used to perform the iterative operation.

[0078] According to an embodiment of the present disclosure, by combining the strategy of tracking the reference lane line and the strategy of correcting the historical control sequence, the objective function is used to calculate the driving costs of the initial control sequences obtained based on different strategies, thereby further improving the convergence speed of the iterative processing of the initial control sequence and improving the control efficiency.

[0079] For scenarios where lane lines are clear and the driving environment is relatively simple, such as in the straight-ahead area of ​​a highway, a strategy of setting the vehicle to completely track the reference lane lines can be applied.

[0080] In some embodiments, based on a reference lane line associated with the target vehicle's driving intention, a second initial control sequence for controlling the target vehicle to travel along the reference lane line is generated, which may include the following operations: extracting each reference position at each reference moment from the reference lane line; and generating a second initial control sequence based on the curvature change state of each reference position on the reference lane line.

[0081] Figure 4 A schematic diagram of controlling a vehicle to completely track a reference lane line according to an embodiment of the present disclosure is schematically shown.

[0082] like Figure 4 As shown, the reference lane line in the schematic diagram includes 5 reference positions, and the intervals between the reference positions can be preconfigured.

[0083] In this embodiment, it is assumed that the vehicle can completely track the reference lane line, which can be understood as the vehicle driving completely according to the five reference positions marked in the schematic diagram, driving to the reference position Pa at the tth time, and driving to the reference position Pb at the t+1th time. Therefore, the second initial control sequence can be generated based on the curvature change of each reference position. As shown in formula (2):

[0084] (2)

[0085] in, represents the curvature of the reference position at the t+1th reference time; represents the curvature of the reference position at the t-th reference time; U represents the second initial control sequence; represents the rate of change of curvature at time t.

[0086] According to an embodiment of the present disclosure, a second initial control sequence is generated based on the curvature change of the lane line, which has a higher degree of matching with the driving scene than related examples and control sequences with random configurations. Therefore, it converges faster during iterative processing.

[0087] For scenarios with low lane line clarity and complex driving environments, such as turning or changing lanes, a strategy for controlling the vehicle to track the reference lane line can be applied.

[0088] In some embodiments, a second initial control sequence for controlling the target vehicle to travel along the reference lane line associated with the target vehicle's driving intention is generated, which may include the following operations: generating a driving state at each moment based on the target vehicle's driving state at the current moment; determining a target reference position at each moment from the reference lane line based on the driving state at each moment; and processing the driving state at each moment and the target reference position at each moment to generate a second initial control sequence.

[0089] Combine the following Figure 5 The process of generating the control information at time t is described in detail.

[0090] Figure 5 A schematic diagram of controlling a vehicle to track a reference lane line according to an actual driving state of the vehicle according to an embodiment of the present disclosure is schematically shown.

[0091] like Figure 5 As shown in the diagram, the front wheel of the vehicle is located at point B, and the rear wheel of the vehicle is located at point A. The target reference position of the vehicle is point T, and the distance between point A and point T is the preview distance l d , the distance between point A and point B represents the wheelbase L of the vehicle. α represents the heading angle of the target reference position point T relative to the current position of the vehicle. δ represents the steering angle of the front wheels of the vehicle. θ represents the heading angle of the front wheels of the vehicle.

[0092] In some embodiments, the driving state at each moment and the target reference position at each moment are processed to generate a second initial control sequence, which may include the following operations: determining the angle of the target reference position at the tth moment relative to the driving position at the tth moment as the heading angle of the target vehicle at the tth moment, t=2, 3, ..., T-1; T is an integer greater than 1; based on the geometric relationship between the wheel steering angle and the vehicle heading angle, performing a geometric transformation on the heading angle at the tth moment to generate the wheel steering angle at the tth moment; based on the wheel steering angle at the tth moment and a predetermined wheelbase, generating the curvature at the tth moment; and generating the second initial control sequence based on the curvature change state between the tth moment and the t-1th moment.

[0093] In some embodiments, the control information at time t may be calculated according to equation (3).

[0094] (3)

[0095] in, Indicates the orientation angle of the target reference position T at time t relative to the current position of the vehicle; represents the steering angle of the front wheels of the vehicle at time t; L represents the wheelbase; represents the preview distance at the tth moment; represents the curvature of the driving position at time t; Indicates the curvature of the traveling position at time t-1.

[0096] In some embodiments, the target reference position at time t+1 may be obtained by recursively calculating along the reference lane line based on the driving speed at time t.

[0097] According to an embodiment of the present disclosure, due to the complex driving environment, a second initial control sequence is recursively obtained by tracking the reference lane line based on the actual driving state of the vehicle, which better matches the actual driving state of the vehicle at the next moment and further improves the convergence speed during iterative processing.

[0098] When the iLQR algorithm is used to solve the vehicle trajectory planning problem, the trajectory planning problem can be expressed as shown in the following formula (4).

[0099] (4)

[0100] Among them, J represents the objective function, st is the constraint condition, is the expected driving state at the kth moment, is the expected driving state at the initial moment, represents the driving cost at the target time, Indicates the driving cost during the driving process, The dynamic constraints representing the expected driving state, represents the dynamic constraints of the control information, and N is the number of maximum planning time steps.

[0101] In the embodiment of the present disclosure, the expected driving state can be expressed as , the control information can be expressed as , the dynamic constraint can be expressed as formula (5).

[0102] (5)

[0103] in, represents the coordinates of the vehicle's expected driving position, Indicates the driving speed, represents the orientation angle, represents the curvature of the driving position, Represents the rate of change of curvature.

[0104] In the embodiment of the present disclosure, the expected driving state may include an expected position and an expected orientation angle; the reference driving state may include a reference position and a reference orientation angle. The objective function may include: a reference line tracking cost function, a target state cost function, a driving comfort cost function, and a safety cost function.

[0105] According to an embodiment of the present disclosure, the reference line tracking cost function is used to constrain the position difference between the expected position at each moment and the reference position at each moment and the angle difference between the expected orientation angle at each moment and the reference orientation angle at each moment.

[0106] In the embodiment of the present disclosure, the position difference between the expected position at each moment and the reference position at each moment can be represented by the projection distance between the expected position at each moment and the reference position at each moment.

[0107] Figure 6 A schematic diagram of a vehicle tracking reference line constraint according to an embodiment of the present disclosure is schematically shown.

[0108] like Figure 6 As shown in Figure 1, Pa represents the current actual position of the vehicle. T represents the reference position on the reference lane line corresponding to the current moment. θt represents the orientation angle of the front wheel of the vehicle at the current moment. Indicates the reference front wheel orientation angle when the vehicle is at the reference position on the reference lane line corresponding to the current moment. d indicates the rotation angle from Pa to P T The projected distance on the tangent line of the reference lane line.

[0109] In the embodiment of the present disclosure, the reference line tracking cost function can be expressed as shown in formula (6).

[0110] (6)

[0111] in, represents the reference line tracking cost function, d i It represents the projection distance between the expected position at the i-th moment and the reference position at the i-th moment. It represents the angle difference between the expected orientation angle at the i-th moment and the reference orientation angle at the i-th moment. Represents the reference line tracking cost weight.

[0112] According to an embodiment of the present disclosure, the target state cost function is used to constrain the error between the expected driving state at the target moment and the target driving state.

[0113] In the embodiment of the present disclosure, the target state cost function can be expressed as shown in formula (7).

[0114] (7)

[0115] in, represents the target state cost function, represents the horizontal coordinate of the target position, The ordinate of the target position. represents the target heading angle, represents the curvature of the target position, The curvature change rate of the target position relative to the driving position at the previous moment, The horizontal coordinate representing the expected position at the target time, The ordinate represents the expected position at the target time. represents the expected heading angle at the target moment, represents the curvature of the expected position at the target moment, It represents the curvature change rate of the expected position at the target moment relative to the expected position at the previous moment.

[0116] According to an embodiment of the present disclosure, the driving comfort cost function is used to constrain the curvature change of the expected driving state at each moment.

[0117] In the embodiment of the present disclosure, the driving comfort cost function can be expressed as shown in formula (8).

[0118] (8)

[0119] in, represents the driving comfort cost function, κ i represents the curvature of the expected position at the i-th moment, dκ i Indicates the rate of change of the curvature of the expected position at the i-th moment relative to the curvature of the expected position at the i-1-th moment. represents the curvature weight, The weight representing the rate of change of curvature.

[0120] According to an embodiment of the present disclosure, the safety cost function is used to constrain the safe driving range of the expected driving state at each moment.

[0121] In some embodiments, the safety cost function may include safety corridor constraints, dynamic constraints of the control sequence, and obstacle constraints.

[0122] Figure 7 A schematic diagram of a safety corridor constraint of a vehicle according to an embodiment of the present disclosure is schematically shown.

[0123] like Figure 7 As shown, in this embodiment, polygons can be used to construct the safety corridor, for example: hexagons are used to construct the hard constraints of the safety corridor (the solid line area in the figure), and quadrilaterals are used to construct the soft constraints of the safety corridor (the dotted line area in the figure).

[0124] For example, from the reference line, sample s (s>1) reference points at a certain distance, find the maximum distance point that can be reached on the left and right of the reference point in the l direction, for example, a circle diameter distance from the obstacle, connect every n points to form a polygon, until the entire trajectory is traversed, and all the sampled polygons form a safe corridor. At this time, the center circle of the front and rear axles of the vehicle is required to be inside the polygon, which is converted into the constraint of the following formula (9). The fourth power of the distance to the straight line is used to punish those that do not meet the constraint. This further increases the flexibility and precision of the safe corridor construction.

[0125] (9)

[0126] in, represents the safety corridor constraint; when the driving position is inside the polygon representing the hard constraint, is 0 when the driving position is outside the polygon representing the hard constraint but inside the polygon representing the soft constraint. is the distance from the driving position to the edge of the polygon representing the hard constraint; Represents the safety corridor constraint weight.

[0127] In some embodiments, the dynamic constraints of the control sequence can represent the range constraints on the vehicle's driving position, orientation angle, curvature, and curvature change rate. The specific range of change can be configured according to actual needs, and the embodiments of the present disclosure do not specifically limit this.

[0128] In some embodiments, an envelope ellipse may be used to construct an obstacle constraint. For example, when the driving position of the vehicle is inside the ellipse, it is determined that there is a risk of collision between the vehicle and the obstacle.

[0129] Figure 8 A schematic diagram of obstacle constraints according to an embodiment of the present disclosure is schematically shown.

[0130] like Figure 8 As shown, the center points of multiple circles can be used to represent the position of the obstacle. Point O represents the center point closest to the vehicle along the touch direction, which can be used to determine whether there is a risk of collision between the current driving position of the vehicle and the obstacle.

[0131] For example, the obstacle constraint can be constructed according to formula (10).

[0132]

[0133] (10)

[0134] in, represents obstacle constraints, represents the vector from the center of the ellipse to the vehicle's driving position, P represents a matrix with the inverse of the square of the major and minor axes of the ellipse as diagonal elements, Represents the obstacle constraint weight.

[0135] The objective function can be expressed as formula (11):

[0136] (11)

[0137] in, is the reference line tracking cost function, is the target state cost function, is the driving comfort cost function, To ensure safe corridors, To control the dynamic constraints of the sequence, Constrained by obstacles.

[0138] According to the embodiments of the present disclosure, the accuracy of trajectory tracking, driving comfort and driving safety are comprehensively considered to construct an objective function, thereby further improving the accuracy of vehicle control.

[0139] In order to reduce the difference between the target control sequence output at the current moment and the historical control sequence output at the previous moment, in some embodiments, the objective function also includes: a historical trajectory tracking cost function, which is used to constrain the position difference between the expected position at each moment and the historical expected position at each moment and the angle difference between the expected orientation angle at each moment and the historical expected orientation angle at each moment.

[0140] For example, the historical trajectory tracking cost can be calculated according to formula (12).

[0141] (12)

[0142] in, represents the historical trajectory tracking cost function, s i Represents the projected distance between the expected position at the i-th moment and the historical expected position at the i-th moment. i It represents the angle difference between the expected heading angle at the i-th moment and the historical expected heading angle at the i-th moment. Represents the historical trajectory tracking weight.

[0143] In some embodiments, in order to further improve control efficiency, the amount of historical control information to be extracted from the historical control sequence may be determined based on the actual driving state of the vehicle.

[0144] When the actual driving state of the vehicle has a large curvature change, the amount of historical control information extracted is small. For example, if the historical control sequence includes the historical control information of the 10th to 20th seconds, only the historical control information of the 17th to 20th seconds can be extracted to participate in the calculation of the historical trajectory tracking cost.

[0145] When the actual driving state of the vehicle is a curvature change of small, the amount of historical control information extracted is large. For example, only the historical control information of eight moments from 12s to 20s can be extracted to participate in the calculation of the historical trajectory tracking cost.

[0146] According to an embodiment of the present disclosure, introducing the historical trajectory tracking cost in the objective function can reduce the difference between the target control sequence output at the current moment and the historical control sequence output at the previous moment, thereby reducing the occurrence of abnormal driving behaviors, such as: turning the steering wheel at a very fast speed or driving in an S-shaped trajectory on the road.

[0147] Fig. 9 An exemplary framework diagram of an application vehicle control method according to an embodiment of the present disclosure is schematically shown.

[0148] like Fig. 9 As shown, the exemplary framework 900 may include an information input module 910 , a problem construction module 920 , an optimization module 930 , and a result evaluation and post-processing module 940 .

[0149] The information input module 910 is used to input rule parameters 911, solution parameters 912 and vehicle parameters 913. Rule parameters 911 include but are not limited to thresholds in dynamic constraints, planning time parameters, planning step parameters, etc. Solution parameters 912 include but are not limited to control sequences, driving state sequences, etc. Vehicle parameters include but are not limited to wheelbase.

[0150] The problem construction module 920 can be used to construct a kinematic model 921, calculate an initial control sequence 922, construct a cost function 923, and construct constraint conditions 924, etc.

[0151] The optimization module 930 can be used to update the main vehicle state setting / solve the control sequence 931, perform local linearization 932 using the iLQR algorithm, iteratively process the initial control sequence by back propagation 933 and forward propagation 934 until convergence to obtain the target control sequence. The process of iteratively processing the initial control sequence using the iLQR algorithm is a relatively mature solution, and the embodiments of the present disclosure will not be described in detail here.

[0152] The result evaluation and post-processing module 940 can evaluate the control result based on the mean variance of curvature / curvature change 941, deviation from the target point / reference lane line 942, steering wheel speed / iteration number 943, etc. It can also participate in a new round of algorithm iteration by interpolation or starting a spare initial control sequence to further optimize the target control sequence.

[0153] Fig.10 An exemplary framework diagram of an application vehicle control method according to another embodiment of the present disclosure is schematically shown.

[0154] like Fig.10 As shown, the exemplary framework 1000 may include a testing module 1010 and an optimizer 1020 .

[0155] The test module 1010 may be used to perform operation S1011 of reading a configuration file, operation S1012 of building a test tool and running a test, and operation S1013 of storing a test result.

[0156] The optimizer 1020 is a core module for performing the above operation S1012. It may include a pre-processing module 1030, a solution module 1040, and a post-processing module 1050. The optimizer 1020 is configured with a pre-built optimization path and a solution path 1021, so that the pre-processing module 1022, the solution module 1023, and the post-processing module 1024 perform the operation of generating a target control sequence according to the optimization path and the solution path 1021.

[0157] The preprocessing module 1030 is used to perform operation S1031 to construct reference lines, safety corridors, obstacle boundaries and historical control sequences; operation S1032 to calculate the starting driving state and initial control sequence and construct reference points and pointers; operation S1033 to perform initialization to start the iLQR algorithm.

[0158] The solution module 1040 is used to perform operation S1041 to initialize the problem to be optimized, and perform operation S1042 to calculate the gradient and driving cost to obtain the target control sequence. Operation S1041 may include operations such as constructing driving costs and constraints, calculating the dynamic changes of each reference line, and calculating the initial control sequence. Operation S1042 may include executing the iLQR algorithm through back propagation and forward propagation to update the initial control sequence until convergence.

[0159] The problem to be optimized 1041 may include a reference line tracking cost 1041_1 , a target state cost 1041_2 , a driving comfort cost 1041_3 , a safety cost constraint 1041_4 , and a dynamic constraint 1041_5 , which corresponds to the objective function in the vehicle control method described above.

[0160] Fig.11A block diagram of a vehicle control device according to an embodiment of the present disclosure is schematically shown.

[0161] like Fig.11 As shown, the vehicle control device 1100 may include a correction module 1110 and an optimization module 1120 .

[0162] The correction module 1110 is used to generate a first initial control sequence by correcting the historical control sequence generated at the historical moment based on the deviation between the expected driving state and the actual driving state of the target vehicle at the current moment; wherein the expected driving state at the current moment is determined according to the historical control sequence.

[0163] The optimization module 1120 is used to iteratively process the first initial control sequence under the dynamic constraint and the driving cost constraint to obtain a target control sequence, and the target control sequence is used to control the driving of the target vehicle.

[0164] According to an embodiment of the present disclosure, the correction module may include: a first determination submodule and an interpolation submodule.

[0165] The first determination submodule is used to determine the expected driving state at time t according to the control information at time t-1, where t=2, 3, ..., T-1.

[0166] The interpolation submodule is used to interpolate between the control information of each adjacent moment in the historical control sequence based on the deviation between the expected driving state at the tth moment and the actual driving state at the tth moment to generate a first initial control sequence.

[0167] According to an embodiment of the present disclosure, the interpolation submodule includes a first determining unit, a first obtaining unit and a first interpolation unit.

[0168] The first determining unit is used to determine the expected driving state at time t+1 according to the control information at time t.

[0169] The first obtaining unit is used to obtain an interpolation coefficient according to a ratio of the deviation to an expected driving state difference; wherein the expected driving state difference indicates a difference between an expected driving state at time t and an expected driving state at time t+1.

[0170] The first interpolation unit is used to interpolate between the control information at each adjacent moment in the historical control sequence based on the interpolation coefficient and the control information at each moment, so as to obtain a first initial control sequence.

[0171] According to an embodiment of the present disclosure, the above-mentioned device also includes: a generating module and a determining module.

[0172] The generating module is used to generate a second initial control sequence for controlling the target vehicle to travel along the reference lane line according to the reference lane line associated with the driving intention of the target vehicle.

[0173] The determination module is used to determine an initial control sequence for performing an iterative operation from a first initial control sequence and a second initial control sequence based on an objective function.

[0174] According to an embodiment of the present disclosure, the generation module includes an extraction submodule and a first generation submodule.

[0175] The extraction submodule is used to extract each reference position at each reference time from the reference lane line.

[0176] The first generating submodule is used to generate a second initial control sequence according to the curvature change state of each reference position on the reference lane line.

[0177] According to an embodiment of the present disclosure, the generation module includes: a second generation submodule, a second determination submodule and a third generation submodule.

[0178] The second generating submodule is used to generate the driving state at each moment based on the driving state of the target vehicle at the current moment.

[0179] The second determination submodule is used to determine the target reference position at each moment from the reference lane line based on the driving state at each moment.

[0180] The third generating submodule is used to process the driving state at each moment and the target reference position at each moment to generate a second initial control sequence.

[0181] According to an embodiment of the present disclosure, the driving state includes: a driving position and a wheel steering angle of the target vehicle. The third generation submodule includes: a second determination unit, a conversion unit, a first generation unit and a second generation unit.

[0182] The second determination unit is used to determine the angle of the target reference position at the tth moment relative to the driving position at the tth moment as the heading angle of the target vehicle at the tth moment, t=2, 3, ..., T-1; T is an integer greater than 1.

[0183] The conversion unit is used to perform geometric conversion on the heading angle at the tth moment based on the geometric relationship between the wheel steering angle and the vehicle heading angle, so as to generate the wheel steering angle at the tth moment.

[0184] The first generating unit is used to generate the curvature at the tth moment based on the wheel steering angle at the tth moment and the predetermined wheelbase.

[0185] The second generating unit is used to generate a third initial control sequence based on the curvature change state at the tth moment and the t-1th moment.

[0186] According to an embodiment of the present disclosure, the determination module includes a first processing unit, a second processing unit and a third determination unit.

[0187] The first processing unit is used to process the first initial control sequence based on the objective function to generate a first generation value.

[0188] The second processing unit is used to process the second initial control sequence based on the objective function to generate a second generation value.

[0189] The third determining unit is used to determine the first initial control sequence as the initial control sequence in response to determining that the first generation value is less than the second generation value; and to determine the second initial control sequence as the initial control sequence in response to determining that the first generation value is greater than or equal to the second generation value.

[0190] According to an embodiment of the present disclosure, the expected driving state includes an expected position and an expected orientation angle; the reference driving state includes a reference position and a reference orientation angle. The objective function includes: a reference line tracking function, a target state cost function, a driving comfort cost function, and a safety cost function.

[0191] The reference line tracking cost function is used to constrain the position difference between the expected position at each moment and the reference position at each moment and the angle difference between the expected orientation angle at each moment and the reference orientation angle at each moment.

[0192] The target state cost function is used to constrain the error between the expected driving state at the target time and the target driving state.

[0193] The driving comfort cost function is used to constrain the curvature change of the expected driving state at each moment.

[0194] The safety cost function is used to constrain the safe driving range of the expected driving state at each moment.

[0195] According to an embodiment of the present disclosure, the objective function also includes: a historical trajectory tracking cost function, which is used to constrain the position difference between the expected position at each moment and the historical expected position at each moment and the angle difference between the expected orientation angle at each moment and the historical expected orientation angle at each moment.

[0196] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, an autonomous driving vehicle, a readable storage medium, and a computer program product.

[0197] According to an embodiment of the present disclosure, an electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method as described above.

[0198] According to an embodiment of the present disclosure, an autonomous driving vehicle includes the above-mentioned electronic device.

[0199] According to an embodiment of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause a computer to execute the method as described above.

[0200] According to an embodiment of the present disclosure, a computer program product includes a computer program, and when the computer program is executed by a processor, the computer program implements the method as described above.

[0201] Fig.12 A schematic block diagram of an example electronic device 1200 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0202] like Fig.12 As shown, the device 1200 includes a computing unit 1201, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1202 or a computer program loaded from a storage unit 1208 into a random access memory (RAM) 1203. In the RAM 1203, various programs and data required for the operation of the device 1200 can also be stored. The computing unit 1201, the ROM 1202, and the RAM 1203 are connected to each other via a bus 1204. An input / output (I / O) interface 1205 is also connected to the bus 1204.

[0203] A number of components in the device 1200 are connected to the I / O interface 1205, including: an input unit 1206, such as a keyboard, a mouse, etc.; an output unit 1207, such as various types of displays, speakers, etc.; a storage unit 1208, such as a disk, an optical disk, etc.; and a communication unit 1209, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1209 allows the device 1200 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0204] The computing unit 1201 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1201 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1201 performs the various methods and processes described above, such as a vehicle control method. For example, in some embodiments, the vehicle control method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 1208. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 1200 via the ROM 1202 and / or the communication unit 1209. When the computer program is loaded into the RAM 1203 and executed by the computing unit 1201, one or more steps of the vehicle control method described above may be performed. Alternatively, in other embodiments, the computing unit 1201 may be configured to perform the vehicle control method in any other appropriate manner (e.g., by means of firmware).

[0205] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0206] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0207] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0208] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0209] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0210] A computer system may include a client and a server. The client and the server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises through computer programs running on respective computers and having a client-server relationship to each other. The server may be a cloud server, a server in a distributed system, or a server combined with a blockchain.

[0211] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0212] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A vehicle control method, comprising: Based on the deviation between the expected driving state and the actual driving state of the target vehicle at the current moment, a first initial control sequence is generated by correcting the historical control sequence generated at the historical moment; wherein the expected driving state at the current moment is determined according to the historical control sequence; and Under the constraints of dynamics and driving cost, the first initial control sequence is iteratively processed to obtain a target control sequence, and the target control sequence is used to control the driving of the target vehicle.

2. The method according to claim 1, wherein: The historical control sequence includes control information at T moments, where T is an integer greater than 1; The method generates a first initial control sequence by correcting a historical control sequence generated at a historical moment based on a deviation between an expected driving state and an actual driving state of the target vehicle at a current moment, including: According to the control information at time t-1, determine the expected driving state at time t, where t=2,3,…,T-1; Based on the deviation between the expected driving state at the tth moment and the actual driving state at the tth moment, interpolation is performed between the control information of each adjacent moment in the historical control sequence to generate the first initial control sequence.

3. The method according to claim 2, wherein: The step of interpolating control information at adjacent moments in the historical control sequence based on the deviation between the expected driving state at the t-th moment and the actual driving state at the t-th moment to generate the first initial control sequence includes: Determine the expected driving state at time t+1 according to the control information at time t; An interpolation coefficient is obtained according to a ratio of the deviation to an expected driving state difference, wherein the expected driving state difference indicates a difference between the expected driving state at the t-th moment and the expected driving state at the t+1-th moment; and Based on the interpolation coefficients and the control information at each moment, interpolation is performed between the control information at each adjacent moment in the historical control sequence to obtain the first initial control sequence.

4. The method according to any one of claims 1 to 3, wherein: The method further comprises: generating, according to a reference lane line associated with the driving intention of the target vehicle, a second initial control sequence for controlling the target vehicle to travel along the reference lane line; Based on the objective function, an initial control sequence for performing an iterative operation is determined from the first initial control sequence and the second initial control sequence.

5. The method according to claim 4, wherein: The generating, according to the reference lane line associated with the driving intention of the target vehicle, a second initial control sequence for controlling the target vehicle to travel along the reference lane line comprises: Extracting each reference position at each reference time from the reference lane line; and The second initial control sequence is generated according to the curvature change state of each reference position on the reference lane line.

6. The method according to claim 4, wherein: The step of generating a second initial control sequence for controlling the target vehicle to travel along the reference lane line according to the reference lane line associated with the driving intention of the target vehicle comprises: Based on the driving state of the target vehicle at the current moment, generating the driving state at each moment; Determining a target reference position at each moment from the reference lane line based on the driving state at each moment; and The driving state at each moment and the target reference position at each moment are processed to generate the second initial control sequence.

7. The method according to claim 6, wherein: The driving state includes: a driving position and a wheel steering angle of the target vehicle; The processing of the driving state at each moment and the target reference position at each moment to generate the second initial control sequence includes: The angle of the target reference position at the tth moment relative to the driving position at the tth moment is determined as the heading angle of the target vehicle at the tth moment, t=2, 3, ..., T-1; T is an integer greater than 1; Based on the geometric relationship between the wheel steering angle and the vehicle heading angle, geometrically transform the heading angle at the tth moment to generate the wheel steering angle at the tth moment; generating a curvature at the tth moment based on the wheel steering angle at the tth moment and a predetermined wheelbase; and The second initial control sequence is generated based on the curvature change states at the tth time and the t-1th time.

8. The method according to claim 4, wherein: The determining, based on the objective function, an initial control sequence for performing an iterative operation from the first initial control sequence and the second initial control sequence comprises: Processing the first initial control sequence based on the objective function to generate a first generation value; Processing the second initial control sequence based on the objective function to generate a second generation value; In response to determining that the first generation value is less than the second generation value, determining the first initial control sequence as the initial control sequence; In response to determining that the first generation value is greater than or equal to the second generation value, the second initial control sequence is determined to be the initial control sequence.

9. The method according to claim 4, wherein: The expected driving state includes an expected position and an expected orientation angle; the reference driving state includes a reference position and a reference orientation angle; The objective function includes: A reference line tracking cost function, used to constrain the position difference between the expected position at each moment and the reference position at each moment and the angle difference between the expected orientation angle at each moment and the reference orientation angle at each moment; A target state cost function is used to constrain the error between the expected driving state at the target time and the target driving state; Driving comfort cost function, used to constrain the curvature change of the expected driving state at each moment; The safety cost function is used to constrain the safe driving range of the expected driving state at each moment.

10. According to the method of claim 8 or 9, the objective function also includes a historical trajectory tracking cost function, which is used to constrain the position difference between the expected position at each moment and the historical expected position at each moment and the angle difference between the expected orientation angle at each moment and the historical expected orientation angle at each moment.

11. A vehicle control device, comprising: a correction module, configured to generate a first initial control sequence by correcting a historical control sequence generated at a historical moment based on a deviation between an expected driving state of the target vehicle at a current moment and an actual driving state; wherein the expected driving state at the current moment is determined according to the historical control sequence; and The optimization module is used to iteratively process the first initial control sequence under the dynamic constraint and the driving cost constraint to obtain a target control sequence, and the target control sequence is used to control the driving of the target vehicle.

12. An electronic device 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, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 10.

13. An autonomous driving vehicle comprising: The electronic device as claimed in claim 12.

14. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-10.

15. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 10.

Images