Automatic driving control method, device and electronic equipment for vehicle

Abstract

The present disclosure provides a method, device and electronic device for controlling an automatic driving of a vehicle, which relates to the field of artificial intelligence technology, and in particular to the field of automatic driving, deep learning and intelligent transportation technology. The specific implementation scheme is as follows: determining the current vehicle state information of the vehicle, and the model parameter sequence corresponding to the vehicle state information, wherein the model parameter sequence includes: the model parameter information of the model predictive controller at each prediction time point in the prediction time domain, wherein the model parameter information at different prediction time points is different; using the model predictive controller, according to the model parameter sequence and the expected trajectory of the vehicle, determining the control quantity sequence; according to the control quantity sequence, performing automatic driving control processing on the vehicle, so that the model parameter sequence can be determined based on the current vehicle state information of the vehicle; and using different model parameter information for different prediction time points to accelerate the convergence speed of the model predictive controller and improve the automatic driving control efficiency of the vehicle.

Description

Technical Field

[0001] The present disclosure relates to the fields of artificial intelligence technology, in particular to autonomous driving, deep learning, and intelligent transportation technology, and in particular to a method, device, and electronic device for autonomous driving control of a vehicle. Background Art

[0002] At present, in the relevant technology, in the lateral and longitudinal control of autonomous driving, model parameter information is determined based on the vehicle's speed, curvature, and acceleration information; and then a model predictive controller is used to control the vehicle's autonomous driving according to the model parameter information.

[0003] In the above scheme, the model parameter information is fixed and the convergence speed of the model predictive controller is slow, thereby reducing the vehicle's autonomous driving control efficiency. Summary of the Invention

[0004] The present disclosure provides a method, device, and electronic device for controlling an automatic driving of a vehicle.

[0005] According to one aspect of the present disclosure, a method for automatic driving control of a vehicle is provided, comprising: determining current vehicle state information of the vehicle, and a model parameter sequence corresponding to the vehicle state information, wherein the model parameter sequence comprises: model parameter information of a model predictive controller at each prediction time point within a prediction time domain, wherein the model parameter information at different prediction time points is different; utilizing the model predictive controller to determine a control quantity sequence according to the model parameter sequence and the expected trajectory of the vehicle, wherein the control quantity sequence comprises: a lateral control quantity and a longitudinal control quantity at each of the prediction time points; and performing automatic driving control processing on the vehicle according to the lateral control quantity and the longitudinal control quantity at the first prediction time point within the prediction time domain.

[0006] According to another aspect of the present disclosure, an automatic driving control device for a vehicle is provided, comprising: a first determination module for determining the current vehicle state information of the vehicle, and a model parameter sequence corresponding to the vehicle state information, wherein the model parameter sequence comprises: model parameter information of a model prediction controller at each prediction time point within a prediction time domain, wherein the model parameter information at different prediction time points is different; a second determination module for determining a control quantity sequence using the model prediction controller according to the model parameter sequence and the expected trajectory of the vehicle, wherein the control quantity sequence comprises: a lateral control quantity and a longitudinal control quantity at each of the prediction time points; and a control module for performing automatic driving control processing on the vehicle according to the lateral control quantity and the longitudinal control quantity at the first prediction time point within the prediction time domain.

[0007] According to another aspect of the present disclosure, there is provided an electronic device, comprising:

[0008] at least one processor; and

[0009] a memory communicatively connected to the at least one processor; wherein,

[0010] 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 so that the at least one processor can execute the automatic driving control method of the vehicle proposed above in the present disclosure.

[0011] 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 automatic driving control method for the vehicle proposed above in the present disclosure.

[0012] According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the steps of the automatic driving control method for a vehicle proposed above in the present disclosure.

[0013] 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

[0014] The accompanying drawings are provided to facilitate a better understanding of the present invention and do not constitute a limitation of the present disclosure.

[0015] Figure 1 is a schematic diagram according to a first embodiment of the present disclosure;

[0016] Figure 2 is a schematic diagram according to a second embodiment of the present disclosure;

[0017] Figure 3 is a schematic diagram according to a third embodiment of the present disclosure;

[0018] Figure 4 is a schematic diagram according to a fourth embodiment of the present disclosure;

[0019] Figure 5 It is a block diagram of an electronic device used to implement the automatic driving control method of a vehicle according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0020] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize 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.

[0021] At present, in the relevant technology, in the lateral and longitudinal control of autonomous driving, model parameter information is determined based on the vehicle's speed, curvature, and acceleration information; and then a model predictive controller is used to control the vehicle's autonomous driving according to the model parameter information.

[0022] In the above scheme, the model parameter information is fixed and the convergence speed of the model predictive controller is slow, thereby reducing the vehicle's autonomous driving control efficiency.

[0023] In response to the above problems, the present disclosure proposes a method, device and electronic device for controlling automatic driving of a vehicle.

[0024] Figure 1 It is a schematic diagram according to the first embodiment of the present disclosure. It should be noted that the automatic driving control method of a vehicle in the embodiment of the present disclosure can be applied to an automatic driving control device of a vehicle, and the device can be configured in an electronic device so that the electronic device can perform the automatic driving control function of the vehicle.

[0025] Among them, the electronic device can be any device with computing capabilities, such as a personal computer (PC), a mobile terminal, a server, etc. The mobile terminal can be, for example, a vehicle-mounted device, a mobile phone, a tablet computer, a personal digital assistant, a wearable device, and other hardware devices with various operating systems, touch screens and / or display screens.

[0026] like Figure 1 As shown, the automatic driving control method of the vehicle may include the following steps:

[0027] Step 101, determine the current vehicle state information of the vehicle and the model parameter sequence corresponding to the vehicle state information, wherein the model parameter sequence includes: model parameter information of the model prediction controller at each prediction time point in the prediction time domain, wherein the model parameter information at different prediction time points is different.

[0028] In an embodiment of the present disclosure, the vehicle's current vehicle state information may include at least one of the following vehicle state parameters: lateral vehicle state parameter, longitudinal vehicle state parameter, expected lateral vehicle state parameter, expected longitudinal vehicle state parameter, lateral vehicle state parameter error, and longitudinal vehicle state parameter error.

[0029] The lateral vehicle state parameter may include at least one of the following parameters: lateral position, lateral heading angle, lateral displacement, lateral velocity, lateral acceleration, lateral curvature, yaw rate, and front wheel angle. The longitudinal vehicle state parameter may include at least one of the following parameters: longitudinal position, longitudinal displacement, longitudinal velocity, longitudinal acceleration, and longitudinal torque. The lateral vehicle state parameter error is the error between the lateral vehicle state parameter and the desired lateral vehicle state parameter. The longitudinal vehicle state parameter error is the error between the longitudinal vehicle state parameter and the desired longitudinal vehicle state parameter.

[0030] The vehicle is provided with a pre-planned desired trajectory, and the vehicle's automatic driving control device is used to combine the desired trajectory with the vehicle's current vehicle state information to perform automatic driving control of the vehicle. The desired trajectory includes multiple desired trajectory points, each of which is provided with corresponding desired lateral vehicle state parameters and desired longitudinal vehicle state parameters.

[0031] Among them, the method for determining the expected lateral vehicle state parameters and the expected longitudinal vehicle state parameters in the vehicle's current vehicle state information is, for example, to select a matching expected trajectory point from the expected trajectory based on the vehicle's current lateral vehicle state parameters and longitudinal vehicle state parameters, and use the expected lateral vehicle state parameters of the matching expected trajectory point as the expected lateral vehicle state parameters in the vehicle's current vehicle state information; and use the expected longitudinal vehicle state parameters of the matching expected trajectory point as the expected longitudinal vehicle state parameters in the vehicle's current vehicle state information.

[0032] Among them, the determination of multiple vehicle state parameters is used to determine the model parameter sequence, which can realize the use of different model parameter sequences for different vehicle state information, and realize the refined distinction of different vehicle state information, thereby improving the accuracy of the determined model parameter sequence, so that the determined model parameter sequence can be better applied to the current vehicle state information, thereby further improving the vehicle's automatic driving control efficiency.

[0033] Among them, the Model Predictive Control (MPC) is used to predict the lateral control quantity and longitudinal control quantity at each prediction time point in the prediction time domain based on the vehicle's current vehicle state information, expected trajectory, and model parameter sequence, so as to perform autonomous driving control processing on the vehicle based on the lateral control quantity and longitudinal control quantity at the first prediction time point in the prediction time domain. Among them, the model predictive controller adopts an advanced process control strategy for prediction processing. For example, a control quantity sequence is first randomly determined, and a predicted trajectory is determined based on the control quantity sequence; based on the predicted trajectory, the expected trajectory, and constraint information, the control quantity sequence is adjusted once; and the above process is repeated until the difference between the predicted trajectory and the expected trajectory meets the specified conditions.

[0034] In the disclosed embodiment, the model parameter information at the prediction time point includes: transverse model parameter information and longitudinal model parameter information. The transverse model parameter information includes: a transverse error term penalty weight, a transverse control amount penalty weight, and a transverse control amount increment penalty weight. The longitudinal model parameter information includes: a longitudinal error term penalty weight, a longitudinal control amount penalty weight, and a longitudinal control amount increment penalty weight.

[0035] The lateral error term penalty weight represents the weight of the lateral vehicle state parameter error term. The longitudinal error term penalty weight represents the weight of the longitudinal vehicle state parameter error term. The different model parameter information at different prediction time points may refer to different lateral error term penalty weights at different prediction time points and / or different longitudinal error term penalty weights at different prediction time points.

[0036] Among them, the model parameter information can include multiple penalty weights, which is convenient for setting the attenuation parameters and then selecting the penalty weights for attenuation processing. Due to the control of the vehicle, the lateral error term and the longitudinal error term at the early prediction time point in the prediction time domain are generally more important, and the lateral error term and the longitudinal error term at the later prediction time point are not important. Therefore, in order to shorten the convergence speed of the model predictive controller, the weights of the lateral error term and the longitudinal error term at the early prediction time point need to be larger, and the weights of the lateral error term and the longitudinal error term at the later prediction time point need to be smaller. That is, the penalty weight of the lateral error term at each prediction time point can show a decaying trend, and / or, the penalty weight of the longitudinal error term at each prediction time point can show a decaying trend.

[0037] Step 102 : Using a model predictive controller, a control variable sequence is determined according to a model parameter sequence and a desired trajectory of the vehicle, wherein the control variable sequence includes: a lateral control variable and a longitudinal control variable at each predicted time point.

[0038] In the disclosed embodiment, the lateral control variable may be, for example, the front wheel steering angle, and the longitudinal control variable may be, for example, the torque.

[0039] Among them, the model predictive controller is used to determine the lateral control amount and longitudinal control amount at each prediction time point in the prediction time domain at each sampling moment based on the vehicle's current vehicle state information and the model parameter sequence, and then determine the predicted trajectory; construct an objective function based on the error between the predicted trajectory and the expected trajectory; and combine the objective function and constraint information to solve and determine the control amount sequence and output it.

[0040] Among them, (1) the model predictive controller can first determine an initial control quantity sequence; (2) the lateral model in the model predictive controller can determine the predicted lateral vehicle state parameters at the first prediction time point based on the vehicle's current lateral vehicle state parameters and the lateral control quantity at the first prediction time point; the longitudinal model in the model predictive controller can determine the predicted longitudinal vehicle state parameters at the first prediction time point based on the vehicle's current longitudinal vehicle state parameters and the lateral control quantity at the first prediction time point; (3) repeating step (2) can obtain the predicted lateral vehicle state parameters and the predicted longitudinal vehicle state parameters at each prediction time point, and then obtain the predicted trajectory.

[0041] Step 103 , performing automatic driving control processing on the vehicle based on the lateral control amount and the longitudinal control amount at the first prediction time point in the prediction time domain.

[0042] In the embodiment of the present disclosure, each predicted time point within the prediction time domain can be sorted in chronological order, where the first predicted time point is the first predicted time point in the sorting result. For example, the sorting result is t1, t2, t3, t4, t5, where t1 is the first predicted time point.

[0043] In the embodiment disclosed herein, taking the lateral control amount as the downward-delivered front wheel angle and the longitudinal control amount as the downward-delivered torque as an example, the vehicle controls the vehicle engine, etc. according to the downward-delivered front wheel angle and the downward-delivered torque, thereby realizing automatic driving control processing.

[0044] The automatic driving control method for a vehicle in an embodiment of the present disclosure determines the current vehicle state information of the vehicle and a model parameter sequence corresponding to the vehicle state information, wherein the model parameter sequence includes: model parameter information of a model predictive controller at each prediction time point in a prediction time domain, wherein the model parameter information at different prediction time points is different; utilizing the model predictive controller, according to the model parameter sequence and the expected trajectory of the vehicle, a control quantity sequence is determined, wherein the control quantity sequence includes: lateral control quantity and longitudinal control quantity at each prediction time point; performing automatic driving control processing on the vehicle according to the lateral control quantity and longitudinal control quantity at the first prediction time point in the prediction time domain, thereby being able to determine the model parameter sequence based on the current vehicle state information of the vehicle; and using different model parameter information for different prediction time points to accelerate the convergence speed of the model predictive controller, thereby improving the determination speed of the control quantity sequence and improving the automatic driving control efficiency of the vehicle.

[0045] In order to accurately determine the model parameter sequence corresponding to the vehicle state information, such as Figure 2 As shown, Figure 2 It is a schematic diagram according to the second embodiment of the present disclosure. In the embodiment of the present disclosure, the model parameter sequence is determined according to the basic model parameter information and the attenuation parameter information corresponding to the vehicle state information. Figure 2 The illustrated embodiment may include the following steps:

[0046] Step 201: Determine the current vehicle status information of the vehicle.

[0047] In an embodiment of the present disclosure, the vehicle's current vehicle state information may include at least one of the following vehicle state parameters: lateral vehicle state parameter, longitudinal vehicle state parameter, expected lateral vehicle state parameter, expected longitudinal vehicle state parameter, lateral vehicle state parameter error, and longitudinal vehicle state parameter error.

[0048] Step 202: Determine basic model parameter information and attenuation parameter information corresponding to the vehicle state information.

[0049] The basic model parameter information is the model parameter information at the first prediction time point in the prediction time domain. The model parameter information at other prediction time points in the prediction time domain can be determined based on the basic model parameter information and the attenuation parameter information.

[0050] In the embodiment of the present disclosure, the process of the vehicle's automatic driving control device executing step 202 can, for example, be to input the vehicle state information into a preset neural network model, obtain the scaling information output by the neural network model; and determine the basic model parameter information and attenuation parameter information corresponding to the vehicle state information based on the scaling information, reference model parameter information, and reference attenuation parameter information.

[0051] The basic model parameter information corresponding to the vehicle state information can be determined based on the scaling information and the reference model parameter information. The reference model parameter information can be the model parameter information of an existing vehicle model, or the model parameter information of a newly added vehicle model that has been manually or automatically adjusted. The reference model parameter information can be set according to actual needs and is not specifically limited here.

[0052] Where the reference model parameter information includes multidimensional control parameters, the scaling ratio information may also be multidimensional, with the number of dimensions being consistent with the number of dimensions of the control parameters in the reference model parameter information. The multidimensional scaling ratios in the scaling ratio information correspond one-to-one with the multidimensional control parameters in the reference model parameter information, respectively indicating the scaling ratios of the corresponding control parameters.

[0053] The neural network model has high accuracy and can consider a wider range of vehicle state parameters, enabling refined differentiation of different vehicle state parameters and improving the accuracy of the resulting model parameter sequence. Furthermore, the use of scaling information, combined with constraints on scaling information, can avoid sudden changes in model parameter information, further improving the accuracy of the resulting basic model parameter information and, consequently, the accuracy of the resulting model parameter sequence.

[0054] In the example where the scaling ratio information includes multi-dimensional scaling ratios, the constraints on the scaling ratio information refer to the constraints on the scaling ratio of each dimension. In the example of the scaling ratio of the first dimension, the constraints refer to the restricted range of the scaling ratio of the first dimension, i.e., the scaling ratio of the first dimension must fall within the restricted range. For example, the restricted range may include the scaling ratio of the first dimension being greater than the first scaling ratio and less than the second scaling ratio.

[0055] The training process of the neural network model may be, for example, as follows: (1) determining an initial neural network model, including an initial policy network, the input of which is vehicle state information, and the output of which is scaling information; (2) the initial policy network provides the scaling information to the control module; (3) the control module determines the model parameter sequence based on the scaling information and the reference model parameter information, and then determines the control quantity sequence in combination with the vehicle state information; (4) the dynamic simulation environment determines the predicted trajectory based on the vehicle state information and the control quantity sequence; combining the predicted trajectory with the expected trajectory and the reward function, a training sample is generated and placed in a memory replay pool; wherein the training sample includes: vehicle state parameters at the current moment, control quantity, reward value, and vehicle state parameters at the next moment; (5) selecting a training sample from the memory replay pool, and training the initial neural network model in combination with the reinforcement learning algorithm and the value network; wherein the input of the value network is scaling information, and the output is a value function for evaluating the scaling information; (6) repeating the above five steps until the trained neural network model meets the specified training convergence conditions.

[0056] The reward function formula can be shown as the following formula (1):

[0057] r(t)=a1r1(t)+a2r2(t)+a3r3(t)+a4r4(t) (1)

[0058] Here, r(t) represents the reward function value at prediction time point t; r1(t) represents the error reward; r2(t) represents the error change rate reward; r3(t) represents the control variable change reward; and r4(t) represents the simulation metric reward. The control variable change reward is -Δu, where Δu represents the control variable change.

[0059] The error return formula and the error change rate return formula can be shown as the following formulas (2) and (3):

[0060]

[0061]

[0062] Wherein, e(t) represents the error at the prediction time point t, which may include the lateral state parameter error and the longitudinal state parameter error.

[0063] The formula for simulation metric return can be shown as the following formula (4):

[0064] r4(t)=∑r 4-i (t) (4)

[0065] In one example, the collision report r 4-1(t) = -500; sudden braking report r 4-2 (t) = -20; turn the direction sharply and return r 4-3 (t) = -20; the reward r of trajectory replanning 4-4 (t)=-200.

[0066] It should be noted that the values of the above four returns are only examples and can be adjusted according to actual needs. No specific restrictions are made here.

[0067] Step 203: Determine the model parameter information at each prediction time point in the prediction time domain according to the basic model parameter information and the attenuation parameter information.

[0068] In the embodiment of the present disclosure, the process of the vehicle's automatic driving control device executing step 203 may, for example, be to determine the serial number of each prediction time point in the prediction time domain; for each prediction time point, determine the attenuation ratio information at the prediction time point based on the attenuation parameter information and the serial number; and determine the model parameter information at the prediction time point based on the basic model parameter information and the attenuation ratio information at the prediction time point.

[0069] Among them, the process of the vehicle's automatic driving control device determining the attenuation ratio information at the predicted time point based on the attenuation parameter information and the serial number can be, for example, taking the attenuation parameter information as d as an example, assuming that the serial number of a certain predicted time point is z, then the attenuation ratio information at the predicted time point is the result obtained by multiplying the attenuation parameter information by z ds, and then obtaining the model parameter information at the predicted time point.

[0070] Among them, taking the attenuation parameter information as the first attenuation parameter for the lateral error term penalty weight as an example, the calculation formula of the lateral error term penalty weight in the model parameter information at the prediction time point can be shown as the following formula (5):

[0071] Q i =Q*d i-1 (5)

[0072] Among them, Q represents the penalty weight of the horizontal error term in the basic model parameter information; Q i Represents the penalty weight of the horizontal error term at the i-1th prediction time point; i-1 represents the serial number of the prediction time point.

[0073] The basic model parameter information may include: a lateral error term penalty weight, a lateral control amount penalty weight, a lateral control amount increment penalty weight, a longitudinal error term penalty weight, a longitudinal control amount penalty weight, and a longitudinal control amount increment penalty weight. The attenuation parameter information may include: a first attenuation parameter for the lateral error term penalty weight and a second attenuation parameter for the longitudinal error term penalty weight.

[0074] Among them, the attenuation ratio information at the prediction time point is determined according to the attenuation parameter information and the serial number, which can ensure that the lateral error term penalty weight and the longitudinal error term penalty weight in the model parameter information at each prediction time point show an attenuation trend, so that the weights of the lateral error term and the longitudinal error term at the early prediction time point need to be larger, and the weights of the lateral error term and the longitudinal error term at the later prediction time point need to be smaller, thereby shortening the convergence speed of the model predictive controller, being able to determine the control quantity sequence in time, and performing automatic driving control processing on the vehicle in time.

[0075] Step 204 : Determine the model parameter sequence corresponding to the vehicle state information based on the model parameter information at each prediction time point within the prediction time domain.

[0076] Step 205 : Using a model predictive controller, a control variable sequence is determined according to the model parameter sequence and the expected trajectory of the vehicle, wherein the control variable sequence includes: a lateral control variable and a longitudinal control variable at each predicted time point.

[0077] Step 206 , performing automatic driving control processing on the vehicle based on the lateral control amount and the longitudinal control amount at the first prediction time point in the prediction time domain.

[0078] It should be noted that the details of step 205 and step 206 can be found in Figure 1 Step 102 and step 103 in the illustrated embodiment will not be described in detail here.

[0079] The automatic driving control method for a vehicle of the disclosed embodiment determines the current vehicle state information of the vehicle; determines the basic model parameter information and attenuation parameter information corresponding to the vehicle state information; determines the model parameter information at each prediction time point in the prediction time domain based on the basic model parameter information and the attenuation parameter information; determines the model parameter sequence corresponding to the vehicle state information based on the model parameter information at each of the prediction time points in the prediction time domain; utilizes a model predictive controller to determine a control quantity sequence based on the model parameter sequence and the expected trajectory of the vehicle, wherein the control quantity sequence includes: lateral control quantities and longitudinal control quantities at each prediction time point; performs automatic driving control processing on the vehicle based on the lateral control quantity and longitudinal control quantity at the first prediction time point in the prediction time domain, thereby being able to determine the model parameter sequence based on the current vehicle state information of the vehicle; and adopts different model parameter information for different prediction time points to accelerate the convergence speed of the model predictive controller, thereby improving the determination speed of the control quantity sequence and improving the automatic driving control efficiency of the vehicle.

[0080] In order to accurately determine the control quantity sequence, such as Figure 3 As shown, Figure 3It is a schematic diagram according to the third embodiment of the present disclosure. In the embodiment of the present disclosure, the lateral model and the longitudinal model in the model predictive controller are used in combination with the model parameter sequence to respectively determine the lateral control amount and the longitudinal control amount at each prediction time point, and then determine the control amount sequence. Figure 3 The illustrated embodiment may include the following steps:

[0081] Step 301, determine the current vehicle state information of the vehicle and the model parameter sequence corresponding to the vehicle state information, wherein the model parameter sequence includes: model parameter information of the model prediction controller at each prediction time point in the prediction time domain, wherein the model parameter information at different prediction time points is different; the model parameter information includes: lateral model parameter information and longitudinal model parameter information.

[0082] Step 302 : Using the lateral model in the model predictive controller, according to the lateral model parameter information at each prediction time point in the model parameter sequence and the expected trajectory, determine the lateral control amount at each prediction time point.

[0083] In the embodiment of the present disclosure, the process of the vehicle's automatic driving control device executing step 302 may, for example, be to determine the lateral vehicle state parameters in the vehicle state information, and the initial lateral control amount at each prediction time point; determine the predicted lateral vehicle state parameters at each prediction time point based on the lateral vehicle state parameters and the initial lateral control amount at each prediction time point using the lateral model; construct a first objective function based on the predicted lateral vehicle state parameters at each prediction time point, the expected lateral vehicle state parameters at each prediction time point in the expected trajectory, and the lateral model parameter information; and adjust the initial lateral control amount at each prediction time point based on the value of the first objective function to obtain the lateral control amount at each prediction time point.

[0084] The formula of the horizontal model can be shown as the following formula (6):

[0085]

[0086] Among them, y e Indicates the lateral displacement in the vehicle status information; Represents the lateral velocity in the vehicle status information; θ e Indicates the lateral heading angle in the vehicle status information; represents the yaw rate in the vehicle state information; δ represents the front wheel turning angle in the vehicle state information; δ desrepresents the lateral control variable, i.e., the front wheel steering angle. Accordingly, the lateral error penalty weight can have five dimensions, where each dimension represents the error penalty weight for lateral displacement, lateral velocity, lateral heading angle, yaw rate, and front wheel angle, respectively.

[0087] The above formula (6) is integrated over time, which can obtain the lateral displacement, lateral velocity, lateral heading angle, yaw rate and front wheel angle at the next predicted time point.

[0088] It should be noted that Formula (6) is illustrated using the example of a lateral model input in which the vehicle state parameters are lateral displacement, lateral velocity, lateral heading angle, yaw rate, and front wheel angle. The vehicle state parameters in the lateral model input can also be replaced with other lateral vehicle state parameters, which are not limited here and can be set according to actual needs.

[0089] The formula of the first objective function of the horizontal model can be shown as the following formula (7):

[0090]

[0091] Where J represents the first objective function; Y i Y represents the predicted lateral vehicle state parameter at the i-th prediction time point obtained based on the lateral vehicle state parameter at the i-1th prediction time point and the lateral control amount at the i-th prediction time point; ri represents the expected lateral vehicle state parameter at the i-th prediction time point; Q represents the lateral error term penalty weight, including the lateral error term penalty weights at each prediction time point; U i represents the lateral control amount at the i-th prediction time point; ΔU i represents the lateral control amount increment at the i-th prediction time point; R1 represents the lateral control amount penalty weight; R2 represents the lateral control amount increment penalty weight.

[0092] Among them, the lateral model is used to determine the predicted lateral vehicle state parameters at each prediction time point based on the lateral vehicle state parameters and the initial lateral control amount at each prediction time point, and then the first objective function is constructed in combination with the lateral model parameter information, and the initial lateral control amount at each prediction time point is adjusted to obtain the lateral control amount at each prediction time point. This can accelerate the convergence speed of the model predictive controller and improve the accuracy of the determined lateral control amount.

[0093] Step 303 : Using the longitudinal model in the model predictive controller, according to the longitudinal model parameter information at each prediction time point in the model parameter sequence and the expected trajectory, the longitudinal control amount at each prediction time point is determined.

[0094] In the embodiment of the present disclosure, the process of the vehicle's automatic driving control device executing step 303 may, for example, be to determine the longitudinal vehicle state parameters in the vehicle state information, and the initial longitudinal control amount at each prediction time point; determine the predicted longitudinal vehicle state parameters at each prediction time point based on the longitudinal vehicle state parameters and the initial longitudinal control amount at each prediction time point using the longitudinal model; construct a second objective function based on the predicted longitudinal vehicle state parameters at each prediction time point, the expected longitudinal vehicle state parameters at each prediction time point in the expected trajectory, and the longitudinal model parameter information; and adjust the initial longitudinal control amount at each prediction time point based on the value of the second objective function to obtain the longitudinal control amount at each prediction time point.

[0095] The formula of the longitudinal model can be shown as the following formula (8):

[0096]

[0097] Where x represents the longitudinal displacement in the vehicle state information; v represents the longitudinal velocity in the vehicle state information; T represents the longitudinal torque in the vehicle state information; T des represents the longitudinal control variable, i.e., the torque. Correspondingly, the longitudinal error penalty weight can have three dimensions, where each dimension represents the error penalty weight of the longitudinal displacement, the error penalty weight of the longitudinal velocity, and the error penalty weight of the longitudinal torque, respectively.

[0098] The above formula (8) is integrated over time, which can obtain the longitudinal displacement, longitudinal velocity and longitudinal torque at the next predicted time point.

[0099] It should be noted that Formula (8) is illustrated using the longitudinal model input vehicle state parameters as the longitudinal displacement, longitudinal velocity, and longitudinal torque. The longitudinal model input vehicle state parameters can also be replaced with other longitudinal vehicle state parameters, which are not limited here and can be set according to actual needs.

[0100] Among them, the formula of the second objective function of the longitudinal model can be determined by referring to the formula of the first objective function. The lateral vehicle state parameters, lateral control quantity, expected lateral vehicle state parameters, lateral error term penalty weight, lateral control quantity, lateral control quantity increment, lateral control quantity penalty weight, and lateral control quantity increment penalty weight in the formula of the first objective function can be replaced by the longitudinal vehicle state parameters, longitudinal control quantity, expected longitudinal vehicle state parameters, longitudinal error term penalty weight, longitudinal control quantity, longitudinal control quantity increment, longitudinal control quantity penalty weight, and longitudinal control quantity increment penalty weight to obtain the second objective function.

[0101] Among them, the longitudinal model is used to determine the predicted longitudinal vehicle state parameters at each prediction time point based on the longitudinal vehicle state parameters and the initial longitudinal control amount at each prediction time point, and then the second objective function is constructed in combination with the longitudinal model parameter information, and the initial longitudinal control amount at each prediction time point is adjusted to obtain the longitudinal control amount at each prediction time point. This can accelerate the convergence speed of the model predictive controller and improve the accuracy of the determined longitudinal control amount.

[0102] Step 304 : determining a control quantity sequence according to the lateral control quantity and the longitudinal control quantity at each predicted time point, wherein the control quantity sequence includes: the lateral control quantity and the longitudinal control quantity at each predicted time point.

[0103] Step 305 , performing automatic driving control processing on the vehicle based on the lateral control amount and the longitudinal control amount at the first prediction time point in the prediction time domain.

[0104] It should be noted that the details of step 301 and step 305 can be found in Figure 1 Step 101 and step 103 in the illustrated embodiment will not be described in detail here.

[0105] The automatic driving control method for a vehicle according to an embodiment of the present disclosure determines the current vehicle state information of the vehicle and a model parameter sequence corresponding to the vehicle state information, wherein the model parameter sequence includes: model parameter information of a model predictive controller at each prediction time point in a prediction time domain, wherein the model parameter information at different prediction time points is different; utilizing a lateral model in the model predictive controller to determine a lateral control amount at each prediction time point based on the lateral model parameter information at each prediction time point in the model parameter sequence and a desired trajectory; utilizing a longitudinal model in the model predictive controller to determine a longitudinal control amount at each prediction time point based on the longitudinal model parameter information at each prediction time point in the model parameter sequence and the desired trajectory; determining a control amount sequence based on the lateral control amount and the longitudinal control amount at each prediction time point; performing automatic driving control processing on the vehicle based on the lateral control amount and the longitudinal control amount at the first prediction time point in the prediction time domain, thereby determining the model parameter sequence based on the current vehicle state information of the vehicle; and utilizing different model parameter information for different prediction time points to accelerate the convergence speed of the model predictive controller, thereby increasing the speed of determining the control amount sequence and improving the automatic driving control efficiency of the vehicle.

[0106] In order to implement the above embodiments, the present disclosure also proposes an automatic driving control device for a vehicle.

[0107] like Figure 4 As shown, Figure 44 is a schematic diagram of a fourth embodiment of the present disclosure. The automatic driving control device 400 for a vehicle includes: a first determination module 410 , a second determination module 420 , and a control module 430 .

[0108] A first determining module 410 is configured to determine current vehicle state information of the vehicle and a model parameter sequence corresponding to the vehicle state information, wherein the model parameter sequence includes model parameter information of a model predictive controller at each prediction time point within a prediction time domain, wherein the model parameter information at different prediction time points is different;

[0109] a second determining module 420 for determining a control variable sequence using the model predictive controller according to the model parameter sequence and the desired trajectory of the vehicle, wherein the control variable sequence includes: a lateral control variable and a longitudinal control variable at each of the predicted time points;

[0110] The control module 430 is used to perform automatic driving control processing on the vehicle based on the lateral control amount and the longitudinal control amount at the first prediction time point in the prediction time domain.

[0111] As a possible implementation of the embodiment of the present disclosure, the first determining module 410 includes: a first determining unit, a second determining unit, a third determining unit, and a fourth determining unit; wherein the first determining unit is configured to determine current vehicle state information of the vehicle; and the second determining unit is configured to determine basic model parameter information and attenuation parameter information corresponding to the vehicle state information.

[0112] The third determination unit is used to determine the model parameter information at each prediction time point in the prediction time domain based on the basic model parameter information and the attenuation parameter information; the fourth determination unit is used to determine the model parameter sequence corresponding to the vehicle state information based on the model parameter information at each prediction time point in the prediction time domain.

[0113] As a possible implementation method of an embodiment of the present disclosure, the second determination unit is specifically used to input the vehicle status information into a preset neural network model to obtain scaling information output by the neural network model; and determine the basic model parameter information and attenuation parameter information corresponding to the vehicle status information based on the scaling information, reference model parameter information and reference attenuation parameter information.

[0114] As a possible implementation method of an embodiment of the present disclosure, the third determination unit is specifically used to determine the serial number of each prediction time point in the prediction time domain; for each prediction time point, determine the attenuation ratio information at the prediction time point based on the attenuation parameter information and the serial number; determine the model parameter information at the prediction time point based on the basic model parameter information and the attenuation ratio information at the prediction time point.

[0115] As a possible implementation method of an embodiment of the present disclosure, the model parameter information includes: lateral model parameter information and longitudinal model parameter information; the lateral model parameter information includes: lateral error term penalty weight, lateral control amount penalty weight and lateral control amount increment penalty weight; the longitudinal model parameter information includes: longitudinal error term penalty weight, longitudinal control amount penalty weight and longitudinal control amount increment penalty weight.

[0116] As a possible implementation method of an embodiment of the present disclosure, the model parameter information includes: a lateral error term penalty weight and a longitudinal error term penalty weight; the attenuation parameter information includes: a first attenuation parameter for the lateral error term penalty weight and a second attenuation parameter for the longitudinal error term penalty weight.

[0117] As a possible implementation method of an embodiment of the present disclosure, the model parameter information includes: lateral model parameter information and longitudinal model parameter information; the second determination module 420 includes: a fifth determination unit, a sixth determination unit and a seventh determination unit; wherein the fifth determination unit is used to utilize the lateral model in the model predictive controller to determine the lateral control amount at each of the predicted time points in the model parameter sequence according to the lateral model parameter information at each of the predicted time points and the expected trajectory; the sixth determination unit is used to utilize the longitudinal model in the model predictive controller to determine the longitudinal control amount at each of the predicted time points in the model parameter sequence according to the longitudinal model parameter information at each of the predicted time points and the expected trajectory; the seventh determination unit is used to determine the control amount sequence according to the lateral control amount and the longitudinal control amount at each of the predicted time points.

[0118] As a possible implementation method of an embodiment of the present disclosure, the fifth determination unit is specifically used to determine the lateral vehicle state parameters in the vehicle state information, and the initial lateral control amount at each of the predicted time points; using the lateral model, according to the lateral vehicle state parameters and the initial lateral control amount at each of the predicted time points, determine the predicted lateral vehicle state parameters at each of the predicted time points; construct a first objective function according to the predicted lateral vehicle state parameters at each of the predicted time points, the expected lateral vehicle state parameters at each of the predicted time points in the expected trajectory, and the lateral model parameter information; according to the value of the first objective function, adjust the initial lateral control amount at each of the predicted time points to obtain the lateral control amount at each of the predicted time points.

[0119] As a possible implementation manner of an embodiment of the present disclosure, the sixth determination unit is specifically used to determine the longitudinal vehicle state parameters in the vehicle state information and the initial longitudinal control amount at each of the prediction time points; using the longitudinal model, based on the longitudinal vehicle state parameters and the initial longitudinal control amount at each of the prediction time points, determine the predicted longitudinal vehicle state parameters at each of the prediction time points; construct a second objective function based on the predicted longitudinal vehicle state parameters at each of the prediction time points, the expected longitudinal vehicle state parameters at each of the prediction time points in the expected trajectory, and the longitudinal model parameter information; and according to the value of the second objective function, adjust the initial longitudinal control amount at each of the prediction time points to obtain the longitudinal control amount at each of the prediction time points.

[0120] As a possible implementation method of an embodiment of the present disclosure, the vehicle state information includes at least one of the following vehicle state parameters: lateral vehicle state parameter, longitudinal vehicle state parameter, expected lateral vehicle state parameter, expected longitudinal vehicle state parameter, lateral vehicle state parameter error, and longitudinal vehicle state parameter error.

[0121] The automatic driving control device of the vehicle in the embodiment of the present disclosure determines the current vehicle state information of the vehicle and the model parameter sequence corresponding to the vehicle state information, wherein the model parameter sequence includes: model parameter information of the model predictive controller at each prediction time point in the prediction time domain, wherein the model parameter information at different prediction time points is different; using the model predictive controller, according to the model parameter sequence and the expected trajectory of the vehicle, a control quantity sequence is determined, wherein the control quantity sequence includes: lateral control quantity and longitudinal control quantity at each prediction time point; according to the lateral control quantity and longitudinal control quantity at the first prediction time point in the prediction time domain, the vehicle is automatically driven and controlled, so that the model parameter sequence can be determined based on the current vehicle state information of the vehicle; and different model parameter information is used for different prediction time points to accelerate the convergence speed of the model predictive controller, thereby improving the determination speed of the control quantity sequence and improving the automatic driving control efficiency of the vehicle.

[0122] In the technical solutions disclosed herein, the collection, storage, use, processing, transmission, provision and disclosure of user personal information are all carried out with the user's consent, comply with relevant laws and regulations, and do not violate public order and good morals.

[0123] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0124] According to an embodiment of the present disclosure, the present disclosure also provides a vehicle, including the automatic driving control device for the vehicle as shown in the fourth embodiment.

[0125] Figure 5 A schematic block diagram of an example electronic device 500 that can be used to implement embodiments 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 assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are provided as examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0126] like Figure 5As shown, the device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. Various programs and data required for the operation of the device 500 can also be stored in the RAM 503. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0127] Various components in device 500 are connected to I / O interface 505, including: an input unit 506, such as a keyboard, mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a magnetic disk, optical disk, etc.; and a communication unit 509, such as a network card, modem, wireless communication transceiver, etc. The communication unit 509 allows device 500 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0128] The computing unit 501 can be a variety of general-purpose and / or specialized processing components with processing and computing capabilities. Some examples of the computing unit 501 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 that run machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 501 performs the various methods and processes described above, such as the autonomous driving control method for a vehicle. For example, in some embodiments, the autonomous driving control method for a vehicle can be implemented as a computer software program that is tangibly contained in a machine-readable medium, such as the storage unit 508. In some embodiments, part or all of the computer program can be loaded and / or installed on the device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded into the RAM 503 and executed by the computing unit 501, one or more steps of the autonomous driving control method for a vehicle described above can be performed. Alternatively, in other embodiments, the computing unit 501 can be configured to perform the autonomous driving control method for a vehicle by any other appropriate means (e.g., by means of firmware).

[0129] Various embodiments of the systems and techniques described 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), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes 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.

[0130] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0131] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can 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.

[0132] 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).

[0133] The systems and techniques described herein can 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 having 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 can 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.

[0134] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0135] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not a limitation herein.

[0136] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A method for controlling an automatic driving of a vehicle, comprising: Determine current vehicle status information of the vehicle; Inputting the vehicle state information into a preset neural network model to obtain scaling information output by the neural network model; Determining basic model parameter information and attenuation parameter information corresponding to the vehicle state information according to the scaling ratio information, the reference model parameter information, and the reference attenuation parameter information; Determining model parameter information at each prediction time point within a prediction time domain according to the basic model parameter information and the attenuation parameter information; Determining a model parameter sequence corresponding to the vehicle state information based on the model parameter information at each prediction time point within the prediction time domain, wherein the model parameter sequence includes: model parameter information of a model prediction controller at each prediction time point within the prediction time domain, wherein the model parameter information at different prediction time points is different; Determining a control variable sequence using the model predictive controller according to the model parameter sequence and the desired trajectory of the vehicle, wherein the control variable sequence includes: a lateral control variable and a longitudinal control variable at each of the predicted time points; The vehicle is automatically driven and controlled according to the lateral control amount and the longitudinal control amount at the first prediction time point in the prediction time domain.

2. The method according to claim 1, wherein The determining, based on the basic model parameter information and the attenuation parameter information, the model parameter information at each prediction time point in the prediction time domain includes: Determining the sequence number of each of the prediction time points in the prediction time domain; For each predicted time point, determining the attenuation ratio information at the predicted time point according to the attenuation parameter information and the sequence number; The model parameter information at the prediction time point is determined according to the basic model parameter information and the attenuation ratio information at the prediction time point.

3. The method according to any one of claims 1 to 2, wherein: The model parameter information includes: transverse model parameter information and longitudinal model parameter information; The lateral model parameter information includes: lateral error term penalty weight, lateral control amount penalty weight and lateral control amount increment penalty weight; The longitudinal model parameter information includes: a longitudinal error term penalty weight, a longitudinal control amount penalty weight, and a longitudinal control amount increment penalty weight.

4. The method according to any one of claims 1 to 2, wherein: The model parameter information includes: horizontal error term penalty weight and vertical error term penalty weight; The attenuation parameter information includes: a first attenuation parameter for the penalty weight of the lateral error term and a second attenuation parameter for the penalty weight of the longitudinal error term.

5. The method according to claim 1, wherein The model parameter information includes: lateral model parameter information and longitudinal model parameter information; the use of the model predictive controller to determine the control amount sequence according to the model parameter sequence and the desired trajectory of the vehicle includes: Determining a lateral control amount at each of the predicted time points according to the lateral model parameter information at each of the predicted time points in the model parameter sequence and the desired trajectory using the lateral model in the model predictive controller; Determining the longitudinal control amount at each of the predicted time points according to the longitudinal model information at each of the predicted time points in the model parameter sequence and the expected trajectory using the longitudinal model in the model predictive controller; The control amount sequence is determined according to the lateral control amount and the longitudinal control amount at each of the predicted time points.

6. The method according to claim 5, wherein: The method of utilizing the lateral model in the model predictive controller to determine the lateral control amount at each of the predicted time points according to the lateral model parameter information at each of the predicted time points in the model parameter sequence and the expected trajectory includes: determining a lateral vehicle state parameter in the vehicle state information and an initial lateral control amount at each of the predicted time points; Determining, using the lateral model, a predicted lateral vehicle state parameter at each of the predicted time points based on the lateral vehicle state parameter and the initial lateral control amount at each of the predicted time points; constructing a first objective function based on the predicted lateral vehicle state parameters at each of the predicted time points, the expected lateral vehicle state parameters at each of the predicted time points in the expected trajectory, and the lateral model parameter information; According to the value of the first objective function, the initial lateral control amount at each of the predicted time points is adjusted to obtain the lateral control amount at each of the predicted time points.

7. The method according to claim 5, wherein: The method of utilizing the longitudinal model in the model predictive controller to determine the longitudinal control amount at each of the predicted time points according to the longitudinal model parameter information at each of the predicted time points in the model parameter sequence and the expected trajectory includes: Determining longitudinal vehicle state parameters in the vehicle state information and initial longitudinal control amounts at each of the predicted time points; Determining, using the longitudinal model, a predicted longitudinal vehicle state parameter at each of the predicted time points based on the longitudinal vehicle state parameter and the initial longitudinal control amount at each of the predicted time points; constructing a second objective function based on the predicted longitudinal vehicle state parameters at each of the predicted time points, the expected longitudinal vehicle state parameters at each of the predicted time points in the expected trajectory, and the longitudinal model parameter information; According to the value of the second objective function, the initial longitudinal control amount at each of the predicted time points is adjusted to obtain the longitudinal control amount at each of the predicted time points.

8. The method according to any one of claims 1 to 2, wherein: The vehicle state information includes at least one of the following vehicle state parameters: a lateral vehicle state parameter, a longitudinal vehicle state parameter, an expected lateral vehicle state parameter, an expected longitudinal vehicle state parameter, a lateral vehicle state parameter error, and a longitudinal vehicle state parameter error.

9. A vehicle automatic driving control device, comprising: a first determining module, configured to determine current vehicle state information of the vehicle and a model parameter sequence corresponding to the vehicle state information, wherein the model parameter sequence includes model parameter information of a model predictive controller at each prediction time point within a prediction time domain, wherein the model parameter information at different prediction time points is different; a second determination module, configured to determine, using the model predictive controller, a control variable sequence according to the model parameter sequence and the desired trajectory of the vehicle, wherein the control variable sequence includes: a lateral control variable and a longitudinal control variable at each of the predicted time points; a control module, configured to perform automatic driving control processing on the vehicle based on the lateral control amount and the longitudinal control amount at the first prediction time point within the prediction time domain; The first determining module includes: a first determining unit, a second determining unit, a third determining unit and a fourth determining unit; Wherein, the first determining unit is used to determine the current vehicle status information of the vehicle; The second determining unit is used to determine the basic model parameter information and the attenuation parameter information corresponding to the vehicle state information; The third determining unit is configured to determine the model parameter information at each prediction time point in the prediction time domain according to the basic model parameter information and the attenuation parameter information; The fourth determining unit is configured to determine a model parameter sequence corresponding to the vehicle state information based on the model parameter information at each prediction time point within the prediction time domain; The second determining unit is specifically configured to: Inputting the vehicle state information into a preset neural network model to obtain scaling information output by the neural network model; According to the scaling information, the reference model parameter information and the reference attenuation parameter information, the basic model parameter information and the attenuation parameter information corresponding to the vehicle state information are determined.

10. The device according to claim 9, wherein The third determining unit is specifically configured to: Determining the sequence number of each of the prediction time points in the prediction time domain; For each predicted time point, determining the attenuation ratio information at the predicted time point according to the attenuation parameter information and the sequence number; The model parameter information at the prediction time point is determined according to the basic model parameter information and the attenuation ratio information at the prediction time point.

11. The device according to any one of claims 9 to 10, wherein: The model parameter information includes: transverse model parameter information and longitudinal model parameter information; The lateral model parameter information includes: lateral error term penalty weight, lateral control amount penalty weight and lateral control amount increment penalty weight; The longitudinal model parameter information includes: a longitudinal error term penalty weight, a longitudinal control amount penalty weight, and a longitudinal control amount increment penalty weight.

12. The device according to any one of claims 9 to 10, wherein: The model parameter information includes: horizontal error term penalty weight and vertical error term penalty weight; The attenuation parameter information includes: a first attenuation parameter for the penalty weight of the lateral error term and a second attenuation parameter for the penalty weight of the longitudinal error term.

13. The device according to claim 9, wherein The model parameter information includes: transverse model parameter information and longitudinal model parameter information; the second determination module includes: a fifth determination unit, a sixth determination unit and a seventh determination unit; The fifth determining unit is configured to determine the lateral control amount at each of the predicted time points according to the lateral model parameter information at each of the predicted time points in the model parameter sequence and the expected trajectory using the lateral model in the model predictive controller; the sixth determining unit is configured to determine the longitudinal control amount at each of the predicted time points according to the longitudinal model parameter information at each of the predicted time points in the model parameter sequence and the expected trajectory using the longitudinal model in the model predictive controller; The seventh determining unit is configured to determine the control amount sequence according to the lateral control amount and the longitudinal control amount at each of the predicted time points.

14. The device according to claim 13, characterized in that The fifth determining unit is specifically configured to: determining a lateral vehicle state parameter in the vehicle state information and an initial lateral control amount at each of the predicted time points; Determining, using the lateral model, a predicted lateral vehicle state parameter at each of the predicted time points based on the lateral vehicle state parameter and the initial lateral control amount at each of the predicted time points; constructing a first objective function based on the predicted lateral vehicle state parameters at each of the predicted time points, the expected lateral vehicle state parameters at each of the predicted time points in the expected trajectory, and the lateral model parameter information; According to the value of the first objective function, the initial lateral control amount at each of the predicted time points is adjusted to obtain the lateral control amount at each of the predicted time points.

15. The device according to claim 13, wherein The sixth determining unit is specifically configured to: Determining longitudinal vehicle state parameters in the vehicle state information and initial longitudinal control amounts at each of the predicted time points; Determining, using the longitudinal model, a predicted longitudinal vehicle state parameter at each of the predicted time points based on the longitudinal vehicle state parameter and the initial longitudinal control amount at each of the predicted time points; constructing a second objective function based on the predicted longitudinal vehicle state parameters at each of the predicted time points, the expected longitudinal vehicle state parameters at each of the predicted time points in the expected trajectory, and the longitudinal model parameter information; According to the value of the second objective function, the initial longitudinal control amount at each of the predicted time points is adjusted to obtain the longitudinal control amount at each of the predicted time points.

16. The device according to any one of claims 9 to 10, wherein: The vehicle state information includes at least one of the following vehicle state parameters: a lateral vehicle state parameter, a longitudinal vehicle state parameter, an expected lateral vehicle state parameter, an expected longitudinal vehicle state parameter, a lateral vehicle state parameter error, and a longitudinal vehicle state parameter error.

17. 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 8.

18. 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 1-8.

19. A computer program product comprising a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 8.

20. A vehicle comprising: An automatic driving control device for a vehicle as claimed in any one of claims 9 to 16.

Images