Control method of hybrid vehicle queue under curve road

By extracting historical driving data and combining radical coefficient estimator and neural network model to predict speed, the problem of low control accuracy of autonomous driving vehicles under curved roads is solved, and more accurate vehicle queue control is achieved.

CN118907091APending Publication Date: 2024-11-08TIANJIN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411205024.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The complex road conditions and driver driving style of curved roads make it difficult to accurately predict the movement status of autonomous vehicles, resulting in lower control accuracy of autonomous vehicles in hybrid traffic.

Method used

By extracting the historical driving data of the pilot vehicle from the mixed vehicle queue, using the radical coefficient estimator, neural network model and constant acceleration prediction model, the speed sequence of the pilot vehicle is predicted, and the autonomous driving vehicle is controlled based on the prediction results to improve the control accuracy .

Benefits of technology

The control accuracy of autonomous vehicles in hybrid vehicle queues is improved to ensure the safety and operation efficiency of vehicle queues.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118907091A_ABST
    Figure CN118907091A_ABST
Patent Text Reader

Abstract

The invention provides a method for controlling a hybrid vehicle queue on a curved road, which can be applied to the technical field of intelligent driving. The method comprises the following steps: extracting vehicle motion characteristics from historical driving data of a pilot vehicle in a hybrid vehicle queue, wherein the hybrid vehicle queue comprises an automatic driving following vehicle and the pilot vehicle driven by a driver; the vehicle motion features are input into an aggressive coefficient estimator, the aggressive coefficient of the driver is obtained, and the aggressive coefficient estimator is constructed based on a multi-feature decision analysis method; the road curvature sequence, the historical speed sequence and the aggressive coefficient are input into a vehicle speed prediction model, a speed prediction sequence of the pilot vehicle is obtained, the vehicle speed prediction model is composed of a neural network model and a constant acceleration prediction model, and the constant acceleration prediction model sets the acceleration of the pilot vehicle in the driving process to be a constant value; and designing a distributed model prediction control algorithm according to the speed prediction sequence of the pilot vehicle to control the autonomous vehicle in the hybrid vehicle queue.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the field of intelligent driving technology, and more specifically, to a control method for a mixed vehicle queue on a curved road. Background Art

[0002] With the advancement of autonomous driving and network communication technology, autonomous vehicles with vehicle-to-vehicle communication functions can significantly improve traffic efficiency. In the future, mixed traffic with human-driven vehicles and autonomous vehicles will exist for a long time. Usually, the speed and direction of the autonomous vehicle are controlled according to the movement status of vehicles near the autonomous vehicle to ensure traffic safety.

[0003] However, the complex road conditions on curved roads and the driver's changing driving style make it difficult to accurately predict the motion state of the vehicle driven by the driver, resulting in low control accuracy of autonomous vehicles in mixed traffic. Summary of the invention

[0004] In view of this, the present disclosure provides a method for controlling a mixed vehicle queue on a curved road.

[0005] One aspect of the present disclosure provides a control method for a mixed vehicle queue on a curved road, comprising: extracting vehicle motion features from historical driving data of a pilot car in the mixed vehicle queue, wherein the mixed vehicle queue includes an autonomous driving vehicle and a pilot car driven by a driver; inputting the vehicle motion features into an aggressive coefficient estimator to obtain an aggressive coefficient of the driver, wherein the aggressive coefficient estimator is constructed based on a multi-feature decision analysis method; inputting a road curvature sequence, a historical speed sequence, and an aggressive coefficient into a vehicle speed prediction model to obtain a speed prediction sequence of the pilot car, wherein the road curvature sequence represents the road curvature information of the pilot car during a driving time period, wherein the historical speed sequence represents the speed information of the pilot car during a driving time period, wherein the vehicle speed prediction model is composed of a neural network model and a constant acceleration prediction model, wherein the constant acceleration prediction model sets the acceleration of the pilot car during driving to a constant value; and controlling the autonomous driving vehicle in the mixed vehicle queue according to the speed prediction sequence of the pilot car.

[0006] Another aspect of the present disclosure provides a control device for a mixed vehicle queue on a curved road, including: an extraction module for extracting vehicle motion features from historical driving data of a pilot car in the mixed vehicle queue, wherein the mixed vehicle queue includes an autonomous driving vehicle and a pilot car driven by a driver; a first input module for inputting the vehicle motion features into an aggressive coefficient estimator to obtain the driver's aggressive coefficient, wherein the aggressive coefficient estimator is constructed based on a multi-feature decision analysis method; a second input module for inputting a road curvature sequence, a historical speed sequence and an aggressive coefficient into a vehicle speed prediction model to obtain a speed prediction sequence of the pilot car, wherein the road curvature sequence represents the road curvature information of the pilot car during the driving time period, and the historical speed sequence represents the speed information of the pilot car during the driving time period, and the vehicle speed prediction model is composed of a neural network model and a constant acceleration prediction model, wherein the constant acceleration prediction model sets the acceleration of the pilot car during driving to a constant value; and a control module for controlling the autonomous driving vehicle in the mixed vehicle queue according to the speed prediction sequence of the pilot car.

[0007] Another aspect of the present disclosure provides an electronic device, comprising:

[0008] one or more processors;

[0009] a memory for storing one or more programs,

[0010] When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described above.

[0011] Another aspect of the present disclosure provides a computer-readable storage medium storing computer-executable instructions, which are used to implement the above method when executed.

[0012] Another aspect of the present disclosure provides a computer program product, which includes computer executable instructions, and when the instructions are executed, are used to implement the method as described above.

[0013] According to an embodiment of the present disclosure, the driver's driving style information is obtained by extracting vehicle motion features from historical driving data of a leading vehicle in a mixed vehicle queue and then inputting the vehicle motion features into an aggressive coefficient estimator to obtain the driver's aggressive coefficient.

[0014] Since the road curvature sequence, historical speed sequence and aggressive coefficient are input into the vehicle speed prediction model, the speed prediction sequence of the pilot vehicle is more accurate, taking into account the influence of road curvature changes and the driver's driving style. At the same time, the vehicle speed prediction model is composed of a neural network model and a constant acceleration prediction model, that is, the fusion of the theoretical model and the neural network can increase the accuracy of the speed prediction sequence of the pilot vehicle. Then, the automatic driving vehicle in the mixed vehicle queue is precisely controlled according to the speed prediction sequence of the pilot vehicle, which improves the control accuracy of the automatic driving vehicle in the mixed vehicle queue. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The above and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0016] Figure 1 A system architecture diagram schematically shows a control method for mixed vehicle queues on a curved road that can be applied according to an embodiment of the present disclosure;

[0017] Figure 2 A flow chart schematically shows a method for controlling a mixed vehicle queue on a curved road according to an embodiment of the present disclosure;

[0018] Figure 3 A schematic diagram of a vehicle speed prediction model according to an embodiment of the present disclosure is schematically shown;

[0019] Figure 4 Schematically shows a mixed vehicle platoon scenario operation diagram according to an embodiment of the present disclosure;

[0020] Figure 5 A schematic diagram of a path coupling expansion look-ahead strategy according to an embodiment of the present disclosure is shown schematically;

[0021] Figure 6 A schematic diagram of an experimental scenario of a method for controlling a mixed vehicle queue on a curved road according to an embodiment of the present disclosure is schematically shown;

[0022] Figure 7 A schematic diagram showing a trajectory diagram of a mixed vehicle platoon traveling on a curved road according to an embodiment of the present disclosure;

[0023] Figure 8 A diagram schematically shows a lateral deviation error diagram of a mixed vehicle platoon traveling on a curved road according to an embodiment of the present disclosure;

[0024] Fig. 9 A schematic diagram of a heading angle of a mixed vehicle platoon traveling on a curved road according to an embodiment of the present disclosure is shown;

[0025] Fig.10A diagram schematically shows a heading angle error diagram of a mixed vehicle platoon traveling on a curved road according to an embodiment of the present disclosure;

[0026] Fig.11 A diagram schematically shows a predicted speed of a pilot vehicle on a curved road according to an embodiment of the present disclosure;

[0027] Fig.12 A vehicle spacing diagram of a mixed vehicle platoon traveling on a curved road according to an embodiment of the present disclosure is schematically shown;

[0028] Fig.13 A block diagram schematically shows a control device for a mixed vehicle queue on a curved road according to an embodiment of the present disclosure;

[0029] Fig.14 A block diagram of an electronic device suitable for implementing a control method for a mixed vehicle queue on a curved road according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0030] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0031] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0032] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0033] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0034] In the embodiments of the present disclosure, the collection, updating, analysis, processing, use, transmission, provision, disclosure, storage, etc. of the data involved (for example, including but not limited to user personal information) are in compliance with the provisions of relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. In particular, necessary measures are taken for user personal information to prevent illegal access to user personal information data and maintain the security of user personal information, network security, and national security.

[0035] In the embodiments of the present disclosure, the user's authorization or consent is obtained before obtaining or collecting the user's personal information.

[0036] The complex road conditions on curved roads and the driver's changing driving style make it difficult to accurately predict the motion state of the vehicle driven by the driver, which will result in low control accuracy of the autonomous driving vehicle in mixed traffic.

[0037] In view of this, an embodiment of the present disclosure provides a control method for a mixed vehicle queue on a curved road, comprising: extracting vehicle motion features from historical driving data of a pilot car in the mixed vehicle queue, wherein the mixed vehicle queue includes an autonomous driving vehicle and a pilot car driven by a driver; inputting the vehicle motion features into an aggressive coefficient estimator to obtain the driver's aggressive coefficient, wherein the aggressive coefficient estimator is constructed based on a multi-feature decision analysis method; inputting a road curvature sequence, a historical speed sequence and an aggressive coefficient into a vehicle speed prediction model to obtain a speed prediction sequence of the pilot car, wherein the road curvature sequence represents the road curvature information of the pilot car during the driving time period, and the historical speed sequence represents the speed information of the pilot car during the driving time period, wherein the vehicle speed prediction model is composed of a neural network model and a constant acceleration prediction model, wherein the constant acceleration prediction model sets the acceleration of the pilot car during driving to a constant value; and controlling the autonomous driving vehicle in the mixed vehicle queue according to the speed prediction sequence of the pilot car.

[0038] Figure 1 The system architecture diagram of the control method for mixed vehicle queues on curved roads according to an embodiment of the present disclosure is schematically shown. It should be noted that: Figure 1 What is shown is merely an example of a system architecture 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 it does not mean that the embodiments of the present disclosure cannot be used in other devices, systems, environments or scenarios.

[0039] like Figure 1As shown, the system architecture 100 according to this embodiment may include a pilot vehicle 101, an autonomous driving vehicle 102, an autonomous driving vehicle 102 in a mixed vehicle platoon, a network 104, and a server 105. The network 104 is used as a medium for providing a communication link between the pilot vehicle 101, the autonomous driving vehicle 102, and the autonomous driving vehicle 102. The network 104 may include various connection types, such as wired and / or wireless communication links, and the like.

[0040] The user may use the pilot car 101, the autonomous driving car 102, and the autonomous driving car 103 to interact with the server 105 through the network 104 to receive or send messages, etc. The pilot car 101, the autonomous driving car 102, and the autonomous driving car 103 may be installed with various communication client applications, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, and / or social platform software, etc. (only as examples).

[0041] The pilot car 101, the autonomous driving car 102, and the autonomous driving car 102 may be various electronic devices having a display screen and supporting web browsing.

[0042] It should be noted that the control method for a mixed vehicle queue on a curved road provided in the embodiment of the present disclosure can generally be executed by the server 105 .

[0043] It should be understood that Figure 1 The number of pilot vehicles, autonomous driving vehicles, networks, and servers in the embodiment is only for illustration purposes. Any number of pilot vehicles, autonomous driving vehicles, networks, and servers may be provided as required.

[0044] Figure 2 A flow chart schematically shows a method for controlling a mixed vehicle queue on a curved road according to an embodiment of the present disclosure.

[0045] like Figure 2 As shown, the method includes operations S210 to S240.

[0046] In operation S210 , vehicle motion features are extracted from historical driving data of a leading vehicle in a mixed vehicle queue.

[0047] According to an embodiment of the present disclosure, a mixed vehicle platoon includes an autonomous driving vehicle and a pilot vehicle driven by a human driver.

[0048] According to an embodiment of the present disclosure, an autonomous driving vehicle needs to be controlled by a control algorithm so that the desired vehicle spacing is maintained between vehicles, thereby reducing air resistance, lowering energy consumption, and ensuring the safety of the vehicle platoon system.

[0049] According to an embodiment of the present disclosure, the pilot vehicle is at a target position of a mixed vehicle queue. For example, the target position may be the first vehicle in the mixed vehicle queue.

[0050] According to an embodiment of the present disclosure, the historical driving data may be heading angle, driving speed, road curvature, road position, driving acceleration, front wheel deviation angle, vehicle spacing, etc.

[0051] According to an embodiment of the present disclosure, the vehicle motion feature is used to characterize the motion state of the vehicle during driving. The vehicle motion state may be related to driving speed, driving acceleration, driving direction, etc.

[0052] In operation S220, the vehicle motion characteristics are input into an aggressive coefficient estimator to obtain the aggressive coefficient of the driver. The aggressive coefficient estimator is constructed based on a multi-feature decision analysis method.

[0053] According to an embodiment of the present disclosure, the aggressiveness coefficient may characterize the driving style of the driver.

[0054] According to an embodiment of the present disclosure, a multi-feature decision analysis method is used to rank the options according to multiple criteria. For example, the multi-feature decision analysis method may be a TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) method. When selecting the option closest to the ideal solution, while staying away from the negative ideal solution. When calculating the driver aggressiveness coefficient, the TOPSIS method can be used to evaluate the aggressiveness of different drivers and compare them based on multiple criteria.

[0055] In operation S230 , the road curvature sequence, the historical speed sequence, and the aggressiveness coefficient are input into a vehicle speed prediction model to obtain a speed prediction sequence of the pilot vehicle.

[0056] According to an embodiment of the present disclosure, the road curvature sequence represents the road curvature information of the pilot vehicle during the driving time period. For example, a data point in the road curvature sequence represents the road curvature of the pilot vehicle at a time point during the driving time period.

[0057] According to an embodiment of the present disclosure, the historical speed sequence represents the speed information of the pilot vehicle during the driving time period. For example, a data point in the historical speed sequence represents the speed of the pilot vehicle at a time point during the driving time period.

[0058] According to an embodiment of the present disclosure, the vehicle speed prediction model is composed of a neural network model and a constant acceleration prediction model. The constant acceleration prediction model sets the acceleration of the pilot vehicle during driving to a constant value.

[0059] In operation S240 , the automatic driving vehicles in the mixed vehicle queue are controlled according to the speed prediction sequence of the pilot vehicle.

[0060] According to an embodiment of the present disclosure, the expected vehicle spacing between vehicles is first calculated, and then the autonomous driving vehicle in the mixed vehicle queue is controlled based on the speed prediction sequence of the pilot vehicle. For example, the control method can be an adaptive control algorithm, a fuzzy control algorithm, etc.

[0061] According to an embodiment of the present disclosure, the driver's driving style information is obtained by extracting vehicle motion features from historical driving data of a leading vehicle in a mixed vehicle queue and then inputting the vehicle motion features into an aggressive coefficient estimator to obtain the driver's aggressive coefficient.

[0062] Since the road curvature sequence, historical speed sequence and aggressive coefficient are input into the vehicle speed prediction model, the speed prediction sequence of the pilot vehicle is more accurate, taking into account the influence of road curvature changes and the driver's driving style. At the same time, the vehicle speed prediction model is composed of a neural network model and a constant acceleration prediction model, that is, the fusion of the theoretical model and the neural network can increase the accuracy of the speed prediction sequence of the pilot vehicle. Then, the automatic driving vehicle in the mixed vehicle queue is precisely controlled according to the speed prediction sequence of the pilot vehicle, which improves the control accuracy of the automatic driving vehicle in the mixed vehicle queue.

[0063] According to an embodiment of the present disclosure, a road curvature sequence, a historical speed sequence and an aggressive coefficient are input into a vehicle speed prediction model to obtain a speed prediction sequence of a pilot car, including: inputting a road curvature sequence, a historical speed sequence and an aggressive coefficient into a neural network model in the vehicle speed prediction model to obtain a speed compensation prediction sequence of the pilot car; inputting the historical speed sequence into a constant acceleration prediction model of the vehicle speed prediction model to obtain a theoretical speed prediction sequence of the pilot car; and obtaining a speed prediction sequence of the pilot car based on the speed compensation prediction sequence and the theoretical speed prediction sequence.

[0064] According to an embodiment of the present disclosure, the neural network model can perform feature learning on the input road curvature sequence, historical speed sequence and aggressiveness coefficient to obtain important information that can accurately predict the speed of the pilot vehicle; the neural network model derives a speed compensation prediction sequence based on the important information that can accurately predict the speed of the pilot vehicle.

[0065] According to an embodiment of the present disclosure, the theoretical speed prediction sequence is constructed based on time, speed and constant acceleration. For example, y=v+at, a is constant acceleration, t is time, v is speed, and y is theoretical speed.

[0066] According to an embodiment of the present disclosure, the theoretical speed prediction sequence is based on the historical speed sequence, the average acceleration is calculated, and then the speed at each future moment is calculated based on the average acceleration. The length of the historical speed sequence and the length of the future speed sequence may be unequal.

[0067] According to an embodiment of the present disclosure, a road curvature sequence, a historical speed sequence and an aggressiveness coefficient are input into a neural network model in a vehicle speed prediction model to obtain a speed compensation prediction sequence for a pilot vehicle, including: extracting time features of the road curvature sequence, the historical speed sequence and the aggressiveness coefficient to obtain a time information sequence; performing residual calculation on the historical speed sequence to obtain a residual speed sequence; and obtaining a speed compensation prediction sequence for a pilot vehicle based on the residual speed sequence and the time information sequence.

[0068] According to an embodiment of the present disclosure, the time information sequence may be composed of a time feature vector arranged based on historical time points. The time feature vector may be composed of the road curvature, speed, and aggressiveness coefficient at a historical time point.

[0069] According to an embodiment of the present disclosure, residual calculation may be implemented by utilizing the residual structure in a neural network model.

[0070] According to the embodiments of the present disclosure, since the residual speed sequence can characterize the speed change rate of the pilot car, and the time information sequence has information such as the aggressiveness coefficient, speed at different road curvatures, and aggressiveness coefficient at different road curvatures that can reflect the driver's driving style information, the driving information of the pilot car is enriched. Therefore, based on the residual speed sequence and the time information sequence, the speed compensation prediction sequence of the pilot car can be accurately obtained.

[0071] According to an embodiment of the present disclosure, time features are extracted from a road curvature sequence, a historical speed sequence, and an aggressiveness coefficient to obtain a time information sequence, including: splicing the road curvature sequence, the historical speed sequence, and the aggressiveness coefficient to obtain a spliced ​​feature vector with an increased number of columns; performing feature learning on the spliced ​​feature vector using a multi-layer perceptron of a neural network model to obtain an initial feature vector; stacking the initial feature vector with the road curvature feature vector to obtain a target feature vector with an increased number of rows; and inputting the target feature vector into a time decoder of the neural network model to obtain a time information sequence.

[0072] According to an embodiment of the present disclosure, a multi-layer perceptron may include a dense encoder and a dense decoder.

[0073] According to an embodiment of the present disclosure, the road curvature feature vector is spliced ​​after the last row of the initial feature vector to achieve stacking of the initial feature vector and the road curvature feature vector. Since the road curvature feature is very important information for the control of mixed vehicle queues, it is possible that part of the information is missing when the spliced ​​feature vector is learned using the multi-layer perceptron of the neural network model. Therefore, the initial feature vector and the road curvature feature vector are stacked, so that the subsequent time decoder of the neural network model is used to decode the target feature vector, that is, as time changes, the information of the road curvature is more detailed, making the speed prediction more accurate.

[0074] According to an embodiment of the present disclosure, the road curvature sequence, the historical speed sequence and the aggressiveness coefficient are spliced ​​to obtain a spliced ​​feature vector with an increased number of columns, which also includes: performing feature projection on the road curvature sequence to obtain a road curvature feature vector; splicing the historical speed sequence, the aggressiveness coefficient and the road curvature feature vector to obtain a spliced ​​feature vector with an increased number of columns.

[0075] According to an embodiment of the present disclosure, the road curvature feature vector is concatenated with the aggressive coefficient, and then concatenated with the feature vector of the historical speed sequence. The aggressive coefficient may also be in a sequence form, for example, the aggressive coefficient sequence may be composed of the aggressive coefficients of the drivers arranged in time based on the historical speed sequence.

[0076] Figure 3 A schematic diagram of a vehicle speed prediction model according to an embodiment of the present disclosure is schematically shown.

[0077] like Figure 3 As shown, for the road curvature sequence Perform feature projection to obtain the road curvature feature vector ; The historical speed sequence v 1-L , Radical coefficient and the road curvature eigenvector The concatenated feature vector is concatenated to obtain a concatenated feature vector with an increased number of columns. The concatenated feature vector is subjected to feature learning by a multi-layer perceptron of a neural network model to obtain an initial feature vector g. For example, the multi-layer perceptron includes a dense encoder and a dense decoder. The concatenated feature vector is decoded by a dense encoder of a neural network model to obtain a decoded vector e; the concatenated feature vector is decoded by a dense decoder of a neural network model to obtain an initial feature vector g. The initial feature vector g and the road curvature feature vector are combined. Stacking is performed to obtain the target feature vector with increasing number of rows . The target feature vector Input the time decoder of the neural network model to obtain the time information sequence.

[0078] The historical speed sequence is input into the linear layer of the neural network model to realize the residual calculation of the historical speed sequence, thereby obtaining the residual speed sequence. According to the residual speed sequence and the time information sequence, the speed compensation sequence of the pilot vehicle is obtained. .

[0079] Input the historical speed sequence into the constant acceleration prediction model of the vehicle speed prediction model to obtain the theoretical speed prediction sequence of the pilot vehicle .

[0080] The speed compensation sequence Theoretical speed prediction sequence Add them together to get the speed prediction sequence of the pilot car .

[0081] At the same time, the sequence can be predicted based on the speed of the pilot car , the position information of the pilot vehicle is calculated. That is, the output of the vehicle speed prediction model can include a position prediction sequence and a speed prediction sequence.

[0082] According to an embodiment of the present disclosure, the vehicle motion characteristics are input into an aggressiveness coefficient estimator to obtain the driver's aggressiveness coefficient, including: using an entropy weight method to calculate the basic feature weights of the vehicle motion characteristics, and determining adaptive weights based on the basic feature weights; calculating the distance between the vehicle motion characteristics and the maximum value of the characteristics, and the distance between the vehicle motion characteristics and the minimum value of the characteristics based on the adaptive weights; determining the driver's aggressiveness coefficient based on the distance between the vehicle motion characteristics and the maximum value of the characteristics, and the distance between the vehicle motion characteristics and the minimum value of the characteristics.

[0083] The entropy weight method is used to calculate the basic feature weights of vehicle motion characteristics. The basic feature weight calculation formula is:

[0084] (1)

[0085] Here, m represents the number of features. represents the information entropy of the j-th vehicle motion feature, is the basic feature weight of the j-th vehicle motion feature.

[0086] (2)

[0087] n represents the number of samples of the pilot car, is the occurrence probability of the jth vehicle motion feature in the i-th sample, is the normalized feature of the jth feature of the i-th sample.

[0088] The adaptive weight is determined according to the basic feature weight. The adaptive weight calculation formula is:

[0089] (3)

[0090] is the adaptive weight of the j-th vehicle motion feature.

[0091] According to an embodiment of the present disclosure, the vehicle motion characteristics include average speed, maximum speed, average longitudinal acceleration, maximum longitudinal acceleration, average lateral acceleration, and maximum lateral acceleration. The formula of the driver's aggressiveness coefficient is as follows:

[0092] (4)

[0093] is the driver's aggressiveness coefficient, represents the distance between the motion feature of the i-th vehicle and the maximum value of the feature, Represents the distance between the i-th vehicle motion feature and the minimum feature value, where i is an integer greater than or equal to 1.

[0094] According to an embodiment of the present disclosure, the above method further includes:

[0095] According to the driver's aggressiveness coefficient, the expected vehicle spacing between vehicles in the mixed vehicle queue is determined. The formula for the expected vehicle spacing is:

[0096] (5)

[0097] In the formula, is the expected inter-vehicle distance of vehicle i in the mixed vehicle platoon, is the basic vehicle spacing, is the heading angle of vehicle i.

[0098] Figure 4 The operating diagram of a mixed vehicle platoon scenario according to an embodiment of the present disclosure is schematically shown.

[0099] like Figure 4 As shown in FIG. 1 , the mixed vehicle platoon is composed of multiple vehicles. Vehicle 0 can be a pilot vehicle. Vehicle i-1, vehicle i, and vehicle N are autonomous vehicles. (x N ,y N ) is the position of vehicle N, L N is the wheelbase of vehicle N, Represents the front wheel deflection angle of vehicle N. d i represents the lateral offset of vehicle i. i,j-1 represents the vehicle distance between vehicle i and vehicle i-1.

[0100] The control goal is to ensure that the mixed vehicle platoon travels along the road with a position error of 0 between the vehicle position and the road centerline, while ensuring that the spacing between adjacent vehicles is equal to the expected vehicle spacing. .

[0101] According to an embodiment of the present disclosure, controlling the autonomous driving vehicles in a mixed vehicle queue according to the speed prediction sequence of the pilot vehicle includes: controlling the autonomous driving vehicles in the mixed vehicle queue according to the speed prediction sequence of the pilot vehicle by a distributed model predictive control algorithm, the distributed model predictive control algorithm including a local model predictive control algorithm, each local model predictive control algorithm controlling one autonomous driving vehicle.

[0102] According to an embodiment of the present disclosure, a local model predictive control algorithm is composed of a vehicle kinematic model, a path coupling extended look-ahead strategy and a local optimal controller.

[0103] Vehicle kinematics model The formula is:

[0104] (6)

[0105] In the formula is the state quantity of vehicle i, is the output of vehicle i, is the control quantity of vehicle i. are the lateral position coordinates, longitudinal position coordinates and heading angle of vehicle i; and represent the speed and front wheel angle of vehicle i respectively, is the sampling period, is the wheelbase of vehicle i.

[0106] Figure 5 A schematic diagram of a path coupling extension look-ahead strategy according to an embodiment of the present disclosure is shown schematically.

[0107] like Figure 5 As shown in FIG. 1 , the figure is a schematic diagram of a path coupling expansion forward-looking strategy provided by an embodiment of the present invention; in some embodiments, the path coupling expansion forward-looking strategy includes

[0108] 1) Definition As the position vector of vehicle i, Mapping to path curve The nearest point on ,in , For waypoints The corresponding heading angle. is the position vector of vehicle j, for Mapping to path curve The nearest point on.

[0109] 2) Calculate the extended look-ahead vector .

[0110] (7)

[0111] in ,function The role of and Mapped to path points and And calculate the curve distance between two points along the path to get .

[0112] 3) In The extended vector You can get the location of the extended forward point .

[0113] (8)

[0114] The formula of the local optimal controller is as follows:

[0115] (9)

[0116] Among them, J i is the cost function, Np is the prediction time domain, l i is the stage cost function, represents the output of vehicle i obtained according to the vehicle kinematic model, The control amount of vehicle i obtained according to the vehicle kinematic model is: is the assumed state output of vehicle i, is the assumed control quantity of vehicle i, is the assumed state output of vehicle i-1, is the driver's aggressiveness factor.

[0117] The constraint condition formulas are shown in formulas (10) to (13).

[0118] (10)

[0119] Represents the assumed state of vehicle i obtained by the vehicle kinematic model.

[0120] (11)

[0121] Represents the identity matrix.

[0122] (12)

[0123] Represents the actual state quantity obtained by the sensor in vehicle i.

[0124] (13)

[0125] Represents the control range of vehicle i. That is, the control amount is limited by the power system and steering system. Function is the stage cost function associated with node i, defined as

[0126] (14)

[0127] In the formula is a positive definite symmetric weight matrix, , represents the predicted position vector of vehicle i, represents the inter-vehicle distance of vehicle i, Represents the position of the extended look-ahead point obtained by combining the predicted position of vehicle i with the assumed position of vehicle i-1. , represents the predicted heading angle of vehicle i, Represents the heading angle of the extended look-ahead point obtained by comparing the predicted position of vehicle i with the assumed position of vehicle i-1.

[0128] It is assumed that the control quantity is taken into account and the calculation method is:

[0129] (15)

[0130] in, represents the assumed speed of vehicle i-1, Indicates the radius of curvature of the reference path.

[0131] Acquire pilot vehicle data and road information data; construct a driver aggressiveness coefficient estimator based on the data; then calculate the driver's aggressiveness coefficient and design the vehicle spacing strategy of the vehicle queue based on the aggressiveness coefficient; then design a vehicle speed prediction model that integrates the mechanism and neural network based on the pilot vehicle data, road information data and the driver aggressiveness coefficient estimator; finally, predict the future state of the pilot vehicle based on the vehicle speed prediction model and use a distributed model predictive control algorithm to control the mixed vehicle queue.

[0132] An aggressive coefficient estimator was constructed based on the pilot car data, and the aggressive coefficient estimation of the human driver was made more reasonable through adaptive weights. A vehicle spacing strategy was designed based on the aggressive coefficient to make the vehicle queue system safer. A vehicle speed prediction model based on the mechanism fusion neural network was designed, which can more accurately predict the future state of the pilot car and improve the control accuracy of the spacing maintenance of the mixed vehicle queue.

[0133] Figure 6 A schematic diagram of an experimental scenario of a method for controlling a mixed vehicle queue on a curved road according to an embodiment of the present disclosure is schematically shown.

[0134] like Figure 6As shown in the figure, the experimental platform of the control method of mixed vehicle queues on curved roads consists of a PC (personal computer), three autonomous vehicles, seven ultra-wideband positioning modules and a router. The specifications of the autonomous vehicle are as follows: maximum speed 1m / s, maximum front wheel deflection angle 0.873rad, wheelbase 0.342m, tire diameter 0.083m, maximum speed 0.7m / s, and a microcomputer as a processor. The vehicle status information such as speed and heading angle is obtained through sensors such as odometer and IMU (Inertial Measurement Unit), and its steering system adopts Ackerman steering structure. UWB (Ultra Wide Band) positioning modules are used for position detection of autonomous vehicle queues, of which four are installed as base stations at the four corners of the experimental site, and three are installed as tags at the center of the rear axle of the vehicle; the router is used to form a local area network. Using local area network communication, signal transmission between the PC and the vehicle is realized through ROS (Robot Operating System). Specifically, the ROS master (manager of the robot operating system) is started on the PC side, and then each ROS node (node ​​of the robot operating system) is started on the vehicle's microcomputer and PC side. The ROS master manages the communication of each ROS node. In the local area network, the vehicle node transmits the status information to the PC node. After the PC node obtains the vehicle status information, it uses the proposed algorithm to generate a control signal, and then transmits it to each vehicle node, thereby realizing the control of the vehicle. Figure 6 The experimental site shown in the figure is 15 meters long and 14 meters wide, with a flat road surface, which can provide good support for the experiment.

[0135] Figure 7 The diagram schematically shows a trajectory diagram of a mixed vehicle queue traveling on a curved road according to an embodiment of the present disclosure.

[0136] Figure 8 A lateral offset error diagram of a mixed vehicle platoon traveling on a curved road according to an embodiment of the present disclosure is schematically shown.

[0137] from Figure 7 and Figure 8 It can be concluded that following car 1 (also called autonomous driving car 1) and following car 2 (also called autonomous driving car 2) can track the curved path position of the pilot car very well, and the pilot car, following car 1 and following car 2 have a small lateral position deviation, which can effectively solve the corner cutting problem.

[0138] Fig. 9 A heading angle diagram of a mixed vehicle platoon traveling on a curved road according to an embodiment of the present disclosure is schematically shown.

[0139] Fig.10 A diagram schematically shows a heading angle error diagram of a mixed vehicle platoon traveling on a curved road according to an embodiment of the present disclosure.

[0140] like Fig. 9 As shown, based on the control method of mixed vehicle queue on a curved road, the following vehicle 1 and the following vehicle 2 are controlled to perform automatic driving along with the automatic driving pilot vehicle. Fig. 9 In the middle are the heading angles of the leading vehicle, the following vehicle 1, and the following vehicle 2 on the curved road. The mixed vehicle platoon can track the heading of the curved path very well. Fig.10 In the figure, there are heading angle errors of the leading vehicle, following vehicle 1, and following vehicle 2 on the curved road. Following vehicle 1 and following vehicle 2 have smaller heading angle errors. Combining the above conclusions, it can be concluded that the control method of the mixed vehicle platoon on the curved road has good position and heading tracking performance on the curved path, reflecting good path tracking performance.

[0141] Fig.11 The predicted speed diagram of the pilot vehicle on a curved road according to an embodiment of the present disclosure is schematically shown.

[0142] Fig.12 A vehicle spacing diagram of a mixed vehicle platoon traveling on a curved road according to an embodiment of the present disclosure is schematically shown.

[0143] like Fig.11 As shown in FIG. 1 , based on the control method for mixed vehicle queues on curved roads, the actual speed of the leading vehicle in the mixed vehicle queue is basically consistent with the predicted speed, which can better estimate the future dynamics of the leading vehicle. Fig.12 As shown in the figure, the hybrid vehicle queue control method considering the future dynamic changes of the pilot vehicle on the curved road can continuously adjust the control amount to make the vehicle queue maintain the desired vehicle spacing, has good vehicle spacing maintenance performance, and can ensure the safety and operation efficiency of the vehicle queue.

[0144] Fig.13 A block diagram of a device for controlling a mixed vehicle queue on a curved road according to an embodiment of the present disclosure is schematically shown.

[0145] like Fig.13 As shown, the control device 1300 for a mixed vehicle queue on a curved road includes an extraction module 1310 , a first input module 1320 , a second input module 1330 , and a control module 1340 .

[0146] The extraction module 1310 is used to extract vehicle motion features from historical driving data of a pilot car in a mixed vehicle queue, where the mixed vehicle queue includes an autonomous driving vehicle and a pilot car driven by a driver. In one embodiment, the extraction module 1310 can be used to perform the operation S210 described above, which will not be described in detail herein.

[0147] The first input module 1320 is used to input the vehicle motion characteristics into the aggressive coefficient estimator to obtain the driver's aggressive coefficient, and the aggressive coefficient estimator is constructed based on the multi-feature decision analysis method. In one embodiment, the first input module 1320 can be used to perform the operation S220 described above, which will not be repeated here.

[0148] The second input module 1330 is used to input the road curvature sequence, the historical speed sequence and the aggressive coefficient into the vehicle speed prediction model to obtain the speed prediction sequence of the pilot vehicle, the road curvature sequence represents the road curvature information of the pilot vehicle during the driving time period, the historical speed sequence represents the speed information of the pilot vehicle during the driving time period, and the vehicle speed prediction model is composed of a neural network model and a constant acceleration prediction model, and the constant acceleration prediction model sets the acceleration of the pilot vehicle during driving to a constant value. In one embodiment, the second input module 1330 can be used to perform the operation S230 described above, which will not be repeated here.

[0149] The control module 1340 is used to control the automatic driving vehicle in the mixed vehicle queue according to the speed prediction sequence of the pilot vehicle. In one embodiment, the control module 1340 can be used to perform the operation S240 described above, which will not be repeated here.

[0150] According to the embodiments of the present invention, any one or more of the modules, submodules, units, and subunits, or at least part of the functions of any one of them can be implemented in one module. According to the embodiments of the present invention, any one or more of the modules, submodules, units, and subunits can be split into multiple modules for implementation. According to the embodiments of the present invention, any one or more of the modules, submodules, units, and subunits can be at least partially implemented as hardware circuits, such as field programmable gate arrays (FPGAs), programmable logic arrays (PLAs), systems on chips, systems on substrates, systems on packages, application specific integrated circuits (ASICs), or can be implemented by hardware or firmware in any other reasonable way of integrating or packaging circuits, or can be implemented in any one of the three implementation methods of software, hardware, and firmware, or in any appropriate combination of any of them. Alternatively, according to the embodiments of the present invention, one or more of the modules, submodules, units, and subunits can be at least partially implemented as computer program modules, and when the computer program modules are run, the corresponding functions can be performed.

[0151] For example, any of the extraction module 1310, the first input module 1320, the second input module 1330, and the control module 1340 can be combined in one module / unit / sub-unit for implementation, or any of the modules / units / sub-units can be split into multiple modules / units / sub-units. Alternatively, at least part of the functions of one or more of these modules / units / sub-units can be combined with at least part of the functions of other modules / units / sub-units and implemented in one module / unit / sub-unit. According to an embodiment of the present disclosure, at least one of the extraction module 1310, the first input module 1320, the second input module 1330, and the control module 1340 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or by any one of the three implementation methods of software, hardware, and firmware, or by a suitable combination of any of them. Alternatively, at least one of the extraction module 1310, the first input module 1320, the second input module 1330, and the control module 1340 may be at least partially implemented as a computer program module, and when the computer program module is executed, a corresponding function may be performed.

[0152] It should be noted that the control device part of the mixed vehicle queue on the curved road in the embodiment of the present disclosure corresponds to the control method part of the mixed vehicle queue on the curved road in the embodiment of the present disclosure. The description of the control device part of the mixed vehicle queue on the curved road specifically refers to the control method part of the mixed vehicle queue on the curved road, which will not be repeated here.

[0153] Fig.14 A block diagram of an electronic device suitable for implementing a control method for a mixed vehicle queue on a curved road according to an embodiment of the present disclosure is schematically shown. Fig.14 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0154] like Fig.14As shown, the electronic device 1400 according to an embodiment of the present disclosure includes a processor 1401, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1402 or a program loaded from a storage part 1408 to a random access memory (RAM) 1403. The processor 1401 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a dedicated microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 1401 may also include an onboard memory for caching purposes. The processor 1401 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0155] In RAM 1403, various programs and data required for the operation of electronic device 1400 are stored. Processor 1401, ROM 1402 and RAM 1403 are connected to each other through bus 1404. Processor 1401 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 1402 and / or RAM 1403. It should be noted that the program can also be stored in one or more memories other than ROM 1402 and RAM 1403. Processor 1401 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in the one or more memories.

[0156] According to an embodiment of the present disclosure, the electronic device 1400 may further include an input / output (I / O) interface 1405, which is also connected to the bus 1404. The electronic device 1400 may further include one or more of the following components connected to the input / output (I / O) interface 1405: an input portion 1406 including a keyboard, a mouse, etc.; an output portion 1407 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 1408 including a hard disk, etc.; and a communication portion 1409 including a network interface card such as a LAN card, a modem, etc. The communication portion 1409 performs communication processing via a network such as the Internet. A drive 1410 is also connected to the input / output (I / O) interface 1405 as needed. A removable medium 1411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1410 as needed, so that a computer program read therefrom is installed into the storage portion 1408 as needed.

[0157] According to an embodiment of the present disclosure, the method flow according to an embodiment of the present disclosure can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable storage medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1409, and / or installed from the removable medium 1411. When the computer program is executed by the processor 1401, the above-mentioned functions defined in the system of the embodiment of the present disclosure are executed. According to an embodiment of the present disclosure, the system, equipment, device, module, unit, etc. described above can be implemented by a computer program module.

[0158] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.

[0159] According to an embodiment of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium. For example, it may include, but is not limited to: 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), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device.

[0160] For example, according to an embodiment of the present disclosure, the computer-readable storage medium may include the ROM 1402 and / or the RAM 1403 described above and / or one or more memories other than the ROM 1402 and the RAM 1403 .

[0161] An embodiment of the present disclosure also includes a computer program product, which includes a computer program, and the computer program contains a program code for executing the method provided by the embodiment of the present disclosure. When the computer program product is run on an electronic device, the program code is used to enable the electronic device to implement the control method for a mixed vehicle queue on a curved road provided by the embodiment of the present disclosure.

[0162] When the computer program is executed by the processor 1401, the above functions defined in the system / device of the embodiment of the present disclosure are executed. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0163] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 1409, and / or installed from the removable medium 1411. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0164] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0165] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions. It can be understood by those skilled in the art that the features recorded in the various embodiments of the present disclosure can be combined and / or combined in a variety of ways, even if such a combination or combination is not explicitly recorded in the present disclosure. In particular, without departing from the spirit and teaching of the present disclosure, the features described in the various embodiments of the present disclosure may be combined and / or combined in a variety of ways. All of these combinations and / or combinations fall within the scope of the present disclosure.

[0166] The embodiments of the present disclosure are described above. However, these embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure. Although the embodiments are described above, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. Without departing from the scope of the present disclosure, those skilled in the art may make a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. A method for controlling a mixed vehicle queue on a curved road, comprising: Extracting vehicle motion features from historical driving data of a pilot vehicle in a mixed vehicle train, the mixed vehicle train including the autonomous vehicle and the pilot vehicle driven by a human driver; Inputting the vehicle motion feature into an aggressive coefficient estimator to obtain the aggressive coefficient of the driver, wherein the aggressive coefficient estimator is constructed based on a multi-feature decision analysis method; Inputting a road curvature sequence, a historical speed sequence and the aggressive coefficient into a vehicle speed prediction model to obtain a speed prediction sequence of the pilot vehicle, wherein the road curvature sequence represents the road curvature information of the pilot vehicle during a driving period, and the historical speed sequence represents the speed information of the pilot vehicle during the driving period. The vehicle speed prediction model is composed of a neural network model and a constant acceleration prediction model, wherein the constant acceleration prediction model sets the acceleration of the pilot vehicle during driving to a constant value; The automated driving vehicle in the mixed vehicle queue is controlled according to the speed prediction sequence of the pilot vehicle.

2. The method according to claim 1, characterized in that The step of inputting the road curvature sequence, the historical speed sequence and the aggressive coefficient into the vehicle speed prediction model to obtain the speed prediction sequence of the pilot vehicle comprises: Inputting the road curvature sequence, the historical speed sequence and the aggressive coefficient into the neural network model in the vehicle speed prediction model to obtain the speed compensation prediction sequence of the pilot vehicle; Inputting the historical speed sequence into the constant acceleration prediction model of the vehicle speed prediction model to obtain the theoretical speed prediction sequence of the pilot vehicle; A speed prediction sequence of the pilot vehicle is obtained according to the speed compensation prediction sequence and the theoretical speed prediction sequence.

3. The method according to claim 2, characterized in that The step of inputting the road curvature sequence, the historical speed sequence and the aggressive coefficient into the neural network model in the vehicle speed prediction model to obtain the speed compensation prediction sequence of the pilot vehicle comprises: Extracting time features from the road curvature sequence, the historical speed sequence and the aggressiveness coefficient to obtain a time information sequence; Performing residual calculation on the historical speed sequence to obtain a residual speed sequence; A speed compensation prediction sequence of the pilot vehicle is obtained according to the residual speed sequence and the time information sequence.

4. The method according to claim 3, characterized in that The extracting time features of the road curvature sequence, the historical speed sequence and the aggressiveness coefficient to obtain the time information sequence comprises: splicing the road curvature sequence, the historical speed sequence and the aggressiveness coefficient to obtain a spliced ​​feature vector with an increased number of columns; Using a multi-layer perceptron of the neural network model to perform feature learning on the concatenated feature vector to obtain an initial feature vector; The initial feature vector and the road curvature feature vector are stacked to obtain a target feature vector with an increased number of rows; The target feature vector is input into the time decoder of the neural network model to obtain the time information sequence.

5. The method according to claim 4, characterized in that The concatenating the road curvature sequence, the historical speed sequence and the aggressive coefficient to obtain a concatenated feature vector with an increased number of columns further includes: Performing feature projection on the road curvature sequence to obtain a road curvature feature vector; The historical speed sequence, the aggressiveness coefficient and the road curvature feature vector are concatenated to obtain a concatenated feature vector with an increased number of columns.

6. The method according to claim 1, characterized in that The step of inputting the vehicle motion characteristics into an aggressive coefficient estimator to obtain the aggressive coefficient of the driver includes: Calculating the basic feature weights of the vehicle motion features using an entropy weight method, and determining adaptive weights based on the basic feature weights; Calculating the distance between the vehicle motion feature and the maximum value of the feature and the distance between the vehicle motion feature and the minimum value of the feature according to the adaptive weight; The aggressiveness coefficient of the driver is determined according to the distance between the vehicle motion feature and a feature maximum value and the distance between the vehicle motion feature and a feature minimum value.

7. The method according to claim 6, characterized in that The vehicle motion characteristics include average speed, maximum speed, average longitudinal acceleration, maximum longitudinal acceleration, average lateral acceleration, and maximum lateral acceleration. The formula of the driver's aggressiveness coefficient is as follows: is the radical coefficient, represents the distance between the motion feature of the i-th vehicle and the maximum value of the feature, Represents the distance between the i-th vehicle motion feature and the minimum feature value, where i is an integer greater than or equal to 1.

8. The method according to claim 1, characterized in that The controlling the automatic driving vehicle in the mixed vehicle queue according to the speed prediction sequence of the pilot vehicle comprises: According to the speed prediction sequence of the pilot vehicle, the autonomous driving vehicles in the mixed vehicle queue are controlled by a distributed model predictive control algorithm, wherein the distributed model predictive control algorithm includes a local model predictive control algorithm, and each local model predictive control algorithm controls one of the autonomous driving vehicles.

9. The method according to claim 8, characterized in that The local model predictive control algorithm is composed of a vehicle kinematic model, a path coupling extended look-ahead strategy and a local optimal controller.

10. The method according to claim 1, characterized in that The method further comprises: According to the driver's aggressiveness coefficient, the expected vehicle spacing between vehicles in the mixed vehicle queue is determined, and the formula for the expected vehicle spacing is: In the formula, is the expected vehicle spacing of vehicle i in the mixed vehicle queue, is the basic vehicle spacing, is the heading angle of the vehicle i.