Reference lane centerline generation method, vehicle cruise control method and application

Abstract

This application discloses a reference lane centerline generation method, a vehicle cruise control method, and an application thereof. The method comprises obtaining an initial motion state quantity of a preceding vehicle; performing motion state prediction based on the initial motion state quantity of the preceding vehicle to obtain a predicted current motion state quantity of the preceding vehicle; obtaining an observed current motion state quantity of the preceding vehicle to correct the predicted current motion state quantity of the preceding vehicle to obtain a corrected current motion state quantity of the preceding vehicle; and fitting a reference lane centerline based on the corrected current motion state quantities of the preceding vehicle and the vehicle's trajectory at multiple moments. In this way, when the lane centerline is lost, the fitted reference lane centerline can be used as the basis for adaptive cruise control and lane keeping control.

Description

Technical Field

[0001] The present application belongs to the field of computer data processing technology, and specifically relates to a reference lane centerline generation method, a vehicle cruise control method and an application. Background Art

[0002] Adaptive cruise control and lane keeping control primarily use lane centerlines and lane centerlines as input. Adaptive cruise control needs to determine whether the vehicle ahead is in the same lane as the vehicle, while the trajectory tracked by lane keeping control vehicles is the lane centerline. However, the lane centerline may not be accurately known in many scenarios, for example: ① The lane centerline is obscured in rainy and snowy weather; ② The on-board camera cannot properly detect the lane centerline due to strong light reflection; ③ In traffic jams, the lane centerlines on both sides are temporarily occupied by vehicles in other lanes. In these scenarios, the lane centerline detection module on the vehicle fails, and the vehicle can no longer use the adaptive cruise control function and lane centering function.

[0003] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the application and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to a person skilled in the art. Summary of the Invention

[0004] The purpose of this application is to provide a reference lane centerline generation method, which is used to solve the problem that the vehicle cannot use adaptive cruise control and lane centering functions when the lane centerline is lost.

[0005] To achieve the above objectives, the present application provides a method for generating a reference lane centerline, the method comprising:

[0006] Obtain the initial motion state of the front vehicle;

[0007] Performing motion state prediction based on the initial motion state quantity of the preceding vehicle to obtain a current motion state prediction quantity of the preceding vehicle;

[0008] Obtaining an observed value of the current motion state of the preceding vehicle to correct the predicted value of the current motion state of the preceding vehicle, thereby obtaining a corrected value of the current motion state of the preceding vehicle;

[0009] Based on the current motion state correction amount of the preceding vehicle and the own vehicle trajectory at multiple moments, a reference lane centerline is generated by fitting.

[0010] In one embodiment, the correction value of the current motion state of the preceding vehicle includes the trajectory of the preceding vehicle;

[0011] Based on the current motion state correction of the preceding vehicle and the own vehicle trajectory at multiple moments, a reference lane centerline is generated by fitting, specifically including:

[0012] The preceding vehicle trajectory and the own vehicle trajectory at multiple moments are fitted based on a cubic spline curve to generate a reference lane centerline, wherein a regularization constraint is imposed on the quadratic term and / or the cubic term of the cubic spline curve.

[0013] In one embodiment, fitting the preceding vehicle trajectory and the host vehicle trajectory at multiple moments based on a cubic spline curve to generate a reference lane centerline specifically includes:

[0014] Establishing a loss function of the sum of squares of deviations of the preceding vehicle trajectory and the own vehicle trajectory relative to the cubic spline curve;

[0015] Under the regularization constraint, the coefficient matrix of the cubic spline curve is solved in a direction that minimizes the loss function to generate a reference lane centerline.

[0016] In one embodiment, the correction value of the current motion state of the preceding vehicle includes the relative position and relative speed of the preceding vehicle relative to the host vehicle;

[0017] The method further comprises:

[0018] Based on the current positioning information and heading angle information of the vehicle, the relative position is converted into a global position, and the relative speed is converted into a global speed.

[0019] In one embodiment, the method specifically includes:

[0020] The current motion state correction amount of the preceding vehicle at time t is used as the initial motion state amount of the preceding vehicle at time t+1 to obtain the current motion state correction amount of the preceding vehicle at time t+1.

[0021] In one embodiment, performing motion state prediction based on the initial motion state of the preceding vehicle to obtain the current motion state prediction of the preceding vehicle specifically includes:

[0022] Performing motion state prediction on the initial motion state of the preceding vehicle using the first state transfer matrix and the state control matrix to obtain a current motion state prediction of the preceding vehicle;

[0023] Obtaining the observed value of the current motion state of the preceding vehicle to correct the predicted value of the current motion state of the preceding vehicle, and obtaining the corrected value of the current motion state of the preceding vehicle, specifically includes:

[0024] Predicting the initial state covariance matrix of the preceding vehicle using the first state transfer matrix and the process excitation noise covariance matrix to obtain a current covariance prediction matrix of the preceding vehicle;

[0025] Calculating a current Kalman gain based on the preceding vehicle's current covariance prediction matrix, the second state transfer matrix, and the measurement noise matrix;

[0026] Based on the current Kalman gain and the error between the current motion state observation value of the preceding vehicle and the current motion state prediction value of the preceding vehicle, the current motion state prediction value of the preceding vehicle is corrected to obtain the current motion state correction value of the preceding vehicle.

[0027] The present application also provides a vehicle cruise control method, the method comprising:

[0028] When the lane centerline is detected to be missing, a reference lane centerline is generated using the method described above;

[0029] Cruise control is performed based on the reference lane centerline.

[0030] The present application also provides an electronic device, comprising:

[0031] at least one processor; and

[0032] A memory storing instructions, which, when executed by the at least one processor, causes the at least one processor to execute the reference lane centerline generating method or the vehicle cruise control method as described above.

[0033] The present application also provides a machine-readable storage medium storing executable instructions, which, when executed, enable the machine to execute the reference lane centerline generation method or vehicle cruise control method as described above.

[0034] Compared with the existing technology, the reference lane centerline generation method according to the present application uses the initial motion state of the preceding vehicle to predict the motion state, and corrects the prediction result based on the current motion state observation of the preceding vehicle, thereby obtaining a reliable motion state correction amount; then, based on the current motion state correction amount of the preceding vehicle at multiple moments and the trajectory of the own vehicle, the reference lane centerline is fitted and generated without relying on the detection of the "real" lane centerline.

[0035] On the other hand, a cubic spline curve is used to fit the reference lane centerline, and regularization constraints can be applied to the quadratic and cubic terms of the cubic spline curve, so that the curvature of the fitted reference lane centerline at each moment changes less. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is an application scenario diagram of a method for generating a lane centerline according to an embodiment of the present application;

[0037] Figure 2 is a flowchart of a method for generating a reference lane centerline according to an embodiment of the present application;

[0038] Figure 3 This is an application scenario diagram of a vehicle cruise control method according to an embodiment of the present application;

[0039] Figure 4 This is a module diagram of a reference lane centerline generating device according to an embodiment of the present application;

[0040] Figure 5 is a module diagram of a vehicle cruise control device according to an embodiment of the present application;

[0041] Figure 6 1 is a hardware structure diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0042] The present application will be described in detail below with reference to the various embodiments shown in the accompanying drawings. However, these embodiments do not limit the present application, and any structural, methodological, or functional modifications made by a person skilled in the art based on these embodiments are included within the scope of protection of the present application.

[0043] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the numbers used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "corresponding to" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0044] Before introducing the embodiments of the present application, the basic technologies and some technical terms involved in the embodiments of the present application are schematically explained:

[0045] Intelligent Traffic System (ITS), also known as Intelligent Transportation System, effectively integrates advanced science and technology (information technology, computer technology, data communication technology, sensor technology, electronic control technology, automatic control theory, operations research and artificial intelligence, etc.) into transportation, service control and vehicle manufacturing, strengthens the connection between vehicles, roads and users, and thus forms a comprehensive transportation system that ensures safety, improves efficiency, improves the environment and saves energy.

[0046] Intelligent Vehicle Infrastructure Cooperative Systems (IVICS), also known as VICS, are a development direction for intelligent transportation systems. They utilize advanced wireless communications and next-generation internet technologies to implement dynamic, real-time information exchange between vehicles and between vehicles and roads. They also utilize the collection and integration of dynamic traffic information across time and space to implement active vehicle safety control and collaborative road management. This approach fully realizes effective coordination between people, vehicles, and roads, ensuring traffic safety and improving traffic efficiency, resulting in a safe, efficient, and environmentally friendly road transportation system.

[0047] ACC (Adaptive Cruise Control) is an automated driving system that dynamically adjusts the vehicle's speed based on the user-set cruising speed and the safe distance between the vehicle ahead and the driver. If the vehicle ahead accelerates, the driver's speed will accelerate to the set speed. If the vehicle ahead decelerates, the driver's speed will decrease to maintain a safe distance between them.

[0048] Ginseng Figure 1 , taking an application scenario of the reference lane centerline generation method provided in an embodiment of the present application as an example. The user can drive the vehicle manually or use the vehicle's intelligent driving system for autonomous driving. Whether in manual driving or autonomous driving, the terminal can collect environmental information based on sensors (e.g., belonging to the perception fusion layer) and provide some decision-making information for the vehicle cruise control.

[0049] For example, sensors may include visual sensors, long-range radar (forward), and short-range radar (corner radar) to provide multi-directional information perception in front, behind, left, and right of the vehicle. The visual sensor may use a 360-degree panoramic camera to capture image information around the vehicle, while the long-range radar and short-range radar can be used to collect information such as the distance to the obstacle in front and the movement of the obstacle in front.

[0050] The terminal implements cruise control of the vehicle when the lane centerline is lost by running the reference lane centerline generation method provided in various embodiments of the present application. The terminal can be a vehicle driven by a user, or an intelligent vehicle-mounted device / module on the vehicle, or a desktop computer, laptop, smartphone, or tablet computer installed in the vehicle while the user is driving, or a portable wearable device carried by the user.

[0051] Alternatively, the terminal can also communicate with the server through the network, and run the vehicle cruise control method of the present application with the assistance of the server or by the server alone. The server can be an independent physical server or a service node in a blockchain system, in which each service node in the blockchain system forms a peer-to-peer (P2P) network, and the P2P protocol is an application layer protocol running on the Transmission Control Protocol (TCP) protocol. In addition, the server can also be a server cluster composed of multiple physical servers, which can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0052] The terminal and the server may be connected via Bluetooth, USB (Universal Serial Bus) or a network, and this application does not impose any restrictions on this.

[0053] Ginseng Figure 2 , an embodiment of the method for generating a reference lane centerline of the present application is introduced. In this embodiment, the method includes:

[0054] S11. Obtain the initial motion state of the preceding vehicle.

[0055] When the vehicle receives a command input by the user, starts the vehicle, the lane center line is lost, or a specific program is run, etc., the vehicle controls the sensors to turn on and detect the vehicle, the road on which the vehicle is located, and other vehicles around the vehicle, thereby obtaining the initial motion state of the preceding vehicle.

[0056] The sensor may be one or more of a camera, an ultrasonic radar, a lidar, a millimeter-wave radar, a global navigation satellite system (GNSS), an inertial navigation system (INS), and the like.

[0057] Sensors can be installed on the front, doors, rear, roof, and interior of a vehicle. By detecting the vehicle, the road it is on, and other vehicles around it, the sensors acquire information such as the vehicle's lateral speed, longitudinal speed, and yaw angle. They also provide information such as the presence of a lane centerline, lane centerline structure, and curbs on the road it is on. They also provide information such as the lateral speed, longitudinal speed, global position, relative position, brake lights, left and right turn signals, instantaneous angular velocity, yaw angle, and heading angle of each vehicle around the vehicle.

[0058] In this embodiment, the obtained initial motion state of the preceding vehicle may include the relative position and relative speed of the preceding vehicle (relative to the vehicle itself). Specifically, the initial motion state of the preceding vehicle may be obtained by an ultrasonic radar or a camera.

[0059] For example, a millimeter-wave radar typically includes a transmitting unit, a receiving unit, and a processing unit. The processing unit can calculate information such as the distance between the preceding vehicle and the ultrasonic radar, the preceding vehicle's shape, the preceding vehicle's relative speed relative to the host vehicle, and the preceding vehicle's lateral and longitudinal speeds relative to the host vehicle based on the time difference between the preceding vehicle's transmission and the reception of the ultrasonic signal by the receiving unit. A camera calculates information such as the preceding vehicle's relative speed, lateral and longitudinal speeds, relative position, instantaneous angular velocity, and yaw angle relative to the host vehicle based on the preceding vehicle's position changes across multiple consecutive frames captured by the camera.

[0060] In some embodiments, it may be desirable to perform the following motion state prediction based on the initial motion state of the preceding vehicle at the current moment. However, due to system settings or human preferences, the vehicle has not yet acquired or stored the motion state before the current moment. In such cases, the initial motion state of the preceding vehicle can be assigned a value through random initialization, for example, setting:

[0061] X=[1,1,0,0]'

[0062] Among them, X represents the initial motion state of the leading vehicle, the first two terms (1,1) in the matrix are the initial positions of the leading vehicle, and the last two terms (0,0) are the initial velocities of the leading vehicle.

[0063] S12. Perform motion state prediction based on the initial motion state quantity of the preceding vehicle to obtain a predicted current motion state quantity of the preceding vehicle.

[0064] Assuming that the above-mentioned initial motion state of the preceding vehicle is an optimal estimate of the actual initial motion state of the preceding vehicle, its state covariance matrix can be expressed as:

[0065] S = state_uncertain*eye(4)

[0066] With p representing position and v representing velocity, the state covariance matrix includes Σ pp ,Σ pv ,Σ vp ,Σ vv The uncertainty between the four states. Similarly, the state covariance matrix can also be randomly initialized.

[0067] Taking into account the impact of external factors on the vehicle, we can also assume that the system error caused by this impact obeys a Gaussian distribution with a mean of 0. Then the process excitation noise covariance can be initialized as:

[0068] Q=[T^2 / 4,0,T^3 / 2,0;0,T^2 / 4,0,T^3 / 2;T^3 / 2,0,T^2,0;0,T^3 / 2,0,T^2].

[0069] Based on the above settings, the first state transfer matrix and the state control matrix are used to predict the initial motion state of the preceding vehicle to obtain the current motion state prediction of the preceding vehicle, which is expressed as:

[0070] X(t) = A*X(t-1) + B*u(t) + Gaussian distribution mean (set to 0)

[0071] Where A is the first state transfer matrix, B is the state control matrix, and u(t) is the state control vector. The state control matrix B represents how acceleration and deceleration change the vehicle's state, and the state control vector u(t) represents the magnitude and direction of the control force. Substituting the initial motion state X of the preceding vehicle into X(t-1) yields a predicted current motion state of the preceding vehicle.

[0072] In this embodiment, in order to use the current motion state observation of the preceding vehicle to correct the current motion state prediction of the preceding vehicle, the first state transfer matrix and the process excitation noise covariance matrix can also be used in the above prediction process to predict the initial state covariance matrix of the preceding vehicle to obtain the current covariance prediction matrix of the preceding vehicle. The current covariance prediction matrix of the preceding vehicle is expressed as:

[0073] S(t)=A*S(t-1)*A'+Q.

[0074] Substituting the state covariance matrix S of the initial state of the preceding vehicle into S(t-1), the current covariance prediction matrix of the preceding vehicle can be obtained.

[0075] S13. Obtain an observed value of the current motion state of the preceding vehicle to correct the predicted value of the current motion state of the preceding vehicle, thereby obtaining a corrected value of the current motion state of the preceding vehicle.

[0076] In this embodiment, the current motion state observation of the preceding vehicle is used to correct the current motion state prediction of the preceding vehicle in order to fuse the results of the prediction and observation and make full use of the uncertainty of both to obtain a more accurate estimate.

[0077] Specifically, the current Kalman gain can be calculated based on the current covariance prediction matrix of the preceding vehicle, the second state transfer matrix, and the measurement noise matrix, and then based on the current Kalman gain and the error between the current motion state observation quantity of the preceding vehicle and the current motion state prediction quantity of the preceding vehicle, the current motion state prediction quantity of the preceding vehicle can be corrected to obtain the current motion state correction quantity of the preceding vehicle.

[0078] After obtaining the predicted value of the current motion state of the preceding vehicle, it is assumed that there should be a specific relationship between it and the actual motion state of the preceding vehicle. Similarly, this relationship can be described here by the second state transfer matrix H. The second state transfer matrix H can be set as:

[0079] H=[1,0,0,0;0,1,0,0]'

[0080] At the same time, for the observation of the current motion state of the preceding vehicle, due to the interference of noise, there must be a difference between it and the actual motion state of the preceding vehicle. It can be assumed that the observation noise is a Gaussian distribution with a mean of 0 and a covariance matrix (i.e., measurement noise matrix) of R, that is,

[0081] R=meas_uncertain*[1,0;0,1].

[0082] Correspondingly, the Kalman gain can be expressed as:

[0083] K = S*H'*inv(H*S*H'+R).

[0084] The correction value of the current motion state of the preceding vehicle can be expressed as:

[0085] X'=X+K*(input-H*X)

[0086] Here, input represents the observed current motion state of the preceding vehicle, and input-H*X represents the error between the observed current motion state of the preceding vehicle and the predicted current motion state of the preceding vehicle. The Kalman gain can be considered a correction weight. The entire correction process aims to eliminate various noise interferences on the actual motion state of the preceding vehicle, which is a filtering process.

[0087] Similarly, the current covariance prediction matrix of the preceding vehicle can be updated based on the Kalman gain, which can be expressed as:

[0088] S'=(eye(4)-K*H)*S.

[0089] It can be seen that the above filtering process can be continuously iterated. The state X and state covariance are first predicted based on the filtering result X at the previous moment and the input input. The state and covariance are then corrected and updated based on the input. After multiple iterative corrections and updates, a relatively accurate motion state of the preceding vehicle can be obtained.

[0090] During the above iterations, the system variables that require attention are the initial motion state of the preceding vehicle, the current covariance prediction matrix of the preceding vehicle, and the Kalman gain. It is understood that during the iterations, the current motion state correction of the preceding vehicle at time t is used as the initial motion state of the preceding vehicle at time t+1 to obtain the current motion state correction of the preceding vehicle at time t+1; and the current covariance prediction matrix of the preceding vehicle at time t is used as the initial state covariance matrix of the preceding vehicle at time t+1 to obtain the current covariance prediction matrix of the preceding vehicle at time t+1.

[0091] S14. Based on the current motion state correction value of the preceding vehicle at multiple moments and the own vehicle trajectory, a reference lane centerline is generated by fitting.

[0092] The data time required for vehicle fitting can be set based on the vehicle system's pre-set settings or the user's style settings. For example, the initial motion state of the preceding vehicle can be continuously acquired over a period of 5 seconds, and then iterated continuously (e.g., at a frequency of 20 Hz) during this period to obtain corrections to the preceding vehicle's current motion state at multiple moments.

[0093] Because the accumulated correction for the preceding vehicle's current motion state includes the preceding vehicle's relative speed and position relative to the host vehicle, and the host vehicle's position and speed may also be constantly changing within the host vehicle's coordinate system (x=0, y=0), in this embodiment, the relative position in the preceding vehicle's current motion state correction can be converted to a global position, and the relative speed can be converted to a global speed, based on the host vehicle's current positioning information and heading angle information at the corresponding moment.

[0094] In this embodiment, the global position of the preceding vehicle's current motion state correction can be used as the preceding vehicle's trajectory. Similarly, the host vehicle's trajectory can also refer to the host vehicle's global position. Furthermore, in various embodiments, the host vehicle's trajectory can be directly obtained based on the aforementioned sensors installed on the vehicle, or the host vehicle's trajectory at each moment can also be corrected using the filtering process described above.

[0095] Exemplarily, the vehicle's positioning information and heading angle information can be obtained based on global navigation positioning data. A global satellite navigation system is an airborne radio navigation positioning system that provides users with all-weather three-dimensional coordinates, velocity, and time information at any location on the Earth's surface or in near-Earth space. For example, the navigation positioning data may include a timestamp, valid longitude and latitude information, and heading angle information.

[0096] In this embodiment, the number of discrete position points included in the preceding vehicle trajectory and the own vehicle trajectory at multiple moments selected for fitting can be the same or different. Considering that too many own vehicle trajectory points may affect subsequent fitting and cause the generated reference lane centerline to shift, in this embodiment, the number of trajectory points in the own vehicle trajectory used for fitting can be set to be less than the number of trajectory points in the preceding vehicle trajectory.

[0097] Specifically, the preceding vehicle trajectory and the own vehicle trajectory at multiple moments may be fitted based on a cubic spline curve to generate a reference lane centerline, wherein the quadratic term and / or the cubic term of the cubic spline curve is subject to regularization constraints.

[0098] It can be understood that for a given set of trajectory points, the interval formed by any two adjacent trajectory points can be approximated by a low-degree polynomial, and the curve at the connection of the trajectory points can be kept smooth (derivative continuous).

[0099] If a set of trajectory points x1, x2, ..., x is given on the interval [a, b] i , and the corresponding function values y1, y2, ..., y i , if the function f(x i ,w) satisfies that on each subinterval of adjacent trajectory points, f(x i ,w) is a polynomial of no more than three degrees and is equal to the corresponding yi. At the same time, f(x i ,w) is a quadratic continuously differentiable function on the interval [a, b], then f(x i ,w) is defined as the cubic spline curve at this time.

[0100] In this embodiment, the cubic spline curve between two adjacent trajectory points can be expressed as:

[0101] f(x i ,w)=w0+w1x i +w2x i 2 +w3x i 3

[0102] It can be seen that the cubic spline curve between two adjacent trajectory points has four unknowns (w0, w1, w2, w3).

[0103] In this embodiment, a loss function is established for the sum of squared deviations of the preceding vehicle trajectory and the own vehicle trajectory relative to the cubic spline curve. Under the constraints of the above-mentioned regularization, the coefficient matrix of the cubic spline curve is solved in the direction of minimizing the loss function to generate the reference lane centerline.

[0104] The loss function can be expressed as:

[0105]

[0106] In this way, the partial derivative of the loss function can be set to 0, expressed as:

[0107]

[0108] Therefore, fitting the coefficients (w0, w1, w2, w3) of the cubic spline curve between adjacent trajectory points, that is, solving the corresponding equations are as follows:

[0109]

[0110] In the various embodiments of the present application, regularization constraints are imposed on the quadratic and / or cubic terms of the cubic spline curve, which is intended to minimize the curvature change of the fitted cubic spline curve at any time. Specifically, a "penalty" can be added to the above-mentioned loss function using, for example, L1 regularization or L2 regularization, and the partial derivative of the "penalty" can also be set to 0 for solution.

[0111] With ginseng Figure 3 , and also introduces an embodiment of the vehicle cruise control method of the present application.

[0112] In this embodiment, when the lane centerline is detected to be missing, a reference lane centerline is generated using the method of the embodiment of the present application as described above, and cruise control is performed based on the reference lane centerline.

[0113] The detection of "lane centerline missing" mentioned here may mean that the sensor is unable to obtain sensing information of the lane centerline, or that there is no marked lane centerline on the current road. Of course, in some embodiments, the reference lane centerline generation method provided in the above embodiments of this application is not limited to application in scenarios where the lane centerline is missing. For example, the reference lane centerline generated by this method can be coordinated with the vehicle's lane centerline detection module in real time, and the two can be integrated through a processor to provide another strategy for adaptive cruise control and lane keeping control.

[0114] Ginseng Figure 4, an embodiment of the device for generating a reference lane centerline of the present application is introduced. In this embodiment, the device for generating a reference lane centerline includes an acquisition module 201, a prediction module 202, an update module 203, and a generation module 204.

[0115] The acquisition module 201 is used to obtain the initial motion state quantity of the leading vehicle; the prediction module 202 is used to perform motion state prediction based on the initial motion state quantity of the leading vehicle to obtain the current motion state prediction quantity of the leading vehicle; the update module 203 is used to obtain the current motion state observation quantity of the leading vehicle to correct the current motion state prediction quantity of the leading vehicle to obtain the current motion state correction quantity of the leading vehicle; the generation module 204 is used to fit the current motion state correction quantity of the leading vehicle and the trajectory of the vehicle at multiple times to generate the reference lane center line.

[0116] In one embodiment, the correction value of the current motion state of the leading vehicle includes the trajectory of the leading vehicle; the generation module 204 is specifically used to fit the trajectory of the leading vehicle and the trajectory of the own vehicle at multiple times based on a cubic spline curve to generate a reference lane centerline, wherein the quadratic term and / or cubic term of the cubic spline curve is subject to regularization constraints.

[0117] In one embodiment, the generation module 204 is specifically used to establish a loss function of the sum of squared deviations of the preceding vehicle trajectory and the own vehicle trajectory relative to the cubic spline curve; under the regularization constraint, the coefficient matrix of the cubic spline curve is solved in a direction that minimizes the loss function to generate a reference lane centerline.

[0118] In one embodiment, the correction value of the current motion state of the leading vehicle includes the relative position and relative speed of the leading vehicle relative to the host vehicle; the generation module 204 is also used to convert the relative position into a global position and the relative speed into a global speed based on the current positioning information and heading angle information of the host vehicle.

[0119] In one embodiment, the generating module 204 is further configured to use the current motion state correction value of the preceding vehicle at time t as the initial motion state value of the preceding vehicle at time t+1 to obtain the current motion state correction value of the preceding vehicle at time t+1.

[0120] In one embodiment, the prediction module 202 is specifically used to use the first state transfer matrix and the state control matrix to predict the motion state of the initial motion state of the leading vehicle to obtain the current motion state prediction of the leading vehicle; the update module 203 is specifically used to use the first state transfer matrix and the process excitation noise covariance matrix to predict the initial state covariance matrix of the leading vehicle to obtain the current covariance prediction matrix of the leading vehicle; based on the current covariance prediction matrix of the leading vehicle, the second state transfer matrix, and the measurement noise matrix, the current Kalman gain is calculated; based on the current Kalman gain and the error between the current motion state observation of the leading vehicle and the current motion state prediction of the leading vehicle, the current motion state prediction of the leading vehicle is corrected to obtain the current motion state correction of the leading vehicle.

[0121] Ginseng Figure 5 This application also provides an embodiment of a vehicle cruise control device. In this embodiment, the vehicle cruise control device includes a lane centerline generation module 211, configured to generate a reference lane centerline using the aforementioned method when a lane centerline is detected to be missing; and a control module 212, configured to perform cruise control based on the reference lane centerline.

[0122] As above Figures 1 to 3 , a lane centerline generation method and a vehicle cruise control method according to embodiments of this specification are described. The details mentioned in the above description of the method embodiment also apply to the lane centerline generation device and vehicle cruise control device of the embodiments of this specification. The above lane centerline generation device and vehicle cruise control device can be implemented using hardware, software, or a combination of hardware and software.

[0123] Figure 6 FIG1 shows a hardware structure diagram of an electronic device according to an embodiment of this specification. Figure 6 As shown, the electronic device 30 may include at least one processor 31, a memory 32 (e.g., a non-volatile memory), a storage 33, and a communication interface 34, and the at least one processor 31, the storage 32, the storage 33, and the communication interface 34 are connected together via an internal bus 35. The at least one processor 31 executes at least one computer-readable instruction stored or encoded in the storage 32.

[0124] It should be understood that the computer executable instructions stored in the memory 32, when executed, cause at least one processor 31 to perform the above combined operations in various embodiments of this specification. Figures 1 to 3 Describes the various operations and functions.

[0125] In the embodiments of the present specification, the electronic device 30 may include but is not limited to: a personal computer, a server computer, a workstation, a desktop computer, a laptop computer, a notebook computer, a mobile electronic device, a smart phone, a tablet computer, a cellular phone, a personal digital assistant (PDA), a handheld device, a messaging device, a wearable electronic device, a consumer electronic device, and the like.

[0126] According to one embodiment, a program product such as a machine-readable medium is provided. The machine-readable medium may have instructions (i.e., the above-mentioned elements implemented in software form), which, when executed by a machine, causes the machine to perform the above-mentioned combined embodiments of the present specification. Figure 1-Figure 3 Specifically, a system or device equipped with a readable storage medium can be provided, on which software program codes for implementing the functions of any of the above-mentioned embodiments are stored, and a computer or processor of the system or device can be enabled to read and execute the instructions stored in the readable storage medium.

[0127] In this case, the program code itself read from the machine-readable medium can implement the functions of any one of the above embodiments, and thus the machine-readable code and the machine-readable storage medium storing the machine-readable code constitute part of this specification.

[0128] Examples of readable storage media include floppy disks, hard disks, magneto-optical disks, optical disks (e.g., CD-ROMs, CD-Rs, CD-RWs, DVD-ROMs, DVD-RAMs, DVD-RWs, DVD-RWs), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, the program code may be downloaded from a server computer or a cloud via a communication network.

[0129] Those skilled in the art will appreciate that the various embodiments disclosed above may be modified and altered in various ways without departing from the essence of the invention. Therefore, the scope of protection of this specification shall be defined by the appended claims.

[0130] It should be noted that not all steps and units in the above processes and system structure diagrams are required, and certain steps or units can be omitted according to actual needs. The execution order of each step is not fixed and can be determined as needed. The device structure described in the above embodiments can be a physical structure or a logical structure, that is, some units may be implemented by the same physical client, or some units may be implemented by multiple physical clients, or may be implemented by certain components in multiple independent devices.

[0131] In the above embodiments, hardware unit or module can be realized by mechanical means or electrical means. For example, a hardware unit, module or processor can include permanent dedicated circuit or logic (such as special processor, FPGA or ASIC) to complete the corresponding operation. Hardware unit or processor can also include programmable logic or circuit (such as general purpose processor or other programmable processor), can be temporarily set up to complete the corresponding operation by software. Concrete implementation (mechanical means or dedicated permanent circuit or temporary circuit) can be determined based on cost and time consideration.

[0132] The specific embodiments described above in conjunction with the accompanying drawings describe exemplary embodiments, but do not represent all embodiments that can be implemented or fall within the scope of protection of the claims. The term "exemplary" used throughout this specification means "used as an example, instance or illustration" and does not mean "preferred" or "having advantages" over other embodiments. For the purpose of providing an understanding of the described technology, the specific embodiments include specific details. However, these technologies can be implemented without these specific details. In some instances, in order to avoid obscuring the concepts of the described embodiments, well-known structures and devices are shown in block diagram form.

[0133] The foregoing description of the present disclosure is provided to enable any person skilled in the art to implement or use the present disclosure. Various modifications to the present disclosure will be readily apparent to those skilled in the art, and the general principles herein may be applied to other variations without departing from the scope of the present disclosure. Therefore, the present disclosure is not limited to the examples and designs described herein, but is intended to be consistent with the widest range of principles and novel features disclosed herein.

Claims

1. A method for generating a reference lane centerline, characterized in that: The method comprises: Acquiring an initial motion state quantity of the preceding vehicle, wherein the initial motion state quantity of the preceding vehicle includes a relative position and relative speed of the preceding vehicle relative to the host vehicle; Performing motion state prediction based on the initial motion state of the preceding vehicle to obtain a predicted current motion state of the preceding vehicle, specifically comprising: performing motion state prediction on the initial motion state of the preceding vehicle using a first state transfer matrix and a state control matrix to obtain a predicted current motion state of the preceding vehicle; Obtaining the current motion state observation quantity of the preceding vehicle to correct the current motion state prediction quantity of the preceding vehicle and obtaining the current motion state correction quantity of the preceding vehicle, specifically comprising: using the first state transfer matrix and the process excitation noise covariance matrix to predict the initial state covariance matrix of the preceding vehicle to obtain the current covariance prediction matrix of the preceding vehicle; calculating the current Kalman gain based on the current covariance prediction matrix of the preceding vehicle, the second state transfer matrix, and the measurement noise matrix; correcting the current motion state prediction quantity of the preceding vehicle based on the current Kalman gain and the error between the current motion state observation quantity of the preceding vehicle and the current motion state prediction quantity of the preceding vehicle to obtain the current motion state correction quantity of the preceding vehicle; Based on the current motion state correction of the preceding vehicle and the own vehicle trajectory at multiple moments, a reference lane centerline is generated by fitting.

2. The method for generating a reference lane centerline according to claim 1, wherein: The correction value of the current motion state of the preceding vehicle includes the trajectory of the preceding vehicle; Based on the current motion state correction of the preceding vehicle and the own vehicle trajectory at multiple moments, a reference lane centerline is generated by fitting, specifically including: The preceding vehicle trajectory and the own vehicle trajectory at multiple moments are fitted based on a cubic spline curve to generate a reference lane centerline, wherein a regularization constraint is imposed on the quadratic term and / or the cubic term of the cubic spline curve.

3. The method for generating a reference lane centerline according to claim 2, wherein: Fitting the preceding vehicle trajectory and the host vehicle trajectory at multiple moments based on a cubic spline curve to generate a reference lane centerline specifically includes: Establishing a loss function of the sum of squares of deviations of the preceding vehicle trajectory and the own vehicle trajectory relative to the cubic spline curve; Under the regularization constraint, the coefficient matrix of the cubic spline curve is solved in a direction that minimizes the loss function to generate a reference lane centerline.

4. The method for generating a reference lane centerline according to claim 1, wherein: The correction value of the current motion state of the preceding vehicle includes the relative position and relative speed of the preceding vehicle relative to the vehicle; The method further comprises: Based on the current positioning information of the vehicle, the relative position is converted into a global position, and the relative speed is converted into a global speed.

5. The method for generating a reference lane centerline according to claim 1, wherein: The method specifically includes: The current motion state correction amount of the preceding vehicle at time t is used as the initial motion state amount of the preceding vehicle at time t+1 to obtain the current motion state correction amount of the preceding vehicle at time t+1.

6. A vehicle cruise control method, characterized in that: The method comprises: When the lane centerline is detected to be missing, generating a reference lane centerline using the method according to any one of claims 1 to 5; Cruise control is performed based on the reference lane centerline.

7. A device for generating a reference lane centerline, characterized in that: include: An acquisition module is used to obtain the initial motion state of the preceding vehicle; A prediction module is configured to perform motion state prediction based on the initial motion state quantity of the preceding vehicle to obtain a predicted value of the current motion state of the preceding vehicle, specifically comprising: performing motion state prediction on the initial motion state quantity of the preceding vehicle using a first state transfer matrix and a state control matrix to obtain a predicted value of the current motion state of the preceding vehicle; An updating module is used to obtain the current motion state observation of the preceding vehicle to correct the current motion state prediction of the preceding vehicle and obtain the current motion state correction of the preceding vehicle, specifically comprising: using the first state transfer matrix and the process excitation noise covariance matrix to predict the initial state covariance matrix of the preceding vehicle to obtain the current covariance prediction matrix of the preceding vehicle; calculating the current Kalman gain based on the current covariance prediction matrix of the preceding vehicle, the second state transfer matrix, and the measurement noise matrix; correcting the current motion state prediction of the preceding vehicle based on the current Kalman gain and the error between the current motion state observation of the preceding vehicle and the current motion state prediction of the preceding vehicle to obtain the current motion state correction of the preceding vehicle; The generation module is used to fit the current motion state correction of the preceding vehicle and the own vehicle trajectory at multiple moments to generate a reference lane centerline.

8. A vehicle cruise control device, characterized in that: include: A lane centerline generation module, configured to generate a reference lane centerline using the method according to any one of claims 1 to 5 when a lane centerline is detected to be missing; A control module is configured to perform cruise control based on the reference lane centerline.

9. A machine-readable storage medium storing executable instructions, which, when executed, enable the machine to execute the reference lane centerline generation method according to any one of claims 1 to 5 or the vehicle cruise control method according to claim 6.

Images