Vehicle lane changing trajectory prediction method, device and equipment and readable storage medium

Abstract

The invention relates to the technical field of automatic driving, and relates to a vehicle lane changing track prediction method, device and equipment and a readable storage medium, and the method comprises the steps: obtaining a discrete data set which comprises the motion track points of at least two vehicles; processing the discrete data set by using an inflection point detection algorithm to obtain an inflection point set; dividing a discrete data set based on the inflection point set, and training the trajectory prediction model by using the divided discrete data set to obtain a trained trajectory prediction model; and according to the trained trajectory prediction model, predicting the lane changing trajectory of the target vehicle to obtain a prediction result, and the target vehicle is the vehicle needing lane changing currently, the invention provides a recognition method for judging the starting point and the ending point of the lane changing process by using the curvature of the discrete point set. Compared with the prior art in which a fixed course angle is adopted for identification, the method has a higher identification rate, so that the division effect of the lane changing stage is better, and the prediction accuracy of the trajectory prediction model is improved.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a vehicle lane change trajectory prediction method, device, equipment and readable storage medium. Background Art

[0002] When dividing the lane changing process into stages, the existing technology generally adopts a threshold to determine the starting and ending points of the vehicle's lane changing behavior stage, such as the heading angle, without considering the characteristics of the entire trajectory. It ignores that different stage division methods will lead to differences in intention recognition results, resulting in inaccuracy in stage division and reducing the accuracy of trajectory prediction. Summary of the invention

[0003] The object of the present invention is to provide a vehicle lane change trajectory prediction method, device, equipment and readable storage medium to improve the above-mentioned problems.

[0004] In order to achieve the above objectives, the present application provides the following technical solutions:

[0005] On the one hand, an embodiment of the present application provides a vehicle lane change trajectory prediction method, the method comprising:

[0006] Acquire a discrete data set, wherein the discrete data set includes motion trajectory points of at least two vehicles;

[0007] Processing the discrete data set using an inflection point detection algorithm to obtain an inflection point set;

[0008] Dividing the discrete data set based on the inflection point set, and training the trajectory prediction model using the divided discrete data set to obtain a trained trajectory prediction model;

[0009] The lane-changing trajectory of the target vehicle is predicted according to the trained trajectory prediction model to obtain a prediction result.

[0010] In a second aspect, an embodiment of the present application provides a vehicle lane change trajectory prediction device, the device comprising:

[0011] An acquisition module, configured to acquire a discrete data set, wherein the discrete data set includes motion trajectory points of at least two vehicles;

[0012] A first processing module, configured to process the discrete data set using an inflection point detection algorithm to obtain an inflection point set;

[0013] A second processing module is used to divide the discrete data set based on the inflection point set, and train the trajectory prediction model using the divided discrete data set to obtain a trained trajectory prediction model;

[0014] The prediction module is used to predict the lane-changing trajectory of the target vehicle according to the trained trajectory prediction model to obtain a prediction result.

[0015] In a third aspect, an embodiment of the present application provides a vehicle lane change trajectory prediction device, the device comprising a memory and a processor. The memory is used to store a computer program; the processor is used to implement the steps of the above-mentioned vehicle lane change trajectory prediction method when executing the computer program.

[0016] In a fourth aspect, an embodiment of the present application provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above-mentioned vehicle lane change trajectory prediction method are implemented.

[0017] The beneficial effects of the present invention are:

[0018] The present invention uses an inflection point detection algorithm to process a discrete data set to obtain an inflection point set, and proposes an identification method that uses the curvature of the discrete point set to determine the starting and ending points of the lane changing process. Compared with the prior art that uses a fixed heading angle for identification, the method has a higher recognition rate and makes the lane changing stage division effect better, thereby improving the accuracy of the trajectory prediction model.

[0019] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 Schematic diagram of the vehicle lane change trajectory prediction method described in an embodiment of the present invention.

[0022] Figure 2 Schematic diagram of the structure of the vehicle lane change trajectory prediction device described in an embodiment of the present invention.

[0023] Figure 3 Schematic diagram of the structure of a vehicle lane-changing trajectory prediction device described in an embodiment of the present invention.

[0024] Figure 4 Schematic diagram of the lane change phase division.

[0025] The following are marked in the figure: 901, acquisition module; 902, first processing module; 903, second processing module; 904, prediction module; 9021, fitting unit; 9022, first processing unit; 9023, second processing unit; 9024, third processing unit; 9031, eleventh processing unit; 9032, twelfth processing unit; 9033, thirteenth processing unit; 9034, marking unit; 9035, division unit; 9041, first acquisition unit; 9042, detection unit; 9043, construction unit; 9044, fourth processing unit; 9045, third acquisition unit; 9046, eighth processing unit; 9047, ninth processing unit; 9048, tenth processing unit; 90431, judgment unit; 90432, second acquisition unit; 90433, fifth processing unit; 90434, sixth processing unit; 90435, seventh processing unit; 800, vehicle lane change trajectory prediction device; 801, processor; 802, memory; 803, multimedia component; 804, I / O interface; 805, communication component. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0027] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0028] Embodiment 1:

[0029] This embodiment provides a method for predicting vehicle lane-changing trajectories. It is understandable that a scenario can be set up in this embodiment, for example, in a highway scenario, an autonomous driving vehicle needs to identify surrounding vehicles to change lanes.

[0030] See also Figure 1, the figure shows that the method includes step S1, step S2, step S3 and step S4, which specifically include:

[0031] Step S1, obtaining a discrete data set, wherein the discrete data set includes motion trajectory points of at least two vehicles;

[0032] Step S2, using an inflection point detection algorithm to process the discrete data set to obtain an inflection point set;

[0033] The step S2 also includes step S21, step S22, step S23 and step S24, which specifically include:

[0034] Step S21, fitting the motion trajectory points in the discrete data set using a smoothing spline curve to obtain a fitted data set;

[0035] In this step, the motion trajectory points in the discrete data set are fitted using a smooth spline curve to ensure the shape of the data set.

[0036] Step S22, normalizing the fitted data set to obtain a normalized data set;

[0037] In this step, the fitted data set is normalized to eliminate the influence of data scale.

[0038] Step S23, finding the local maximum point in the normalized data set to obtain a candidate inflection point set;

[0039] In this step, any point in the normalized data set is selected as the point to be processed (xi, yi), the horizontal coordinate of the point to be processed is kept unchanged, and the horizontal coordinate of the point to be processed is subtracted from the vertical coordinate to obtain the calculation result; a new data set is constructed according to the calculation result and the horizontal coordinate of the point to be processed, and the local maximum point in the new data set is found to obtain a set of candidate inflection points.

[0040] Step S24: select points with local maximum values ​​from the candidate inflection point set to obtain the inflection point set.

[0041] Step S3, dividing the discrete data set based on the inflection point set, and using the divided discrete data set to train the trajectory prediction model to obtain a trained trajectory prediction model;

[0042] In this step, if Figure 4As shown, the inflection point set can be used to divide the discrete data set into different stages of lane changing, specifically: the inflection point set can be divided into a set before the lane changing point and a set after the lane changing point. Suppose the total duration of the vehicle trajectory is t, the first inflection point from the lane changing point to time 0 is point T2, and the first inflection point traversed from the lane changing point to time t is point T3. How to determine the key node T2 in a complete trajectory will affect the accuracy and authenticity of the subsequent model. The recognition method of the present application using the curvature of the discrete point set to determine the starting and ending points of the lane changing process has a higher recognition rate than the prior art that uses a fixed heading angle for recognition, which makes the lane changing stage division effect better, thereby improving the accuracy of the trajectory prediction model prediction.

[0043] At the same time, the trajectory prediction model in this step includes two parts, one is the intention recognition module and the other is the trajectory prediction module. This application uses lanes to mark the divided discrete data sets for training the intention recognition module, and marks the divided discrete data sets with the lane change category of the vehicle for training the trajectory prediction module. The trajectory feature information and traffic flow feature information are sent to the intention recognition module to predict the vehicle's lane changing intention, and the output result is sent to the trajectory prediction module to output the predicted trajectory. It can be understood that the intention recognition module of this application adopts a network structure that combines the BiGRU model and the Attention mechanism.

[0044] It should be noted that before using the divided discrete data set to train the intention recognition module, the present application cropped the divided discrete data set to improve the efficiency of data processing without affecting the prediction accuracy. The specific process is: determine the time at which the lane change point is located to obtain the first time; traverse from the lane change point to time 0, with 1s as the interval, and obtain sequences of 3s in length forward, all of which are marked as corresponding lane change categories until they slide over the starting turning point of the lane change behavior, and mark them as straight-ahead sequences, and determine that the sequence in the time window is a time series; crop the divided discrete data set based on the time window, and after obtaining the cropped discrete data set, use lane division to train the intention recognition module.

[0045] The step S3 also includes step S31, step S32, step S33, step S34 and step S35, which specifically include:

[0046] Step S31, selecting an inflection point corresponding to any vehicle in the divided discrete data set to obtain a first inflection point;

[0047] Step S32, calculating the difference between the horizontal coordinates of the first inflection point in the first frame and the third frame, and determining the running direction of the vehicle corresponding to the first inflection point;

[0048] Step S33, determining a lane change category of the vehicle corresponding to the first inflection point according to the running direction;

[0049] Step S34, labeling the data in the divided discrete data set according to the lane change category to obtain a labeled data set;

[0050] Step S35: Divide the labeled data set according to the lane change category of the vehicle to obtain a training set.

[0051] It should be noted that the lane numbers of the data set are also marked in this application.

[0052] Step S4: predicting the lane-changing trajectory of the target vehicle according to the trained trajectory prediction model to obtain a prediction result.

[0053] The step S4 also includes step S41, step S42, step S43 and step S44, which specifically include:

[0054] Step S41, obtaining a preset distance threshold;

[0055] In this step, the present application does not limit the setting of the distance threshold, and the setting of the distance threshold varies depending on the type of vehicle.

[0056] Step S42: Detecting vehicles around the target vehicle according to the preset distance threshold to obtain a detection result;

[0057] In this step, only vehicles within the distance threshold are detected, and vehicles outside the distance threshold are invalid vehicles and have no effect on the vehicle's lane changing decision.

[0058] Step S43, constructing trajectory feature information according to the detection result;

[0059] The step S43 also includes step S431, step S432, step S433, step S434 and step S435, which specifically include:

[0060] Step S431, judging whether there is a first vehicle around the target vehicle according to the detection result, and obtaining a judgment result;

[0061] Step S432: when the judgment result is that there is a first vehicle around the target vehicle, acquiring historical driving data of the first vehicle and historical driving data of the target vehicle, wherein the first vehicle is a vehicle around the target vehicle, and the target vehicle is a vehicle that needs to make a trajectory prediction decision;

[0062] In this step, the first vehicle is a vehicle that affects the driving decision of the target vehicle. The present application takes into account the influence of the first vehicle on the driving decision of the target vehicle, and further improves the prediction accuracy of the target vehicle.

[0063] Step S433: determining a first historical feature according to the historical driving data of the first vehicle;

[0064] In this step, the first historical feature is specifically in

[0065] Δx t , Δy t Respectively represent the relative values ​​of the lateral coordinate and longitudinal coordinate of the target vehicle and the first vehicle at time t;

[0066] They represent the relative values ​​of the lateral speed and longitudinal speed of the target vehicle and the first vehicle at time t respectively; They respectively represent the relative values ​​of the lateral acceleration and longitudinal acceleration of the target vehicle and the first vehicle at time t.

[0067] Step S434, determining a second historical feature according to the historical driving data of the target vehicle;

[0068] In this step, the second historical feature is specifically Among them, x t ,y t are the lateral and longitudinal coordinates of the target vehicle at time t respectively;

[0069] are the lateral speed and longitudinal speed of the target vehicle at time t respectively; are the lateral acceleration and longitudinal acceleration of the target vehicle at time t, respectively.

[0070] Step S435: Determine trajectory feature information according to the first historical feature and the second historical feature.

[0071] It should be noted that, when the judgment result is that the first vehicle does not exist around the target vehicle, the characteristic information is recorded as infinite.

[0072] Step S44: predicting the lane-changing trajectory of the target vehicle according to the trajectory feature information.

[0073] After step S44, steps S45, S46, S47 and S48 are also included, which specifically include:

[0074] Step S45, dividing a time-space domain on the lane, and obtaining the size of the time-space domain;

[0075] Step S46, determining the sum of the moving distances of all vehicles in the time-space domain, the sum of the travel time of all vehicles in the time-space domain, and the traffic volume of all vehicles in the time-space domain;

[0076] Step S47, calculating according to the size of the spatiotemporal domain, the sum of the moving distances of all vehicles in the spatiotemporal domain, the sum of the travel time of all vehicles in the spatiotemporal domain, and the traffic volume of all vehicles in the spatiotemporal domain to obtain traffic flow characteristic information;

[0077] In this step, the traffic flow characteristic information is specifically:

[0078]

[0079]

[0080]

[0081] In the above formula, Q(A) is the total traffic volume in the spatiotemporal domain A; t(A) is the sum of the travel time of all vehicles in the spatiotemporal domain A; d(A) is the sum of the moving distances of all vehicles in the spatiotemporal domain A; |A| represents the size of the spatiotemporal region, which is the value of the time length multiplied by the space length; q(A), k(A) and v(A) represent the average traffic flow, average traffic density and spatial average speed respectively.

[0082] Step S48: predicting the lane-changing trajectory of the target vehicle based on the trajectory feature information and the traffic flow feature information.

[0083] In this step, the trajectory feature information and the traffic flow feature information are standardized, and the standardized traffic flow features and the standardized trajectory feature information are input into the trajectory prediction model to effectively improve the prediction accuracy.

[0084] In this embodiment, from the perspective of traffic characteristics, during the driving process of the vehicle, there may be a process of changing lanes from a high-flow, high-density lane to a low-flow, low-density lane. However, most studies use only the historical trajectories of the environmental vehicle and the predicted vehicle as input, without considering the traffic flow characteristics during the vehicle lane changing process, and do not consider the differences in driving characteristics of vehicles in different lanes. Therefore, the present application uses trajectory feature information and traffic flow feature information to predict lane changing trajectories, which can effectively improve the accuracy of the predicted trajectory, thereby better assisting the lane changing decision of the autonomous driving vehicle.

[0085] Embodiment 2:

[0086] like Figure 2As shown, this embodiment provides a vehicle lane change trajectory prediction device, the device includes an acquisition module 901, a first processing module 902, a second processing module 903 and a prediction module 904, which specifically includes:

[0087] An acquisition module 901 is used to acquire a discrete data set, where the discrete data set includes motion trajectory points of at least two vehicles;

[0088] A first processing module 902 is used to process the discrete data set using an inflection point detection algorithm to obtain an inflection point set;

[0089] A second processing module 903 is used to divide the discrete data set based on the inflection point set, and train the trajectory prediction model using the divided discrete data set to obtain a trained trajectory prediction model;

[0090] The prediction module 904 is used to predict the lane-changing trajectory of the target vehicle according to the trained trajectory prediction model to obtain a prediction result.

[0091] In a specific implementation manner disclosed in this embodiment, the first processing module 902 further includes a fitting unit 9021, a first processing unit 9022, a second processing unit 9023 and a third processing unit 9024, which specifically include:

[0092] A fitting unit 9021, configured to fit the motion trajectory points in the discrete data set using a smooth spline curve to obtain a fitted data set;

[0093] A first processing unit 9022 is used to perform normalization processing on the fitted data set to obtain a normalized data set;

[0094] The second processing unit 9023 is used to find the local maximum point in the normalized data set to obtain a candidate inflection point set;

[0095] The third processing unit 9024 is used to select points with local maximum values ​​from the candidate inflection point set to obtain the inflection point set.

[0096] In a specific implementation manner disclosed in this embodiment, the second processing module 903 further includes an eleventh processing unit 9031, a twelfth processing unit 9032, a thirteenth processing unit 9033, a labeling unit 9034 and a dividing unit 9035, which specifically include:

[0097] An eleventh processing unit 9031 is used to select an inflection point corresponding to any vehicle in the inflection point set to obtain a first inflection point;

[0098] A twelfth processing unit 9032 is used to calculate the difference between the horizontal coordinates of the first frame and the third frame of the first inflection point, and determine the running direction of the vehicle corresponding to the first inflection point;

[0099] A thirteenth processing unit 9033 is used to determine the lane change category of the vehicle corresponding to the first turning point according to the running direction;

[0100] A labeling unit 9034, configured to label the data in the inflection point set according to the lane change category to obtain a labeled data set;

[0101] The division unit 9035 is used to divide the labeled data set according to the lane change category of the vehicle to obtain a training set.

[0102] In a specific implementation manner disclosed in this embodiment, the prediction module 904 further includes a first acquisition unit 9041, a detection unit 9042, a construction unit 9043 and a fourth processing unit 9044, which specifically include:

[0103] A first acquiring unit 9041 is used to acquire a preset distance threshold;

[0104] The detection unit 9042 is used to detect vehicles around the target vehicle according to the preset distance threshold and obtain a detection result;

[0105] A construction unit 9043, configured to construct trajectory feature information according to the detection result;

[0106] The fourth processing unit 9044 is used to predict the lane-changing trajectory of the target vehicle according to the trajectory feature information.

[0107] In a specific implementation manner disclosed in this embodiment, the construction unit 9043 further includes a judgment unit 90431, a second acquisition unit 90432, a fifth processing unit 90433, a sixth processing unit 90434 and a seventh processing unit 90435, which specifically include:

[0108] A judgment unit 90431 is used to judge whether there is a first vehicle around the target vehicle according to the detection result, and obtain a judgment result;

[0109] A second acquisition unit 90432 is used to acquire historical driving data of the first vehicle and historical driving data of the target vehicle when the judgment result is that there is a first vehicle around the target vehicle, the first vehicle is a vehicle around the target vehicle, and the target vehicle is a vehicle that needs to make a trajectory prediction decision;

[0110] a fifth processing unit 90433, configured to determine a first historical feature according to the historical driving data of the first vehicle;

[0111] A sixth processing unit 90434, configured to determine a second historical feature according to the historical driving data of the target vehicle;

[0112] The seventh processing unit 90435 is used to determine trajectory feature information according to the first historical feature and the second historical feature.

[0113] In a specific implementation manner disclosed in this embodiment, the fourth processing unit 9044 further includes a third acquisition unit 9045, an eighth processing unit 9046, a ninth processing unit 9047 and a tenth processing unit 9048, which specifically include:

[0114] The third acquisition unit 9045 is used to divide a time-space domain on the lane and obtain the size of the time-space domain;

[0115] An eighth processing unit 9046, configured to determine the sum of the moving distances of all vehicles in the time-space domain, the sum of the travel time of all vehicles in the time-space domain, and the traffic volume of all vehicles in the time-space domain;

[0116] A ninth processing unit 9047, configured to obtain traffic flow characteristic information by performing calculation according to the size of the spatiotemporal domain, the sum of the moving distances of all vehicles in the spatiotemporal domain, the sum of the travel time of all vehicles in the spatiotemporal domain, and the traffic volume of all vehicles in the spatiotemporal domain;

[0117] The tenth processing unit 9048 is used to predict the lane-changing trajectory of the target vehicle according to the trajectory feature information and the traffic flow feature information.

[0118] It should be noted that, regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.

[0119] Embodiment 3:

[0120] Corresponding to the above method embodiment, a vehicle lane-changing trajectory prediction device is also provided in this embodiment. The vehicle lane-changing trajectory prediction device described below and the vehicle lane-changing trajectory prediction method described above can refer to each other.

[0121] Figure 3 FIG. 8 is a block diagram of a vehicle lane change trajectory prediction device 800 according to an exemplary embodiment. Figure 3 As shown, the vehicle lane-changing trajectory prediction device 800 may include: a processor 801 and a memory 802. The vehicle lane-changing trajectory prediction device 800 may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0122] The processor 801 is used to control the overall operation of the vehicle lane change trajectory prediction device 800 to complete all or part of the steps in the above-mentioned vehicle lane change trajectory prediction method. The memory 802 is used to store various types of data to support the operation of the vehicle lane change trajectory prediction device 800, and these data may include, for example, instructions for any application or method operating on the vehicle lane change trajectory prediction device 800, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 802 or sent via the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, and the above-mentioned other interface modules can be a keyboard, a mouse, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the vehicle lane change trajectory prediction device 800 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include: Wi-Fi module, Bluetooth module, NFC module.

[0123] In an exemplary embodiment, the vehicle lane change trajectory prediction device 800 can be implemented by one or more application specific integrated circuits (Application Specific Integrated Circuit, referred to as ASIC), digital signal processors (Digital Signal Processor, referred to as DSP), digital signal processing devices (Digital Signal Processing Device, referred to as DSPD), programmable logic devices (Programmable Logic Device, referred to as PLD), field programmable gate arrays (Field Programmable Gate Array, referred to as FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned vehicle lane change trajectory prediction method.

[0124] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned vehicle lane change trajectory prediction method are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 802 including program instructions, and the above-mentioned program instructions can be executed by the processor 801 of the vehicle lane change trajectory prediction device 800 to complete the above-mentioned vehicle lane change trajectory prediction method.

[0125] Embodiment 4:

[0126] Corresponding to the above method embodiment, a readable storage medium is also provided in this embodiment. The readable storage medium described below and the vehicle lane change trajectory prediction method described above can refer to each other.

[0127] A readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the vehicle lane change trajectory prediction method of the above method embodiment.

[0128] The readable storage medium may specifically be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or other readable storage medium that can store program codes.

[0129] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0130] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A vehicle lane-changing trajectory prediction method, characterized in that: include: Acquire a discrete data set, wherein the discrete data set includes motion trajectory points of at least two vehicles; Processing the discrete data set using an inflection point detection algorithm to obtain an inflection point set; Dividing the discrete data set based on the inflection point set, and training the trajectory prediction model using the divided discrete data set to obtain a trained trajectory prediction model; The lane-changing trajectory of the target vehicle is predicted according to the trained trajectory prediction model to obtain a prediction result.

2. The vehicle lane-changing trajectory prediction method according to claim 1, characterized in that: The discrete data set is processed using an inflection point detection algorithm to obtain an inflection point set, including: Using a smoothing spline curve to fit the motion trajectory points in the discrete data set to obtain a fitted data set; Normalizing the fitted data set to obtain a normalized data set; Finding the local maximum point in the normalized data set to obtain a candidate inflection point set; Points with local maximum values ​​are screened out from the candidate inflection point set to obtain the inflection point set.

3. The vehicle lane-changing trajectory prediction method according to claim 1, characterized in that: Predicting the lane-changing trajectory of the target vehicle according to the trained trajectory prediction model includes: Get the preset distance threshold; Detecting vehicles around the target vehicle according to the preset distance threshold to obtain a detection result; Constructing trajectory feature information according to the detection results; The lane-changing trajectory of the target vehicle is predicted based on the trajectory feature information.

4. The vehicle lane-changing trajectory prediction method according to claim 3, characterized in that: After predicting the lane-changing trajectory of the target vehicle according to the trajectory feature information, the method further includes: Divide a spatiotemporal domain on the lane and obtain the size of the spatiotemporal domain; Determine the sum of the moving distances of all vehicles in the time-space domain, the sum of the travel time of all vehicles in the time-space domain, and the traffic volume of all vehicles in the time-space domain; Calculating according to the size of the spatiotemporal domain, the sum of the moving distances of all vehicles in the spatiotemporal domain, the sum of the travel time of all vehicles in the spatiotemporal domain, and the traffic volume of all vehicles in the spatiotemporal domain to obtain traffic flow characteristic information; The lane-changing trajectory of the target vehicle is predicted based on the trajectory feature information and the traffic flow feature information.

5. A vehicle lane-changing trajectory prediction device, characterized in that: include: An acquisition module, configured to acquire a discrete data set, wherein the discrete data set includes motion trajectory points of at least two vehicles; A first processing module, configured to process the discrete data set using an inflection point detection algorithm to obtain an inflection point set; A second processing module is used to divide the discrete data set based on the inflection point set, and train the trajectory prediction model using the divided discrete data set to obtain a trained trajectory prediction model; The prediction module is used to predict the lane-changing trajectory of the target vehicle according to the trained trajectory prediction model to obtain a prediction result.

6. The vehicle lane-changing trajectory prediction device according to claim 5, characterized in that: The first processing module comprises: A fitting unit, used for fitting the motion trajectory points in the discrete data set using a smooth spline curve to obtain a fitted data set; A first processing unit, configured to perform normalization processing on the fitted data set to obtain a normalized data set; A second processing unit is used to find the local maximum point in the normalized data set to obtain a candidate inflection point set; The third processing unit is used to select points with local maximum values ​​from the candidate inflection point set to obtain the inflection point set.

7. The vehicle lane-changing trajectory prediction device according to claim 5, characterized in that: The prediction module comprises: A first acquisition unit, used to acquire a preset distance threshold; A detection unit, used to detect vehicles around the target vehicle according to the preset distance threshold to obtain a detection result; A construction unit, configured to construct trajectory feature information according to the detection result; The fourth processing unit is used to predict the lane-changing trajectory of the target vehicle according to the trajectory feature information.

8. The vehicle lane-changing trajectory prediction device according to claim 7, characterized in that: After the fourth processing unit, the method further includes: A third acquisition unit is used to divide a time-space domain on the lane and obtain the size of the time-space domain; an eighth processing unit, for determining the sum of the moving distances of all vehicles in the time-space domain, the sum of the travel time of all vehicles in the time-space domain, and the traffic volume of all vehicles in the time-space domain; a ninth processing unit, configured to obtain traffic flow characteristic information by performing calculations according to the size of the spatiotemporal domain, the sum of the moving distances of all vehicles in the spatiotemporal domain, the sum of the travel time of all vehicles in the spatiotemporal domain, and the traffic volume of all vehicles in the spatiotemporal domain; The tenth processing unit is used to predict the lane-changing trajectory of the target vehicle according to the trajectory characteristic information and the traffic flow characteristic information.

9. A vehicle lane change trajectory prediction device, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the vehicle lane change trajectory prediction method as claimed in any one of claims 1 to 4 when executing the computer program.

10. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the vehicle lane change trajectory prediction method according to any one of claims 1 to 4 are implemented.

Images