A trajectory point optimization method, device, equipment and medium

Abstract

The present application provides a trajectory point optimization method and device, electronic equipment and computer readable storage medium, wherein the trajectory point optimization method comprises: obtaining steering angle data and GPS data of a vehicle; determining valid GPS data in the GPS data according to the steering angle data; and generating a corresponding trajectory of the vehicle according to the valid GPS data. In the present application, the steering angle data of the vehicle is obtained at the same time as the GPS data, the validity of the GPS data is determined according to the steering angle data, the valid GPS data for vehicle positioning is obtained, and the vehicle driving trajectory is generated according to the valid GPS data.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of automobile positioning, and in particular to a trajectory point optimization method and device, an electronic device and a computer readable storage medium. BACKGROUND

[0002] ADAS (Advanced-Driving Assistance System) is a system that uses various sensors installed on a vehicle, such as millimeter wave radar, laser radar, monocular / dual camera, and satellite navigation, to sense the environment around the vehicle at all times during driving, collect data, identify, detect, and track static and dynamic objects, and combine navigation map data to perform system calculations and analysis.

[0003] In the ADAS data collection scheme, the ADAS data collected by each vehicle can be presented in real time or played back afterwards using a software interface, for example, GPS (Global Positioning System) data of each road sampling vehicle can be collected, and then the positions and trajectories of each road sampling vehicle can be displayed on a map. In the prior art, non-important position points can be found based on position points in GPS data and filtered out to optimize the GPS data, reduce the number of position points to be processed, and not significantly change the trajectory display effect. However, in this scheme, at least three position points are required, and only the second position point can be determined whether to discard, so only the middle position point can be determined whether to discard, and the determination is based on the angle offset between position points. If the GPS position points are missing, the angle offset cannot accurately reflect the influence of the position points on the trajectory display effect, and thus the optimized GPS data cannot guarantee the accuracy of the trajectory display effect.

[0004] Therefore, the prior art still needs to be improved and developed. SUMMARY

[0005] The embodiments of the present application provide a trajectory point optimization method, device, electronic device, and computer readable storage medium, thereby solving the problems of poor real-time performance and easy errors of the existing trajectory point optimization scheme.

[0006] In one aspect, the present application provides a trajectory point optimization method, comprising the following steps:

[0007] obtaining steering angle data and GPS data of a vehicle;

[0008] determining valid GPS data in the GPS data according to the steering angle data;

[0009] generating a corresponding trajectory of the vehicle according to the valid GPS data.

[0010] In some embodiments of the present application, the GPS data comprises target GPS data and non-target GPS data, and before the step of determining valid GPS data from the GPS data according to the steering angle data, the method further comprises:

[0011] determining a target road corresponding to the target GPS data, and obtaining execution reference information of the target road;

[0012] wherein the execution reference information is used to determine a required number of non-target GPS data between two adjacent target GPS data or a required interval time between two adjacent target GPS data from a plurality of GPS data corresponding to the target road.

[0013] In some embodiments of the present application, the execution reference information is determined by the following steps:

[0014] obtaining road labels of a plurality of preset roads;

[0015] determining the execution reference information of the target road according to the road labels of the plurality of roads and preset execution reference information corresponding to the road labels, wherein the execution reference information corresponding to at least two road labels is different.

[0016] In some embodiments of the present application, the steering angle data comprises a steering angle and a steering angle acquisition time corresponding to the steering angle, and the GPS data comprises a coordinate point and a coordinate point acquisition time corresponding to the coordinate point, wherein the step of determining valid GPS data from the GPS data according to the steering angle data comprises:

[0017] for any target GPS data in the GPS data, obtaining a plurality of steering angle data in a first preset time period covering the coordinate point acquisition time according to the coordinate point acquisition time in the target GPS data;

[0018] obtaining an average value of the steering angle in the first preset time period according to a plurality of steering angles corresponding to a plurality of steering angle data;

[0019] if the average value is greater than a preset steering angle threshold, determining the target GPS data as the valid GPS data.

[0020] In some embodiments of the present application, the step of determining valid GPS data from the GPS data according to the steering angle data is performed by a trained neural network model, the steering angle data corresponding to the target GPS data is input into the neural network model, and the neural network model outputs a judgment result of whether the target GPS data is valid or invalid. The training method of the neural network model comprises:

[0021] Obtaining training data, the training data being a plurality of historical steering angle data with valid labels or invalid labels;

[0022] Inputting the plurality of historical steering angle data with valid labels or invalid labels into the neural network model;

[0023] Calculating a training loss value based on the judgment result of the historical steering angle data with valid labels or invalid labels and the labels corresponding to the historical steering angle data with valid labels or invalid labels, and updating the neural network model according to the training loss value.

[0024] In some embodiments of the present application, the step of determining valid GPS data from the GPS data according to the steering angle data is performed by a trained neural network model, the steering angle data corresponding to the target GPS data is input into the neural network model, and the neural network model outputs a judgment result of whether the target GPS data is valid or invalid. The training method of the neural network model comprises:

[0025] Obtaining training data, the training data being a plurality of historical steering angle data with different probability value labels, wherein the historical steering angle data with probability value labels includes a plurality of historical steering angle data with valid labels or invalid labels, and the probability value label is a proportion of the historical steering angle data with valid labels in the historical steering angle data matrix;

[0026] Inputting the plurality of historical steering angle data with different probability value labels into the probability value generation model;

[0027] Calculating a training loss value based on the judgment result of the historical steering angle data with probability value labels and the probability value labels corresponding to the historical steering angle data with probability value labels, and updating the probability value generation model according to the training loss value.

[0028] If the valid probability value of the target GPS data is greater than a preset valid probability threshold, it is determined that the target GPS data is valid GPS data.

[0029] In some embodiments of the present application, the effective probability threshold is preset with multiple values, and the effective probability threshold corresponds to the road label and the execution reference information of the target road corresponding to the target GPS data; wherein the effective probability threshold is determined according to the road label of the target road.

[0030] In another aspect, the present application provides a trajectory point optimization device, comprising:

[0031] a data acquisition module, configured to acquire steering angle data and GPS data of a vehicle;

[0032] a data determination module, configured to determine effective GPS data in the GPS data according to the steering angle data;

[0033] a trajectory generation module, configured to generate a corresponding trajectory of the vehicle according to the effective GPS data.

[0034] In another aspect, the present application further provides an electronic device, comprising:

[0035] a memory, configured to store computer readable instructions;

[0036] a processor, configured to read the computer readable instructions stored in the memory to perform the steps in the trajectory point optimization method.

[0037] In a fourth aspect, the present application further provides a computer readable storage medium, which stores computer readable instructions, and when the computer readable instructions are executed by a processor of a computer, the computer performs the steps in the trajectory point optimization method.

[0038] Beneficial effects: According to the present application, the steering angle data of the vehicle is acquired at the same time as the GPS data of the vehicle, and the effectiveness of the GPS data is determined according to the steering angle data, so as to acquire the effective GPS data for vehicle positioning, and generate a vehicle driving trajectory according to the effective GPS data. On the one hand, the acquired GPS data can be judged in real time to determine whether it is the effective GPS data. On the other hand, the steering angle data is independent of the GPS data, and will not change with the errors and omissions of the position points in the GPS data. The effectiveness of the GPS data is confirmed based on the steering angle data, which can improve the accuracy of the effective GPS data. In addition, by screening the GPS data, the data processing amount required for fitting the trajectory can be reduced, and the processing efficiency can be improved. In addition, in the present embodiment, only the effective GPS data can be saved, and the complete GPS data acquired does not need to be saved. BRIEF DESCRIPTION OF DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description only show some of the embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort based on these drawings.

[0040] Figure 1 is an embodiment flow diagram of the trajectory point optimization method provided by the embodiments of the present application;

[0041] Figure 2 is an embodiment structure diagram of the trajectory point optimization device provided by the embodiments of the present application;

[0042] Figure 3 is an embodiment structure diagram of the electronic device provided by the embodiments of the present application. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort fall within the scope of protection of the present application.

[0044] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, and are only used to facilitate the description of the present application and simplify the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation of the present application. In addition, the terms "first", "second", "third", "fourth" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features limited by "first", "second", "third", "fourth" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0045] In this application, the word "exemplary" is used to mean "serving as an example, instance, or illustration." Any implementation described as "exemplary" is not necessarily to be construed as preferred or advantageous over other implementations. The following description is presented to enable any person skilled in the art to make and use the application. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present application. It will be apparent, however, to one skilled in the art that the present application can be practiced without using these specific details. In other instances, well-known structures and processes are not elaborated

[0046] It should be noted that the method provided in the embodiments of the present application is executed in a computer device, and the processing objects of each computer device exist in the form of data or information, such as time, which is actually time information. It can be understood that if the size, quantity, position and the like are mentioned in subsequent embodiments, the corresponding data exist, so that the computer device can process, and details are not described herein.

[0047] The embodiments of the present application provide a trajectory point optimization method and device, electronic equipment and computer readable storage medium, which are described in detail as follows.

[0048] Please refer to Figure 1 , Figure 1 The flowchart of the trajectory point optimization method provided by the embodiments of the present application is shown in the figure, and the trajectory point optimization method comprises the following steps.

[0049] S100, acquiring steering angle data and GPS data of a vehicle.

[0050] The steering angle data can be data obtained by detecting the rotation of the steering wheel at the steering wheel by a detector, or data obtained by detecting the steering of the wheels at the wheels by a detector. The GPS data is data or coordinates used for vehicle positioning collected in fixed time frequency in the process of vehicle driving.

[0051] S200, determining valid GPS data in the GPS data according to the steering angle data.

[0052] In the embodiments, the GPS data includes valid GPS data and invalid GPS data, and the valid GPS data in the GPS data can be determined according to the steering angle data.

[0053] Specifically, for any GPS data, one or more groups of steering angle data closest in time can be found, and it is determined whether the GPS data is valid GPS data based on the found steering angle data, for example, when the vehicle has no obvious steering, i.e. for GPS data at a time, the data of the one or more groups of steering angle data closest in time found are all small, then the GPS data at the time is determined as invalid GPS data, and when the vehicle has obvious steering, for GPS data at another time, the values of the one or more groups of steering angle data closest in time found are large, then the GPS data at the other time is determined as valid GPS data, and the data is saved.

[0054] S300, generating a corresponding trajectory of the vehicle according to the valid GPS data.

[0055] Understandably, when a certain amount of valid GPS data is obtained, a corresponding trajectory of the vehicle can be generated according to the valid GPS data.

[0056] In the embodiment, by obtaining the steering angle data of the vehicle at the same time as obtaining the GPS data of the vehicle, and determining the validity of the GPS data according to the steering angle data, the valid GPS data is obtained, and a corresponding trajectory of the vehicle is generated according to the valid GPS data. On the one hand, the real-time GPS data obtained can be judged to determine whether it is valid GPS data, and on the other hand, the validity of the GPS data is confirmed based on the steering angle data, which can improve the judgment accuracy of the valid GPS data. In addition, in the embodiment, only valid GPS data can be saved, and the complete GPS data obtained does not need to be saved.

[0057] In a specific embodiment, the GPS data includes target GPS data and non-target GPS data, and before step S200, the method further includes:

[0058] M210, determining a target road corresponding to the target GPS data, and obtaining execution reference information of the target road;

[0059] The execution reference information is used to determine the number of non-target GPS data required to be reached between two adjacent groups of target GPS data or the interval time required to be reached between two adjacent groups of target GPS data in a plurality of GPS data corresponding to the target road.

[0060] In the embodiment, considering that the road condition during the driving of the vehicle is constantly changing, when the vehicle drives on a road with branches, more GPS data is required for the positioning of the vehicle, and when the vehicle drives on a straight road with simple road condition, less GPS data is required for the positioning of the vehicle. Therefore, when the GPS data for the positioning of the vehicle is acquired at a fixed frequency, all the acquired GPS data does not need to be used to perform the step S200 to determine whether the GPS data is valid.

[0061] For example, when the vehicle drives on a straight road with simple road condition, the GPS data of the vehicle is collected every 10 seconds, and 6 GPS data collected within one minute is taken as one group. Every 5 groups of GPS data, the first GPS data in the first group, the first GPS data in the sixth group, the first GPS data in the eleventh group, and the like are taken as target GPS data to perform the step S200 to determine whether the target GPS data is valid, and the other GPS data is non-target GPS data, which does not need to continue to perform the step S200. It can be seen that the target GPS data refers to the GPS data which needs to be determined whether it is valid.

[0062] When the vehicle drives on a road with branches with complex road condition, the GPS data of the vehicle is still collected every 10 seconds, and 6 GPS data collected within one minute is taken as one group. Every one group of GPS data, the first GPS data in the first group, the first GPS data in the second group, the first GPS data in the third group, and the like are taken as target GPS data to perform the step S200 to determine whether the target GPS data is valid, and the other GPS data is non-target GPS data, which does not need to continue to perform the step S200.

[0063] Therefore, in consideration of the influence of road conditions on target GPS data, it is necessary to determine the target road corresponding to the target GPS data. Understandably, no matter what kind of road the vehicle travels on, GPS data for vehicle positioning is collected at a fixed frequency, which includes target GPS data and non-target GPS data. Then, the execution reference information of the target road can be obtained according to the target road, wherein the execution reference information is used to determine the number of non-target GPS data required to be reached between two adjacent groups of target GPS data in the plurality of GPS data corresponding to the target road or the interval time required to be reached between two adjacent groups of target GPS data, which can also be understood as: the execution reference data determines the frequency of target GPS data. Taking the number and interval time as examples for description, they can be inversely proportional to the complexity and / or bending degree of the road, for example, the more complex and / or the higher the bending degree of the road, the smaller the value of the number and interval time; correspondingly, taking the frequency as an example for description, it can be proportional to the complexity and / or bending degree of the road, for example, the more complex and / or the higher the bending degree of the road, the greater the value of the frequency. Based on this, different road labels can reflect different road complexity and / or bending degree.

[0064] The execution reference information can be the number, interval time, frequency itself, or other arbitrary data used to calculate or represent the number, interval time, and frequency, which can be understood as being used to determine the number of non-target GPS data required to be reached between two adjacent groups of target GPS data or the interval time required to be reached between two adjacent groups of target GPS data.

[0065] In the embodiment, the acquired GPS data for vehicle positioning is classified into target GPS data and non-target GPS data in advance, and the non-target GPS data is directly discarded, while the target GPS data is further confirmed for validity according to the steering angle data. If the target GPS data is determined to be valid GPS data, it is retained, and if the target GPS data is determined to be invalid GPS data, it is discarded, thereby effectively improving the screening efficiency of screening a plurality of valid GPS data for generating a vehicle trajectory from all acquired GPS data.

[0066] In a specific embodiment, the execution reference information is determined by the following steps:

[0067] Obtaining road labels of a plurality of preset roads;

[0068] According to the road labels of the plurality of roads and the preset execution reference information corresponding to the road labels, determining the execution reference information of the target road; wherein the execution reference information corresponding to at least two road labels is different.

[0069] In the embodiment, the road is divided into different types and with different road labels, such as straight road, curve road without branch, curve road with branch, etc. In addition, for the same type of road, different road labels can be formed according to different bending degrees and complexity, for example, for curve road without branch, there can be curve road without branch with low bending degree and curve road without branch with medium bending degree. In addition, when dividing the road, it can be divided based on length or based on specific intersections (such as crossroads and / or T-shaped intersections), for example, the road between adjacent crossroads or the road between intersections with K crossroads can be divided, and K can be greater than 2.

[0070] The road with different road labels has different execution reference information, for example, if the execution reference information is used to determine the required interval time length between two adjacent sets of target GPS data, the interval time length corresponding to the road label of straight road can be greater than the interval time length corresponding to the road label of curve road without branch. The road with different road labels and the corresponding execution reference information are pre-set, and when the target road is determined, the execution reference information of the target road can be obtained, that is, when the vehicle travels to the target road, the required number of non-target GPS data between two adjacent sets of target GPS data or the required interval time length between two adjacent sets of target GPS data is determined, and the target GPS data that needs to perform step S200 is determined.

[0071] The steering angle data includes a steering angle and a steering angle acquisition time corresponding to the steering angle, and the GPS data includes a coordinate point and a coordinate point acquisition time corresponding to the coordinate point. In a specific embodiment, step S200 includes:

[0072] S210, for any target GPS data in the GPS data, a plurality of steering angle data in a first preset time period covering the coordinate point acquisition time in the target GPS data is obtained according to the coordinate point acquisition time in the target GPS data.

[0073] The acquisition time of the target GPS data and the acquisition time of the steering angle data can be the same or different. In the embodiment, a plurality of steering angle data in a first preset time period covering the coordinate point acquisition time of the target GPS data is obtained.

[0074] S220, according to a plurality of steering angles corresponding to a plurality of steering angle data, an average value of the steering angle in the first preset time period is obtained.

[0075] S230, if the average value is greater than a preset steering angle threshold value, determining that the target GPS data is the valid GPS data.

[0076] In the embodiment, in order to improve the authenticity of the target GPS data as valid GPS data, when the average value of the steering angle is greater than the preset steering angle threshold value, further, a plurality of steering angle data in a second preset time period adjacent to the first preset time period is obtained, and the plurality of steering angle data in the second preset time period is compared with the steering angle threshold value, if the plurality of steering angle data in the second preset time period is greater than the steering angle threshold value, it is determined that the target GPS data is valid GPS data, otherwise, it is determined to be invalid GPS data.

[0077] In a specific embodiment, step S200 is executed by a trained neural network model, the steering angle data corresponding to the target GPS data is input into the neural network model, and the judgment result of the validity of the target GPS data is output by the neural network model, wherein the training method of the neural network model comprises:

[0078] obtaining training data, the training data being a plurality of historical steering angle data with valid labels or invalid labels;

[0079] inputting the plurality of historical steering angle data with valid labels or invalid labels into the neural network model;

[0080] calculating a training loss value based on the judgment result of the historical steering angle data with valid labels or invalid labels and the labels corresponding to the historical steering angle data with valid labels or invalid labels, and updating the neural network model according to the training loss value.

[0081] In another specific embodiment, step S200 is executed by a trained neural network model, the steering angle data corresponding to the target GPS data is input into the neural network model, and the validity probability value of the GPS data is output by the neural network model, wherein the training method of the neural network model comprises:

[0082] obtaining training data, the training data being a plurality of historical steering angle data matrices with different probability value labels, wherein the historical steering angle data matrix with probability value labels includes a plurality of historical steering angle data with valid labels or invalid labels, and the probability value label is the proportion of the historical steering angle data with the valid label in the historical steering angle data matrix;

[0083] inputting the plurality of historical steering angle data matrices with different probability value labels into the probability value generation model;

[0084] calculate a training loss value based on the determination result of the historical steering angle data matrix with the probability value label and the probability value label corresponding to the historical steering angle data matrix with the probability value label, and update the probability value generation model according to the training loss value;

[0085] If the effective probability value of the target GPS data is greater than a preset effective probability threshold, the target GPS data is determined as the effective GPS data.

[0086] In an embodiment, the effective probability threshold is preset with multiple values, and the effective probability threshold corresponds to a road label of the target road corresponding to the target GPS data; wherein the effective probability threshold is determined according to the road label of the target road.

[0087] In this embodiment, considering that the road conditions are constantly changing during the driving of the vehicle, when the vehicle drives on a road with a fork and the road condition is relatively complex, more GPS data is required for vehicle positioning, the effective probability threshold needs to be increased, and more GPS data needs to be retained as effective GPS data to generate the driving trajectory of the vehicle. When the vehicle drives on a straight road with a simple road condition, less GPS data is required for vehicle positioning, and the effective probability threshold needs to be reduced, and less GPS data needs to be retained as effective GPS data. Therefore, the effective probability threshold needs to be determined according to the road label of the target road.

[0088] In order to better implement the trajectory point optimization method in the embodiments of the present application, on the basis of the trajectory point optimization method, the embodiments of the present application also provide a trajectory point optimization device, as shown in Figure 2 The trajectory point optimization device 600 includes:

[0089] The data acquisition module 601 is configured to acquire the steering angle data and the GPS data of the vehicle.

[0090] The data determination module 602 is configured to determine the effective GPS data in the GPS data according to the steering angle data.

[0091] The trajectory generation module 603 is configured to generate the corresponding trajectory of the vehicle according to the effective GPS data.

[0092] The embodiments of the present application also provide an electronic device, which includes:

[0093] The memory stores computer readable instructions;

[0094] The processor reads the computer readable instructions stored in the memory to perform the steps in the trajectory point optimization method.

[0095] The embodiment of the present application further provides an electronic device integrating any trajectory point optimization device provided by the embodiment of the present application. As shown in Figure 3 The embodiment of the present application further provides an electronic device integrating any trajectory point optimization device provided by the embodiment of the present application. As shown in

[0096] The electronic device can include a processor 701 with one or more processing cores, a memory 702 with one or more computer readable storage media, a power supply 703, an input unit 704, and the like. Those skilled in the art can understand that the structure of the computer device shown in the embodiment of the present application does not constitute a limitation on the computer device, and can include more or fewer components than those shown in the figure, or combine certain components, or different component arrangements. Among them: Figure 3

[0097] The processor 701 is the control center of the electronic device, and connects various parts of the computer device through various interfaces and lines. By running or executing software programs and / or modules stored in the memory 702, and calling data stored in the memory 702, the processor 701 performs various functions and processes data of the electronic device, thereby overall monitoring the computer device. Optionally, the processor 701 can include one or more processing cores; preferably, the processor 701 can integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, etc., and the modem processor mainly processes wireless communication. It can be understood that the above-mentioned modem processor can also not be integrated into the processor 701.

[0098] The memory 702 can be used to store software programs and modules. The processor 701 executes various functions and data processing by running the software programs and modules stored in the memory 702. The memory 702 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application program required by a function (such as a sound playing function, an image playing function, etc.), and the like; the data storage area can store data created according to the use of the computer device, etc. In addition, the memory 702 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state memory device. Accordingly, the memory 702 can further include a memory controller to provide access for the processor 701 to the memory 702.

[0099] ​The electronic device can further include a power supply 703 for supplying power to each component. Preferably, the power supply 703 can be logically connected to the processor 701 through a power management system, so that the power management system can perform functions such as management of charging, discharging, and power consumption management. The power supply 703 can also include one or more DC or AC power sources, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and the like.

[0100] The electronic device can further include an input unit 704 for receiving input digital or character information, and generating keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function controls.

[0101] Although not shown, the electronic device can further include a display unit and the like, which will not be described here. In particular, in the present embodiment, the processor 701 in the electronic device loads one or more executable files corresponding to processes of one or more application programs into the memory 702, and runs the application programs stored in the memory 702 according to the following instructions, so as to implement various functions, such as:

[0102] obtaining steering angle data and GPS data of a vehicle;

[0103] determining valid GPS data from the GPS data according to the steering angle data;

[0104] generating a corresponding trajectory of the vehicle according to the valid GPS data.

[0105] It can be understood by those skilled in the art that all or part of the steps in the various methods of the above embodiments can be completed by instructions, or by controlling related hardware by the instructions, which can be stored in a computer readable storage medium and loaded and executed by a processor.

[0106] To this end, an embodiment of the present application provides a computer readable storage medium, which can include a read only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and the like. A computer program is stored on the storage medium, and the computer program is loaded by a processor to execute steps of any trajectory point optimization method provided by the embodiments of the present application. For example, the computer program loaded by the processor can execute the following steps:

[0107] obtaining steering angle data and GPS data of a vehicle;

[0108] determining valid GPS data from the GPS data according to the steering angle data;

[0109] generate a respective trajectory of the vehicle based on the valid GPS data.

[0110] In the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the detailed description of other embodiments above, which will not be repeated here.

[0111] In the implementation, each of the units or structures above can be implemented as an independent entity, or can be combined as the same or several entities. The specific implementation of each unit or structure can be referred to the method embodiments above, which will not be repeated here.

[0112] The specific implementation of each operation above can be referred to the embodiments above, which will not be repeated here.

[0113] The trajectory point optimization method, device, equipment and storage medium provided by the embodiments of the present application are described in detail above, and the principle and implementation mode of the present application are described by applying specific examples. The above embodiment is only used to help understand the method and its core idea of the present application; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range will be changed, and the above description should not be understood as the limitation of the present application.

Claims

1. A trajectory point optimization method, characterized by, The method comprises the following steps: obtaining steering angle data and GPS data of a vehicle; determining valid GPS data in the GPS data according to the steering angle data; generating a corresponding trajectory of the vehicle according to the valid GPS data; wherein the GPS data comprises target GPS data and non-target GPS data, and before the step of determining valid GPS data in the GPS data according to the steering angle data, the method further comprises the following steps: determining a target road corresponding to the target GPS data, and obtaining execution reference information of the target road; wherein the execution reference information is used to determine the number of non-target GPS data required to be reached between two adjacent groups of target GPS data or the interval time required to be reached between two adjacent groups of target GPS data in a plurality of GPS data corresponding to the target road.

2. The trajectory point optimization method of claim 1, wherein, The execution reference information is determined by the following steps: obtaining road labels of a plurality of preset roads; determining the execution reference information of the target road according to the road labels of the plurality of roads and the preset execution reference information corresponding to the road labels; wherein the execution reference information corresponding to at least two road labels is different.

3. The trajectory point optimization method of claim 1, wherein, The steering angle data comprises a steering angle and a steering angle acquisition time corresponding to the steering angle, and the GPS data comprises a coordinate point and a coordinate point acquisition time corresponding to the coordinate point, wherein the step of determining valid GPS data in the GPS data according to the steering angle data comprises: for any target GPS data in the GPS data, obtaining a plurality of steering angle data in a first preset time period covering the coordinate point acquisition time in the target GPS data according to the coordinate point acquisition time in the target GPS data; obtaining an average value of the steering angle in the first preset time period according to a plurality of steering angles corresponding to a plurality of steering angle data; if the average value is greater than a preset steering angle threshold value, determining that the target GPS data is valid GPS data.

4. The trajectory point optimization method of claim 1, wherein, The step of determining valid GPS data in the GPS data according to the steering angle data is performed by a trained neural network model, the steering angle data corresponding to the target GPS data is input into the neural network model, and the judgment result of whether the target GPS data is valid or invalid is output by the neural network model, wherein the training method of the neural network model comprises: obtaining training data, the training data being a plurality of historical steering angle data with valid labels or invalid labels; inputting a plurality of the historical steering angle data with valid labels or invalid labels into the neural network model; calculating a training loss value based on the judgment result of the historical steering angle data with valid labels or invalid labels and the labels corresponding to the historical steering angle data with valid labels or invalid labels, and updating the neural network model according to the training loss value.

5. The trajectory point optimization method of claim 1, wherein, The step of determining valid GPS data in the GPS data according to the steering angle data is performed by a trained neural network model, the steering angle data corresponding to the target GPS data is input into the neural network model, and the valid probability value of the GPS data is output by the neural network model, wherein the training method of the neural network model comprises: Obtain training data, the training data being a plurality of historical steering angle data matrices with different probability value labels, wherein the historical steering angle data matrix with a probability value label includes a plurality of historical steering angle data with valid labels or invalid labels, and the probability value label is the proportion of the historical steering angle data with the valid label in the historical steering angle data matrix; Input a plurality of historical steering angle data matrices with different probability value labels into a probability value generation model; Calculate a training loss value based on the judgment result of the historical steering angle data matrix with a probability value label and the probability value label corresponding to the historical steering angle data matrix with a probability value label, and update the probability value generation model according to the training loss value; If the valid probability value of the target GPS data is greater than a preset valid probability threshold, the target GPS data is determined as the valid GPS data.

6. The trajectory point optimization method of claim 5, wherein, The valid probability threshold has a plurality of preset values, and the valid probability threshold corresponds to the road label of the target road corresponding to the target GPS data and execution reference information; wherein the valid probability threshold is determined according to the road label of the target road.

7. A trajectory point optimization apparatus characterized by comprising: Comprise: Data acquisition module, acquiring steering angle data and GPS data of a vehicle; Data determination module, determining valid GPS data in the GPS data according to the steering angle data; Trajectory generation module, generating a corresponding trajectory of the vehicle according to the valid GPS data; Wherein, the GPS data includes target GPS data and non-target GPS data, the data determination module determines the target road corresponding to the target GPS data and acquires execution reference information of the target road; wherein the execution reference information is used to determine the number of non-target GPS data required to be reached between adjacent two groups of target GPS data in a plurality of GPS data corresponding to the target road or the interval time required to be reached between adjacent two groups of target GPS data.

8. An electronic device, comprising: Comprise: Memory, storing computer readable instructions; Processor, reading computer readable instructions stored in memory, to perform the steps of the method of any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, Computer readable instructions stored thereon, when executed by the processor of the computer, cause the computer to perform the steps of the method of any one of claims 1-6.

Images