Data processing method and computer equipment

Abstract

The invention provides a data processing method and computer equipment, and the method comprises the steps: a terminal obtains a first track dotting data set of a vehicle in a preset time range, and obtains the driving state data of the vehicle in the preset time range; the terminal sends the first track dotting data set to the cloud, so that the cloud performs error correction processing on the first track dotting data set based on a preset error correction strategy to obtain a second track dotting data set; the terminal receives a second track dotting data set sent by the cloud, and verifies the second track dotting data set based on the driving state data to obtain a verification result; and the terminal writes the second track dotting data set into a preset track file under the condition that the verification result indicates that the verification is passed. Based on the scheme, the accuracy of the track dotting data in the track file can be effectively improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular, to a data processing method and a computer device. Background Art

[0002] In the development of autonomous driving technology, trajectory files, consisting of a series of track-point data, are used to guide the training and optimization of autonomous driving algorithms. This track-point data is collected by a terminal deployed on the vehicle during actual driving. Due to limitations in positioning accuracy, the track-point data may not always match the vehicle's actual driving trajectory, making it difficult to guarantee the accuracy of the track-point data in the trajectory file. Summary of the Invention

[0003] The embodiments of the present application provide a data processing method and a computer device.

[0004] A first aspect of an embodiment of the present application provides a data processing method, which is applied to a terminal, the terminal being communicatively connected to a cloud, and the terminal being deployed on a vehicle; the method comprising:

[0005] Obtaining a first trajectory point data set of the vehicle within a preset time range, and obtaining driving status data of the vehicle within the preset time range;

[0006] Sending the first track dot data set to the cloud, so that the cloud performs error correction processing on the first track dot data set based on a preset error correction strategy to obtain a second track dot data set;

[0007] receiving a second track dot data set sent by the cloud, and verifying the second track dot data set based on the driving status data to obtain a verification result;

[0008] When the verification result indicates that the verification is passed, the second track dotting data set is written into a preset track file.

[0009] In an optional embodiment of the present application, the second trajectory dot data set is verified based on the driving status data to obtain a verification result, including: generating a first driving trajectory line based on the driving status data; generating a second driving trajectory line based on the second trajectory dot data set; determining the geometric similarity between the first driving trajectory line and the second driving trajectory line; comparing the geometric similarity with a preset geometric similarity threshold to obtain a verification result, the verification result indicating that the verification is passed when the geometric similarity is not less than the geometric similarity threshold, and the verification result indicating that the verification is failed when the geometric similarity is less than the geometric similarity threshold.

[0010] In an optional embodiment of the present application, generating a first driving trajectory based on driving status data includes: determining a plurality of target time points within a preset time range; determining a driving speed and a driving direction of the vehicle at each of the plurality of target time points based on the driving status data; and determining a vehicle position corresponding to each target time point based on the driving speed and driving direction of the vehicle at each target time point;

[0011] A first driving trajectory line is generated based on the vehicle position corresponding to each target time point.

[0012] In an optional embodiment of the present application, generating a second driving trajectory line based on the second trajectory dot data set includes: determining the position information corresponding to each second trajectory dot data in the second trajectory dot data set; and generating the second driving trajectory line based on the position information corresponding to each second trajectory dot data.

[0013] In an optional embodiment of the present application, determining the geometric similarity between the first driving trajectory line and the second driving trajectory line includes: calculating a geometric distance deviation value between the first driving trajectory line and the second driving trajectory line based on a preset geometric distance algorithm; and normalizing the geometric distance deviation value between the first driving trajectory line and the second driving trajectory line to obtain the geometric similarity between the first driving trajectory line and the second driving trajectory line.

[0014] In an optional embodiment of the present application, after verifying the second trajectory dotting data set based on the driving status data to obtain a verification result, it also includes: when the verification result indicates that the verification fails, the control terminal and / or the vehicle pushes trajectory dotting abnormality information.

[0015] A second aspect of an embodiment of the present application provides a data processing method, which is applied to a cloud, wherein the cloud is in communication with a terminal, and the terminal is deployed in a vehicle; the method includes:

[0016] The first track point data set of the vehicle within a preset time range sent by the receiving terminal;

[0017] Based on a preset error correction strategy, error correction processing is performed on the first track dot data set to obtain a second track dot data set;

[0018] The second trajectory dot data set is sent to the terminal, so that the terminal verifies the second trajectory dot data set based on the driving status data of the vehicle within a preset time range to obtain a verification result, and writes the second trajectory dot data set into a preset trajectory file if the verification result indicates that the verification passes.

[0019] In an optional embodiment of the present application, based on a preset error correction strategy, the first trajectory dot data set is error corrected to obtain a second trajectory dot data set, including: in a preset high-precision map, determining the target road section where each first trajectory dot data in the first trajectory dot data set is located; based on the high-precision map, determining the physical property constraints and traffic rule constraints of the target road section; based on the physical property constraints and traffic rule constraints of the target road section, performing error correction on each first trajectory dot data to obtain second trajectory dot data corresponding to each first trajectory dot data; and determining the second trajectory dot data set based on each second trajectory dot data.

[0020] In an optional embodiment of the present application, based on the physical characteristics constraints and traffic rule constraints of the target road section, each first trajectory dotted data is error-corrected to obtain second trajectory dotted data corresponding to each first trajectory dotted data, including: based on the physical characteristics constraints and traffic rule constraints of the target road section, and the predicted trajectory constraints of associated traffic participants, each first trajectory dotted data is error-corrected to obtain second trajectory dotted data corresponding to each first trajectory dotted data; wherein, the associated traffic participants are traffic participants other than vehicles in the target road section, and the predicted trajectory constraints of the associated traffic participants are determined based on the vehicle's perception data of the associated traffic participants.

[0021] According to a third aspect of an embodiment of the present application, a computer device is provided, comprising: a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any of the above methods when executing the computer program.

[0022] A fourth aspect of the embodiments of the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of any of the above methods when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0024] Figure 1 A schematic diagram of a data processing scenario provided for one embodiment of the present application;

[0025] Figure 2 A flowchart of a data processing method provided in one embodiment of the present application;

[0026] Figure 3 A flowchart of a verification process provided for one embodiment of the present application;

[0027] Figure 4A flowchart for generating a first driving trajectory line provided in one embodiment of the present application;

[0028] Figure 5 A flowchart for generating a second driving trajectory line provided in one embodiment of the present application;

[0029] Figure 6 A flowchart for determining geometric similarity provided in one embodiment of the present application;

[0030] Figure 7 A flowchart of a data processing method provided in one embodiment of the present application;

[0031] Figure 8 A flowchart of an error correction process provided for one embodiment of the present application;

[0032] Figure 9 A schematic diagram of the structure of a data processing device provided in one embodiment of the present application;

[0033] Figure 10 A schematic diagram of the structure of a data processing device provided in one embodiment of the present application;

[0034] Figure 11 A schematic diagram of the computer device structure provided for one embodiment of the present application. DETAILED DESCRIPTION

[0035] The following will clearly and thoroughly describe the technical solutions in this application in conjunction with the accompanying drawings. In the description of the embodiments of this application, unless otherwise specified, " / " means or, for example, A / B can mean A or B: "and / or" in the text is only a description of the association relationship of associated objects, indicating that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more than two.

[0036] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of technical features indicated. Therefore, features defined as "first" or "second" may explicitly or implicitly include one or more of such features.

[0037] See Figure 1 , Figure 1A schematic diagram of a data processing scenario provided for an embodiment of the present application. In the research and development of autonomous driving technology, a trajectory file is a file consisting of a series of trajectory dot data, which is used to guide the training and optimization of autonomous driving algorithms. The trajectory dot data is collected in real time by a terminal deployed on the vehicle during the actual driving process of the vehicle. However, during the data collection process, due to the limitations of the accuracy and stability of the positioning technology, the trajectory dot data sometimes deviates from the actual driving trajectory of the vehicle. Figure 1 As shown, the driving trajectory reflected by the dotted data has a certain deviation from the actual driving trajectory of the vehicle.

[0038] Specifically, during the collection of trajectory dot data, the vehicle's Global Positioning System (GPS) unit can provide key information such as the vehicle's location, speed, and timestamp. In addition, in order to enhance positioning accuracy and stability, other sensors such as the vehicle's inertial navigation system (INS) and wheel speed sensors can be combined as auxiliary to form a multi-sensor fusion positioning solution. However, although these positioning technologies can provide information about the vehicle's location to a certain extent, they each have certain limitations. For example, GPS signals may be affected by factors such as building obstruction, atmospheric interference, or uneven satellite distribution, resulting in reduced positioning accuracy, position offset, or signal loss. Similarly, auxiliary positioning technologies such as inertial navigation systems and wheel speed sensors may also be interfered with by environmental factors, sensor noise, or calibration errors, thereby affecting the accuracy of the data.

[0039] Therefore, in related technologies, it is difficult to guarantee the accuracy of the trajectory dot data in the trajectory file.

[0040] In response to the above-mentioned problems, the main solutions provided by the embodiments of the present application include: the terminal first obtains the first trajectory dot data set of the vehicle within a preset time range, and obtains the driving status data of the vehicle within the preset time range; then, the terminal sends the first trajectory dot data set to the cloud; after the cloud receives the first trajectory dot data set, it performs error correction processing on the first trajectory dot data set based on a preset error correction strategy to obtain a second trajectory dot data set, and sends the second trajectory dot data set to the terminal; accordingly, the terminal receives the second trajectory dot data set sent by the cloud, verifies the second trajectory dot data set based on the driving status data to obtain a verification result. If the verification result indicates that the verification is passed, the terminal writes the second trajectory dot data set to the preset trajectory file. In this way, after error correction in the cloud and verification by the terminal, the second trajectory dot data set is conditionally written to the trajectory file, which can effectively improve the accuracy of the trajectory dot data in the trajectory file.

[0041] based on Figure 1 The following is a schematic diagram of the scene Figure 2 - Figure 8 , the data processing method provided in the embodiments of the present application is introduced in detail.

[0042] See Figure 2 , Figure 2 This is a flow chart of a data processing method provided in one embodiment of the present application. Figure 2 As shown, the method of the embodiment of the present application is applied to a terminal, the terminal is communicatively connected to the cloud, and the terminal is deployed on a vehicle. The method may include the following steps S101-S104.

[0043] S101, obtaining a first track dot data set of a vehicle within a preset time range, and obtaining driving state data of the vehicle within the preset time range.

[0044] Specifically, the terminal involved in this embodiment refers to a device with data collection, processing and communication capabilities. The terminal can obtain the vehicle's trajectory data and driving status data, and transmit and interact with the cloud.

[0045] In one possible implementation, the terminal is deployed in the vehicle, which means that the terminal and the vehicle are connected through relevant communication interfaces to achieve data transmission and interaction, such as Universal Serial Bus (USB) interfaces, wireless network interfaces, etc. In this way, the terminal can be an independent device, such as a vehicle-mounted data recorder. In another possible implementation, the terminal is deployed in the vehicle, which means that the terminal is integrated into the vehicle and is part of the vehicle. For example, the terminal can be the vehicle's own intelligent control system or on-board computer.

[0046] During the driving process of the vehicle, the terminal will obtain the first trajectory dot data set of the vehicle within the preset time range. Among them, the preset time range refers to a specific time period, which can be set according to actual needs, such as one minute, five minutes, etc. The terminal will obtain multiple first trajectory dot data based on the preset time interval within the preset time range. The first trajectory dot data contains information such as the vehicle's position, speed, and timestamp. Then, based on the multiple first trajectory dot data, the terminal will combine and organize them according to the time sequence or other rules to determine the first trajectory dot data set. For example, when the preset time range is five minutes and the preset time interval is to obtain one first trajectory dot data per second, the terminal will obtain 300 first trajectory dot data within these five minutes, and combine these 300 first trajectory dot data into the first trajectory dot data set.

[0047] On the other hand, the terminal also needs to obtain the vehicle's driving status data within a preset time range. The vehicle's driving status data within the preset time range refers to data describing the vehicle's driving conditions within the preset time range. It should be noted that the driving status data includes at least the vehicle's speed and direction at different points in time within the preset time range. Furthermore, the driving status data may also include other driving-related information such as the vehicle's acceleration, steering angle, braking status, and other information at different points in time within the preset time range.

[0048] Therefore, the process of obtaining the vehicle's driving status data within a preset time range can specifically be manifested as the terminal collecting the driving status data in real time within the preset time range through the vehicle's sensors or control system. For example, the terminal can obtain the vehicle's driving speed at different time points within the preset time range through the vehicle's speed sensor; obtain the vehicle's driving direction at different time points within the preset time range through the vehicle's direction sensor; and obtain the vehicle's driving acceleration at different time points within the preset time range through the vehicle's acceleration sensor.

[0049] S102 : Sending the first track dot data set to the cloud, so that the cloud performs error correction processing on the first track dot data set based on a preset error correction strategy to obtain a second track dot data set.

[0050] Specifically, based on the communication connection relationship between the terminal and the cloud, after obtaining the first track marking data set, the terminal can send the first track marking data set to the cloud.

[0051] In one possible implementation, the terminal sends the first trajectory dot data set to the cloud. Specifically, the terminal sends the first trajectory dot data set in the form of a data packet to the cloud server via a wireless network connection, such as 4G, 5G, or Wi-Fi. During this process, the terminal first encapsulates the first trajectory dot data set and adds the necessary communication protocol headers to ensure the integrity and accuracy of the data during transmission. Then, the encapsulated data packet is sent to the cloud server via the wireless network interface. After receiving the data packet, the cloud server decapsulates it and extracts the first trajectory dot data set.

[0052] Correspondingly, after receiving the first track dot data set, the cloud performs error correction processing on the first track dot data set based on a preset error correction strategy to obtain a second track dot data set.

[0053] Specifically, an error correction strategy refers to a pre-defined processing rule or algorithm used to modify the first trajectory point data set to eliminate or reduce potential deviations or errors, thereby generating a more accurate second trajectory point data set. The specific implementation of the error correction strategy can be based on a variety of data sources and / or constraints, including but not limited to high-precision maps, vehicle perception data of associated traffic participants, and physical models.

[0054] In some possible implementations, the error correction strategy can perform error correction processing on the first trajectory dotting data set based on a preset high-precision map. For example, the target road section where each first trajectory dotting data in the first trajectory dotting data set is located is determined in the high-precision map (or HD Map), and each first trajectory dotting data is corrected based on the physical characteristic constraints and traffic rule constraints of the target road section. In another possible implementation, the error correction strategy can further perform error correction processing in combination with the predicted trajectory constraints of the associated traffic participants. For example, based on the physical characteristic constraints and traffic rule constraints of the target road section, as well as the predicted trajectory constraints of the associated traffic participants, each first trajectory dotting data is corrected. In yet another possible implementation, the error correction strategy can correct each first trajectory dotting data based on the physical characteristic constraints, traffic rule constraints and predicted trajectory constraints of the associated traffic participants of the high-precision map, combined with a preset statistical model or machine learning model.

[0055] High-precision maps are a type of prior data used for autonomous driving, representing map elements such as roads, lanes, roadside traffic signs, and ground markings. The cloud performs error correction on the first trajectory dot data set based on the high-precision map. This means that the cloud matches each first trajectory dot data in the first trajectory dot data set with the high-precision map to check whether it complies with the map's physical characteristics, traffic rules, and other constraints. If a first trajectory dot data set is found to not comply with the high-precision map's constraints, the cloud will correct it based on the high-precision map's information to obtain accurate second trajectory dot data. Of course, if a first trajectory dot data set is found to comply with the high-precision map's constraints, the cloud will determine it as the second trajectory dot data set, without modifying the first trajectory dot data set. After the cloud performs error correction on each first trajectory dot data set, the corresponding second trajectory dot data set is obtained. Finally, the second trajectory dot data set is determined based on each second trajectory dot data set.

[0056] In another possible implementation, the cloud performs error correction processing on the first trajectory dot data set based on the high-precision map and the predicted trajectory constraints of the associated traffic participants. Specifically, the cloud will match each first trajectory dot data in the first trajectory dot data set with the high-precision map, and at the same time, combined with the predicted trajectory constraints of the associated traffic participants, check whether it conforms to the physical characteristics, traffic rules and predicted trajectory constraints of the associated traffic participants in the map. If it is found that a certain first trajectory dot data does not meet the above constraints, the cloud will correct the first trajectory dot data based on the information of the high-precision map and the predicted trajectory constraints of the associated traffic participants to obtain more accurate second trajectory dot data. If it is found that a certain first trajectory dot data meets the above constraints, the cloud will determine the first trajectory dot data as the second trajectory dot data without modifying the first trajectory dot number. After the cloud performs error correction processing on each first trajectory dot data, the second trajectory dot data corresponding to each first trajectory dot data is obtained, and finally the second trajectory dot data set is determined based on each second trajectory dot data.

[0057] In another possible implementation, the cloud performs error correction on the first trajectory dot data set based on the physical characteristics constraints of the high-precision map, traffic rules constraints, and predicted trajectory constraints of associated traffic participants, combined with a preset statistical model or machine learning model. Specifically, the cloud will match each first trajectory dot data in the first trajectory dot data set with the high-precision map, and at the same time, combine it with the predicted trajectory constraints of the associated traffic participants, and use a preset statistical model or machine learning model to check whether it complies with the physical characteristics, traffic rules, and predicted trajectory constraints of the associated traffic participants in the map. If it is found that a certain first trajectory dot data does not meet the above constraints, the cloud will correct the first trajectory dot data based on the information of the high-precision map, the predicted trajectory constraints of the associated traffic participants, and the output results of the statistical model or machine learning model to obtain more accurate second trajectory dot data. If it is found that a certain first trajectory dot data meets the above constraints, the cloud will determine the first trajectory dot data as the second trajectory dot data, and there is no need to modify the first trajectory dot data. After error correction processing is performed on each first track dot data in the cloud, the second track dot data corresponding to each first track dot data is obtained, and finally the second track dot data set is determined based on each second track dot data.

[0058] It is understandable that during the above-mentioned error correction process, the cloud may modify some of the first trajectory dot data in the first trajectory dot data set, which indicates that there are deviations or errors in these first trajectory dot data. Through correction by the high-precision map, the predicted trajectory constraints of the associated traffic participants, and the statistical model or machine learning model, it can be made to more accurately reflect the actual driving trajectory of the vehicle. During the above-mentioned error correction process, the cloud may also not modify the first trajectory dot data in the first trajectory dot data set, which indicates that these first trajectory dot data are already sufficiently accurate and consistent with the information in the high-precision map, the predicted trajectory constraints of the associated traffic participants, and the output results of the statistical model or machine learning model, and no further correction is required.

[0059] S103: Receive a second track dot data set sent by the cloud, and verify the second track dot data set based on the driving status data to obtain a verification result.

[0060] Specifically, after obtaining the second track dot data set, the cloud sends the second track dot data set to the terminal based on the communication connection relationship between the cloud and the terminal. Correspondingly, the terminal receives the second track dot data set.

[0061] Furthermore, the second track dotted data set is verified based on the vehicle's driving status data within a preset time range to obtain a verification result. The driving status data reflects the vehicle's actual driving conditions within the preset time range, including driving speed, driving direction, etc. Therefore, the second track dotted data set can be verified to obtain a verification result by comparing the driving status data with corresponding information in the second track dotted data set.

[0062] It is understood that the verification result may indicate a pass or a fail. If the verification passes, it indicates that the second track dot data set is highly consistent with the driving status data and the data can be considered accurate. If the verification fails, it indicates that there are significant differences between the second track dot data set and the driving status data, and the second track dot data set may need to be further processed or discarded.

[0063] In one possible implementation, a first driving trajectory can be generated based on driving state data, and a second driving trajectory can be generated based on a second trajectory dotted data set. A verification result is then determined based on the geometric similarity between the first and second driving trajectory lines. Geometric similarity can be measured by calculating the distance, angular deviation, or other geometric features between the two driving trajectory lines. This verification result represents the degree of consistency between the second trajectory dotted data set and the vehicle's actual driving trajectory. A higher geometric similarity indicates a higher data accuracy.

[0064] In one possible implementation, it is possible to check whether each second trajectory dot data in the second trajectory dot data set is consistent with the record of the driving status data at the corresponding time point, such as whether the speed, direction, etc. match, and then determine the verification result based on the matching result of each second trajectory dot data.

[0065] In one possible implementation, a statistical analysis can be performed on the driving state data and the second track dot data set to calculate statistical indicators such as the correlation coefficient or error distribution between the two. If the statistical indicators show a high correlation or a small error distribution, the verification is considered passed; otherwise, the verification is considered failed.

[0066] In one possible implementation, an initial machine learning model can be trained based on multiple sample pairs to obtain a target machine learning model. Any target sample pair among the multiple sample pairs includes a driving state data sample and a trajectory mark data set sample within the same time range, and the target sample pair is annotated with a similarity score between the driving state data sample and the trajectory mark data set sample. That is, during the training process, the initial machine learning model learns the association between the driving state data samples, the trajectory mark data set samples, and the similarity scores, enabling the target machine learning model to receive the driving state data and the trajectory mark data set as input and output a similarity score. In actual applications, the driving state data and the second trajectory mark data set are input into the target machine learning model to obtain a similarity score between the driving state data and the second trajectory mark data set, output by the target machine learning model. A verification result is then determined based on the similarity score. If the similarity score between the driving state data and the second trajectory mark data set is not less than a preset similarity score threshold, the verification result indicates a passed verification; conversely, if the similarity score between the driving state data and the second trajectory mark data set is less than the preset similarity score threshold, the verification result indicates a failed verification.

[0067] S104 : When the verification result indicates that the verification is passed, the second track dotting data set is written into a preset track file.

[0068] Specifically, if the verification result indicates a pass, then the second trajectory dot data set is highly consistent with the driving status data and has high data accuracy. This data can be considered reliable and valid, reflecting the vehicle's actual driving trajectory. In this case, the second trajectory dot data set can be written to a preset trajectory file. A trajectory file is a file pre-set on the terminal / cloud / vehicle for storing trajectory dot data. This trajectory file has a specific format, storage path, and naming convention.

[0069] The second track dot data set is written into a preset track file, specifically, each second track dot data in the second track dot data set is written into the track file according to a preset format and sequence.

[0070] In one possible implementation, the trajectory file is in Comma-Separated Values ​​(CSV) format. This format uses commas to separate each trajectory point data, and the entire trajectory file is stored in plain text. The CSV format also offers advantages such as a simple structure, convenient data exchange, and strong compatibility. For trajectory point data, the CSV format clearly displays the various parameters of each trajectory point data, facilitating subsequent data processing and analysis.

[0071] In this embodiment, the terminal first obtains a first trajectory dot data set of the vehicle within a preset time range, and obtains the driving status data of the vehicle within the preset time range; then, the terminal sends the first trajectory dot data set to the cloud; after the cloud receives the first trajectory dot data set, it performs error correction processing on the first trajectory dot data set based on a preset error correction strategy to obtain a second trajectory dot data set, and sends the second trajectory dot data set to the terminal; accordingly, the terminal receives the second trajectory dot data set sent by the cloud, verifies the second trajectory dot data set based on the driving status data, and obtains a verification result. If the verification result indicates that the verification is passed, the terminal writes the second trajectory dot data set to the preset trajectory file. In this way, after error correction in the cloud and verification by the terminal, the second trajectory dot data set is conditionally written to the trajectory file, which can effectively improve the accuracy of the trajectory dot data in the trajectory file.

[0072] In one embodiment, based on the above Figure 2 In the embodiment shown, the following steps may be further performed after step S103:

[0073] When the verification result indicates that the verification fails, the control terminal and / or the vehicle pushes the track marking abnormality information.

[0074] Specifically, if a large amount of abnormal track dotting data is written into the track file, it will make it difficult to modify the track file later, or even make the track file completely unusable. To this end, this embodiment proposes to push track dotting abnormality information when the track dotting data is abnormal, to prevent abnormal track dotting data from being continuously written into the track file.

[0075] If the verification result indicates a failure, this indicates a significant discrepancy between the second track dot data set and the driving status data, raising concerns about data accuracy and potentially failing to reflect the vehicle's actual driving trajectory. In this case, the terminal and / or vehicle can be controlled to push a track dot anomaly message. This track dot anomaly message indicates a problem with the track dot data and requires further inspection or processing. This track dot anomaly message can include at least one of a track dot anomaly text prompt, a track dot anomaly video prompt, a track dot anomaly image prompt, and a track dot anomaly audio prompt.

[0076] In one possible implementation, the terminal can be controlled to push information about abnormal tracking and marking. For example, the terminal's display component can be controlled to display at least one of a text message, a video message, or an image message indicating abnormal tracking and marking; or the terminal's audio component can be controlled to play an audio message indicating abnormal tracking and marking.

[0077] In one possible implementation, the vehicle can be controlled to push track marking anomaly information. For example, the vehicle's display component can be controlled to display at least one of a track marking anomaly prompt text, a track marking anomaly prompt video, or a track marking anomaly prompt image; or the vehicle's audio component can be controlled to play a track marking anomaly prompt audio.

[0078] In a possible implementation, the terminal and the vehicle can be controlled to push the track marking abnormality information at the same time.

[0079] In this embodiment, when the verification result indicates that the verification has failed, the control terminal and / or the vehicle pushes the track dotting abnormality information, so that the driver or system maintenance personnel can quickly understand the abnormality of the track dotting data and take corresponding measures, such as re-collecting data, checking the sensor status or adjusting the data processing algorithm, etc., to avoid the problem that the track file is difficult to correct or even unusable at a later time.

[0080] See Figure 3 , Figure 3 A flow chart of a verification process provided in one embodiment of the present application is as follows: Figure 3 As shown, the method of the embodiment of the present application may include the following steps S201-S204, and steps S201-S204 may be used as Figure 2 The illustrated embodiment is a detailed step of "verifying the second track dot data set based on the driving state data to obtain a verification result" in step S103.

[0081] S201, generating a first driving trajectory line based on driving state data;

[0082] S202, generating a second driving trajectory line based on the second trajectory dot data set;

[0083] S203, determining the geometric similarity between the first driving trajectory line and the second driving trajectory line;

[0084] S204 , comparing the geometric similarity with a preset geometric similarity threshold to obtain a verification result. If the geometric similarity is not less than the geometric similarity threshold, the verification result indicates that the verification is passed; if the geometric similarity is less than the geometric similarity threshold, the verification result indicates that the verification is failed.

[0085] Specifically, this embodiment verifies the second track dot data set from the aspect of geometric features to determine whether the second track dot data set conforms to the actual driving state of the vehicle.

[0086] First, a first driving trajectory needs to be generated based on the driving status data. Specifically, this process involves converting the vehicle's driving status data within a preset time range, including speed and direction at different time points, into a continuous trajectory, i.e., the first driving trajectory, using an algorithm or mathematical model.

[0087] Then, a second driving trajectory line needs to be generated based on the second trajectory dot data set. The specific process is: each second trajectory dot data in the second trajectory dot data set is connected in chronological order to form a continuous trajectory line, namely the second driving trajectory line.

[0088] Furthermore, the geometric similarity between the first and second driving trajectories is determined. Geometric similarity refers to the degree of similarity between the two driving trajectories in terms of shape, direction, position, etc. Various algorithms can be used to determine geometric similarity, such as calculating the distance, angular deviation, and curvature similarity between the two driving trajectories to comprehensively evaluate the geometric similarity between the first and second driving trajectories.

[0089] It can be understood that geometric similarity is a numerical value that represents the degree of similarity between two driving trajectories. A higher geometric similarity indicates a greater similarity between the first and second driving trajectories in terms of shape, direction, and position; a lower geometric similarity indicates a lower similarity. Furthermore, a preset geometric similarity threshold is introduced. This threshold represents an acceptable range of similarity between two driving trajectories and is used to determine whether the geometric similarity reaches an acceptable level.

[0090] It can be understood that if the geometric similarity is not less than the geometric similarity threshold, it can be considered that the degree of consistency between the second trajectory dot data set and the actual driving trajectory of the vehicle is acceptable, and the verification result indicates that the verification is passed; if the geometric similarity is less than the geometric similarity threshold, it can be considered that the degree of consistency between the second trajectory dot data set and the actual driving trajectory of the vehicle is unacceptable, and the verification result indicates that the verification is failed.

[0091] In this embodiment, by generating a first driving trajectory line and a second driving trajectory line and calculating the geometric similarity between them, the consistency of the second trajectory dot data set with the vehicle's actual driving trajectory can be effectively assessed. Comparing the geometric similarity with a preset geometric similarity threshold can quantitatively determine the accuracy of the second trajectory dot data set, ensuring that the second trajectory dot data set is only written to the trajectory file when it is sufficiently accurate. This effectively improves the accuracy of the trajectory dot data in the trajectory file.

[0092] See Figure 4 , Figure 4 A schematic diagram of a process for generating a first driving trajectory line is provided in an embodiment of the present application, such as Figure 4 As shown, the method of the embodiment of the present application may include the following steps S301-S304, and steps S301-S304 may be used as Figure 3 The detailed steps of step S201 of the embodiment are shown.

[0093] S301, determining multiple target time points within a preset time range;

[0094] S302, determining a driving speed and a driving direction of the vehicle at each of a plurality of target time points based on the driving state data;

[0095] S303, determining the vehicle position corresponding to each target time point based on the vehicle's driving speed and driving direction at each target time point;

[0096] S304: Generate a first driving trajectory line based on the vehicle position corresponding to each target time point.

[0097] Specifically, a target time point refers to a specific time point selected within a preset time range. Multiple target time points may or may not be equally spaced. Taking equally spaced target time points as an example, multiple target time points may be determined within the preset time range based on uniform sampling. For example, uniform sampling may select one target time point every second.

[0098] Furthermore, the vehicle's speed and direction at each of the multiple target time points are determined based on the driving status data. In some cases, the driving status data may directly record the vehicle's speed and direction at each target time point; in other cases, the driving status data may record the vehicle's speed and direction at other time points within a preset time range. Based on the vehicle's speed and direction at other time points, and using an interpolation algorithm or a physical model (such as a uniformly accelerated linear motion model), the vehicle's speed and direction at each target time point are inferred.

[0099] It should be noted that the above driving speed is expressed in units such as "meters per second (m / s)" or "kilometers per hour (km / h)", and the driving direction is expressed in degrees (°) or a direction vector. For example, the driving direction can be the angle relative to true north, measured in degrees, such as "30°" for northeast; or it can be expressed as a direction vector on a two-dimensional plane, such as "(0.707,0.707)" for northeast.

[0100] Furthermore, based on the vehicle's speed and direction at each target time point, the vehicle position corresponding to each target time point is determined. In one possible implementation, a virtual position can be assigned to the first target time point within a preset time range as a starting point, and then based on the vehicle's speed and direction at that target time point, the vehicle position at the next target time point is calculated, and so on, to determine the vehicle position corresponding to all target time points. In another possible implementation, based on the vehicle's speed and direction at each target time point, combined with a physical model (such as a uniformly accelerated linear motion model or a uniformly accelerated curved motion model), the specific motion trajectory of the vehicle between each target time point can be calculated, thereby determining the vehicle position corresponding to each target time point.

[0101] Furthermore, a first driving trajectory is generated based on the vehicle positions corresponding to each target time point. In one possible implementation, the vehicle positions corresponding to each target time point can be sequentially connected to form a continuous trajectory, i.e., the first driving trajectory. In another possible implementation, a smooth trajectory can be generated based on the vehicle positions corresponding to each target time point using an interpolation algorithm or a curve fitting algorithm, and used as the first driving trajectory.

[0102] In one possible implementation, a machine learning algorithm can be combined to generate the first driving trajectory. For example, a supervised learning approach can be used, using historical driving trajectory data as training samples to train a machine learning model to predict the vehicle position corresponding to each target time point. New driving status data is then input into the trained model to predict the vehicle position corresponding to each target time point. Furthermore, the vehicle positions corresponding to each target time point are sequentially connected to form a continuous trajectory, i.e., the first driving trajectory. Alternatively, an interpolation algorithm or curve fitting algorithm can be used based on the vehicle positions corresponding to each target time point to generate a smooth trajectory, which serves as the first driving trajectory.

[0103] In this embodiment, multiple target time points are determined within a preset time range. Based on the vehicle's speed and direction at each target time point, the vehicle's position corresponding to each target time point is determined, thereby generating a first driving trajectory. This approach accurately reproduces the vehicle's motion trajectory during actual driving. Furthermore, by combining physical models with interpolation or machine learning algorithms, the accuracy of the first driving trajectory is improved while also making the generated first driving trajectory smoother and more continuous, thereby better reflecting the vehicle's actual driving state.

[0104] See Figure 5 , Figure 5 A schematic diagram of a process for generating a second driving trajectory line is provided in an embodiment of the present application, such as Figure 5 As shown, the method of the embodiment of the present application may include the following steps S401-S402, and steps S401-S402 may be used as Figure 3 The detailed steps of step S202 of the embodiment are shown.

[0105] S401, determining position information corresponding to each second track dot data in the second track dot data set;

[0106] S402 : Generate a second driving trajectory line based on the position information corresponding to each second trajectory dot data.

[0107] Specifically, location information is a fundamental attribute of track dot data. It can be expressed as geographic coordinates such as longitude, latitude, and altitude (optional), describing the vehicle's spatial position at a specific point in time. Therefore, determining the location information corresponding to each second track dot data set in the second track dot data set actually involves extracting the geographic coordinate information contained in each second track dot data set, namely, longitude, latitude, and altitude (optional).

[0108] Furthermore, based on the position information corresponding to each second trajectory dot data, a second driving trajectory line is generated. Regarding this process, in one possible implementation, the position points corresponding to each second trajectory dot data can be determined based on the position information corresponding to each second trajectory dot data; and based on the time sequence of each second trajectory dot data, the position points corresponding to each second trajectory dot data can be connected by a straight line or a simple curve (such as a broken line) to form a continuous second driving trajectory line.

[0109] In a possible implementation, after obtaining the second driving trajectory, algorithms such as Kalman filtering, particle filtering, or Bezier curve fitting may be used to smooth the second driving trajectory to reduce the influence of noise and error.

[0110] In this embodiment, by determining the position information corresponding to each second trajectory dot data point, this position information can be directly used to generate the second driving trajectory line, ensuring that the second driving trajectory line retains the accuracy of the second trajectory dot data point. Furthermore, because the position information is directly used, complex trajectory fitting or interpolation processing is not required, thereby improving the efficiency and accuracy of the second driving trajectory line generation.

[0111] See Figure 6 , Figure 6 A schematic diagram of a process for determining geometric similarity is provided in one embodiment of the present application, such as Figure 6 As shown, the method of the embodiment of the present application may include the following steps S501-S502, and steps S501-S502 may be used as Figure 3 The detailed steps of step S203 of the embodiment are shown.

[0112] S501, calculating a geometric distance deviation value between a first driving trajectory line and a second driving trajectory line based on a preset geometric distance algorithm;

[0113] S502 : Normalize the geometric distance deviation between the first driving trajectory line and the second driving trajectory line to obtain the geometric similarity between the first driving trajectory line and the second driving trajectory line.

[0114] Specifically, the preset geometric distance algorithm refers to a mathematical algorithm used to measure the geometric differences between two driving trajectory lines. The geometric distance algorithm can comprehensively consider the shape, direction and position of the trajectory lines and evaluate the similarity by calculating the minimum distance or maximum deviation value between them.

[0115] It should be noted that the first and second driving trajectories may not be uniform in spatial scale. In this case, they must first be spatially aligned or normalized to ensure they are comparable at the same scale. For example, operations such as translation, rotation, or scaling can be performed to spatially align the starting and ending points of the first and second driving trajectories, or to normalize their coordinates to the same reference frame.

[0116] Regarding the process of calculating the geometric distance deviation value between the first driving trajectory line and the second driving trajectory line based on the preset geometric distance algorithm, it can be specifically expressed as follows: for each point on the first driving trajectory line, a point closest to it is found on the second driving trajectory line, and then the distance between these point pairs is calculated, and the maximum value, minimum value or average value is taken as the geometric distance deviation value.

[0117] In one possible implementation, the geometric distance algorithm can employ the Fréchet distance algorithm. The Fréchet distance is a metric specifically designed to measure the similarity between two curves. Its core consideration is the maximum deviation between the curves, rather than relying solely on the simple Euclidean distance. Specifically, the first and second driving trajectories are first discretized, with the first trajectory converted into a first set of discrete points, and the second trajectory converted into a second set of discrete points. Next, the Fréchet distance algorithm is used to calculate the maximum deviation between the first and second discrete point sets. The Fréchet distance algorithm compares each point in the first and second discrete point sets to determine the maximum deviation between them. This maximum deviation is the Fréchet distance. Finally, this maximum deviation is determined as the geometric distance deviation between the first and second trajectories. This geometric distance deviation reflects the degree of geometric similarity between the two trajectories. A smaller geometric distance deviation indicates greater similarity between the two trajectories, while a larger geometric distance deviation indicates greater difference between the two trajectories.

[0118] To more intuitively understand the process of calculating the geometric distance deviation between the first driving trajectory and the second driving trajectory based on the Fréchet distance algorithm, a simple example is provided below:

[0119] Assume that the first and second driving trajectories are discretized and converted into a series of discrete points. Specifically, the first driving trajectory is converted into a first discrete point set, which includes five discrete points, denoted as A1, A2, A3, A4, and A5. The second driving trajectory is converted into a second discrete point set, which includes five discrete points, denoted as B1, B2, B3, B4, and B5.

[0120] Next, the Fréchet distance algorithm is used to calculate the maximum deviation between the first and second discrete point sets. The points in the first and second discrete point sets are compared one by one to find the maximum deviation between them. For example, the distance between A1 and B1 is compared, the distance between A2 and B2, and so on. Variations in the distances between adjacent points are also considered to ensure the comprehensiveness and accuracy of the calculation. Finally, a maximum deviation value, the Fréchet distance, is obtained. This value reflects the degree of geometric similarity between the first and second driving trajectories. Assuming the maximum deviation value is d, the smaller d is, the more similar the first and second driving trajectories are; the larger d is, the greater the difference between the first and second driving trajectories.

[0121] Furthermore, the geometric distance deviation value between the first driving trajectory line and the second driving trajectory line is normalized to obtain the geometric similarity between the first driving trajectory line and the second driving trajectory line. The purpose of the normalization process is to convert the geometric distance deviation value into a value between 0 and 1, so that the geometric similarity is more comparable and universal. Specifically, linear normalization, logarithmic normalization or other normalization methods can be used to map the geometric distance deviation value to the interval [0,1] to obtain the geometric similarity. Among them, the closer the geometric similarity is to 1, the more similar the two driving trajectory lines are geometrically; the closer the geometric similarity is to 0, the greater the geometric difference between the two driving trajectory lines. It can be understood that if the above-mentioned geometric distance algorithm is the Fréchet distance algorithm, then the "geometric distance deviation value between the first driving trajectory line and the second driving trajectory line" is negatively correlated with the "geometric similarity between the first driving trajectory line and the second driving trajectory line".

[0122] In this embodiment, a preset geometric distance algorithm is used to calculate the geometric distance deviation between the first and second driving trajectories, enabling a quantitative assessment of the differences in geometric characteristics between the two trajectories. This geometric distance deviation is then normalized and converted into geometric similarity, making the similarity between different trajectories comparable and universal. This effectively improves the accuracy and reliability of the verification of the second trajectory dot data set.

[0123] See Figure 7 , Figure 7 A flow chart of a data processing method provided in one embodiment of the present application is shown as follows: Figure 7 As shown, the method of the embodiment of the present application is applied to the cloud, the cloud is communicated with the terminal, and the terminal is deployed on the vehicle. The method may include the following steps S601-S603.

[0124] S601: receiving a first track dot data set of a vehicle within a preset time range sent by a terminal.

[0125] Specifically, the present embodiment involves the communication connection relationship between the terminal, the vehicle, the cloud and the terminal, and the first track dot data set of the vehicle within a preset time range. Figure 2 The relevant description has been made in the illustrated embodiment and will not be repeated here.

[0126] Specifically, the cloud referred to in this embodiment refers to a remote server or server cluster with computing and storage capabilities, which communicates with the terminal via the internet. The cloud runs a corresponding operating system and database management system, capable of processing and analyzing the first trajectory dotting data set sent by the terminal. For example, the cloud can be a single remote server; in another example, the cloud can be a server cluster consisting of multiple computing nodes, where the multiple computing nodes are connected via a high-speed network and collectively provide data processing and storage services to the terminal.

[0127] After obtaining the first track data set of the vehicle within the preset time range, the terminal sends the first track data set to the cloud. Correspondingly, the cloud receives the first track data set.

[0128] S602 : Based on a preset error correction strategy, error correction is performed on the first track dot data set to obtain a second track dot data set.

[0129] Specifically, after receiving the first track dot data set, the cloud performs error correction on the first track dot data set based on a preset error correction strategy to obtain a second track dot data set.

[0130] An error correction strategy is a pre-defined processing rule or algorithm used to modify the first trajectory point data set to eliminate or reduce potential deviations or errors, thereby generating a more accurate second trajectory point data set. The specific implementation of the error correction strategy can be based on a variety of data sources and / or constraints, including but not limited to high-precision maps, vehicle perception data of associated traffic participants, and physical models.

[0131] In some possible implementations, the error correction strategy can perform error correction processing on the first trajectory dotting data set based on a preset high-precision map. For example, the target road section where each first trajectory dotting data in the first trajectory dotting data set is located is determined in the high-precision map (or HD Map), and each first trajectory dotting data is corrected based on the physical characteristic constraints and traffic rule constraints of the target road section. In another possible implementation, the error correction strategy can further perform error correction processing in combination with the predicted trajectory constraints of the associated traffic participants. For example, based on the physical characteristic constraints and traffic rule constraints of the target road section, as well as the predicted trajectory constraints of the associated traffic participants, each first trajectory dotting data is corrected. In yet another possible implementation, the error correction strategy can correct each first trajectory dotting data based on the physical characteristic constraints, traffic rule constraints and predicted trajectory constraints of the associated traffic participants of the high-precision map, combined with a preset statistical model or machine learning model.

[0132] High-precision maps are a type of prior data used for autonomous driving, representing map elements such as roads, lanes, roadside traffic signs, and ground markings. The cloud performs error correction on the first trajectory dot data set based on the high-precision map. This means that the cloud matches each first trajectory dot data in the first trajectory dot data set with the high-precision map to check whether it complies with the map's physical characteristics, traffic rules, and other constraints. If a first trajectory dot data set is found to not comply with the high-precision map's constraints, the cloud will correct it based on the high-precision map's information to obtain accurate second trajectory dot data. Of course, if a first trajectory dot data set is found to comply with the high-precision map's constraints, the cloud will determine it as the second trajectory dot data set, without modifying the first trajectory dot data set. After the cloud performs error correction on each first trajectory dot data set, the corresponding second trajectory dot data set is obtained. Finally, the second trajectory dot data set is determined based on each second trajectory dot data set.

[0133] In another possible implementation, the cloud performs error correction processing on the first trajectory dot data set based on the high-precision map and the predicted trajectory constraints of the associated traffic participants. Specifically, the cloud will match each first trajectory dot data in the first trajectory dot data set with the high-precision map, and at the same time, combined with the predicted trajectory constraints of the associated traffic participants, check whether it conforms to the physical characteristics, traffic rules and predicted trajectory constraints of the associated traffic participants in the map. If it is found that a certain first trajectory dot data does not meet the above constraints, the cloud will correct the first trajectory dot data based on the information of the high-precision map and the predicted trajectory constraints of the associated traffic participants to obtain more accurate second trajectory dot data. If it is found that a certain first trajectory dot data meets the above constraints, the cloud will determine the first trajectory dot data as the second trajectory dot data without modifying the first trajectory dot number. After the cloud performs error correction processing on each first trajectory dot data, the second trajectory dot data corresponding to each first trajectory dot data is obtained, and finally the second trajectory dot data set is determined based on each second trajectory dot data.

[0134] In another possible implementation, the cloud performs error correction on the first trajectory dot data set based on the physical characteristics constraints of the high-precision map, traffic rules constraints, and predicted trajectory constraints of associated traffic participants, combined with a preset statistical model or machine learning model. Specifically, the cloud will match each first trajectory dot data in the first trajectory dot data set with the high-precision map, and at the same time, combine it with the predicted trajectory constraints of the associated traffic participants, and use a preset statistical model or machine learning model to check whether it complies with the physical characteristics, traffic rules, and predicted trajectory constraints of the associated traffic participants in the map. If it is found that a certain first trajectory dot data does not meet the above constraints, the cloud will correct the first trajectory dot data based on the information of the high-precision map, the predicted trajectory constraints of the associated traffic participants, and the output results of the statistical model or machine learning model to obtain more accurate second trajectory dot data. If it is found that a certain first trajectory dot data meets the above constraints, the cloud will determine the first trajectory dot data as the second trajectory dot data, and there is no need to modify the first trajectory dot data. After error correction processing is performed on each first track dot data in the cloud, the second track dot data corresponding to each first track dot data is obtained, and finally the second track dot data set is determined based on each second track dot data.

[0135] It is understandable that during the above-mentioned error correction process, the cloud may modify some of the first trajectory dot data in the first trajectory dot data set, which indicates that there are deviations or errors in these first trajectory dot data. Through correction by the high-precision map, the predicted trajectory constraints of the associated traffic participants, and the statistical model or machine learning model, it can be made to more accurately reflect the actual driving trajectory of the vehicle. During the above-mentioned error correction process, the cloud may also not modify the first trajectory dot data in the first trajectory dot data set, which indicates that these first trajectory dot data are already sufficiently accurate and consistent with the information in the high-precision map, the predicted trajectory constraints of the associated traffic participants, and the output results of the statistical model or machine learning model, and no further correction is required.

[0136] S603: Send the second trajectory dot data set to the terminal, so that the terminal can verify the second trajectory dot data set based on the driving status data of the vehicle within a preset time range to obtain a verification result, and write the second trajectory dot data set into a preset trajectory file if the verification result indicates that the verification is passed.

[0137] Specifically, after obtaining the second track dot data set, the cloud will send the second track dot data set to the terminal. Correspondingly, the terminal receives the second track dot data set sent by the cloud, verifies the second track dot data set based on the driving status data, and obtains a verification result; if the verification result indicates that the verification is passed, the terminal writes the second track dot data set into the preset track file. Regarding how the terminal processes the second track dot data set, Figure 2 This is clearly described in the illustrated embodiment and will not be described in detail here.

[0138] In this embodiment, the terminal first obtains a first trajectory dot data set of the vehicle within a preset time range, and obtains the driving status data of the vehicle within the preset time range; then, the terminal sends the first trajectory dot data set to the cloud; after the cloud receives the first trajectory dot data set, it performs error correction processing on the first trajectory dot data set based on a preset error correction strategy to obtain a second trajectory dot data set, and sends the second trajectory dot data set to the terminal; accordingly, the terminal receives the second trajectory dot data set sent by the cloud, verifies the second trajectory dot data set based on the driving status data, and obtains a verification result. If the verification result indicates that the verification is passed, the terminal writes the second trajectory dot data set to the preset trajectory file. In this way, after error correction in the cloud and verification by the terminal, the second trajectory dot data set is conditionally written to the trajectory file, which can effectively improve the accuracy of the trajectory dot data in the trajectory file.

[0139] See Figure 8 , Figure 8 A flowchart of an error correction process provided by an embodiment of the present application is as follows: Figure 8 As shown, the method of the embodiment of the present application may include the following steps S701-S704, and steps S701-S704 may be used as Figure 7 The detailed steps of step S602 in the embodiment are shown.

[0140] S701, determining a target road section where each first track dot data in the first track dot data set is located in a preset high-precision map;

[0141] S702, determining the physical property constraints and traffic rule constraints of the target road section based on the high-precision map;

[0142] S703, based on the physical property constraints and traffic rule constraints of the target road segment, performing error correction processing on each first track dot data to obtain second track dot data corresponding to each first track dot data;

[0143] S704 : Determine a second track dotting data set based on each second track dotting data.

[0144] Specifically, because high-precision maps are detailed and accurate map data that includes a rich set of map elements, such as roads, lanes, roadside traffic signs, and ground markings, and the first-track dot data includes location information, the target road section where each first-track dot data point in the first-track dot data set is located can be determined in the high-precision map.

[0145] Furthermore, regarding the process of determining the physical characteristic constraints and traffic rule constraints of the target road section based on the HD map, the specific process is as follows: First, the road data in the HD map is parsed to extract the physical characteristic information of the target road section, such as the road width, curvature, slope, road surface material, etc. This information together constitutes the physical characteristic constraints of the target road section. The physical characteristic constraints characterize the characteristics of the target road section in terms of geometry, road conditions, etc., and have a direct impact on the vehicle's driving trajectory. Secondly, the traffic rule data in the HD map is parsed to extract the traffic rule information of the target road section, such as speed limit, one-way street, no turn, traffic light signal, etc. This information together constitutes the traffic rule constraints of the target road section. Traffic rule constraints characterize the requirements of the target road section in terms of traffic laws and driving rules, and also have a significant impact on the vehicle's driving trajectory.

[0146] After determining the physical characteristic constraints and traffic rule constraints of the target road section, the cloud can perform error correction processing on each first trajectory dot data based on the physical characteristic constraints and traffic rule constraints. The specific process of error correction processing includes: matching each first trajectory dot data with the target road section in the high-precision map to check whether it meets the physical characteristic constraints and traffic rule constraints; if it is found that a first trajectory dot data does not meet the constraints, the cloud will correct the first trajectory dot data according to the information of the high-precision map, such as adjusting its position, speed or direction and other parameters to make it meet the physical characteristic constraints and traffic rule constraints; the corrected first trajectory dot data becomes the corresponding second trajectory dot data. After error correction processing is performed on all first trajectory dot data, the second trajectory dot data corresponding to each first trajectory dot data is obtained, and the first trajectory dot data and the second trajectory dot data are equal in quantity. Further, the cloud determines the second trajectory dot data set based on each second trajectory dot data. It can be understood that each second track dot data in the second track dot data set has been error-corrected by the high-precision map and is therefore closer to the actual driving trajectory of the vehicle.

[0147] In this embodiment, the first trajectory dot data set is corrected based on a preset high-precision map. By determining the target road section where each first trajectory dot data is located and performing error correction based on the physical characteristics constraints and traffic rules constraints of the high-precision map, the accuracy and reliability of the second trajectory dot data can be significantly improved, effectively avoiding trajectory deviations caused by the limitations of positioning technology, and ensuring that the second trajectory dot data is more consistent with actual road conditions.

[0148] In one embodiment, Figure 8 Step S703 of the illustrated embodiment may be further refined to include the following steps:

[0149] Based on the physical property constraints and traffic rule constraints of the target road section, as well as the predicted trajectory constraints of the associated traffic participants, error correction processing is performed on each first trajectory dotted data to obtain second trajectory dotted data corresponding to each first trajectory dotted data;

[0150] The associated traffic participants are traffic participants other than vehicles in the target road section, and the predicted trajectory constraints of the associated traffic participants are determined based on the vehicle's perception data of the associated traffic participants.

[0151] Specifically, associated traffic participants refer to traffic participants other than vehicles in the target road section, including but not limited to other vehicles, pedestrians, bicycles, etc. The predicted trajectory constraints of associated traffic participants are determined based on the vehicle's perception data of the associated traffic participants, and are used to describe the possible driving trajectories of associated traffic participants in the same time and space range as the target vehicle. It should be noted that the same time and space range means that the target vehicle and the associated traffic participants are in the same road section or adjacent road sections at the same time, and the positional relationship between them can be obtained through the vehicle's perception data (such as cameras, radars, lidars, etc.). The vehicle's perception data of associated traffic participants includes information such as the position, speed, and direction of the associated traffic participants.

[0152] The process of correcting each first trajectory dot data based on the physical characteristic constraints and traffic rule constraints of the target road section, as well as the predicted trajectory constraints of the associated traffic participants, includes the following steps: First, the cloud matches each first trajectory dot data with the target road section in the high-precision map to check whether it complies with the physical characteristic constraints and traffic rule constraints; Second, the cloud determines the predicted trajectory constraints of the associated traffic participants based on the vehicle's perception data of the associated traffic participants, and checks whether each first trajectory dot data complies with the predicted trajectory constraints of the associated traffic participants; If it is found that a certain first trajectory dot data does not comply with the above constraints, the cloud will correct the first trajectory dot data according to the physical characteristic constraints, traffic rule constraints and predicted trajectory constraints of the associated traffic participants in the high-precision map, such as adjusting its parameters such as position, speed or direction to make it comply with all constraints; The corrected first trajectory dot data becomes the corresponding second trajectory dot data.

[0153] For example, assume that when a target vehicle is traveling on a target road section, the cloud receives a first set of trajectory dotted data, including the first trajectory dotted data P1. First, the cloud matches P1 with the target road section in the HD map and checks whether it complies with the physical property constraints (such as lane width W) and traffic rule constraints (such as speed limit V_max). Second, based on the vehicle's perception data of the associated traffic participant (such as pedestrian A), the cloud determines that pedestrian A's predicted trajectory constraint is trajectory T_A and checks whether P1 complies with T_A. If P1 does not comply with the above constraints, the cloud corrects P1's position from (x1, y1) to (x2, y2), its speed from v1 to v2, and its direction from θ1 to θ2 based on the HD map's physical property constraints, traffic rule constraints, and pedestrian A's predicted trajectory constraints, so that it complies with all constraints. The corrected data becomes the second trajectory dotted data P2. If P1 complies with the above constraints, P1 is directly determined to be P2.

[0154] In this embodiment, by introducing constraints on the predicted trajectories of associated traffic participants, the error correction conditions are further refined, ensuring that the second trajectory dotted data not only conforms to the physical characteristics and traffic regulations of the target road segment but also reflects the driving trajectories of associated traffic participants. This processing approach ensures that the second trajectory dotted data more closely matches actual road conditions, further improving the accuracy of the trajectory dotted data in the trajectory file.

[0155] It should be understood that although the steps in the flowchart are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the figure may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0156] Based on the above Figure 1 The following is a scene diagram of Figure 9 The data processing device provided in the embodiment of the present application is introduced in detail. It should be noted that, Figure 9 The data processing device in the present application is used to execute Figure 2 - Figure 6 For the convenience of explanation, only the part related to the embodiment of the present application is shown. For the specific technical details not disclosed, please refer to the present application. Figure 2 - Figure 6 Specifically, the data processing device 800 is applied to a terminal, the terminal is connected to the cloud for communication, and the terminal is deployed on a vehicle; the data processing device 800 may include an acquisition unit 801, a sending unit 802, a verification unit 803, and a writing unit 804, specifically as follows:

[0157] An acquisition unit 801 is configured to acquire a first track dot data set of a vehicle within a preset time range, and to acquire driving state data of the vehicle within the preset time range;

[0158] The sending unit 802 is configured to send the first track dot data set to the cloud, so that the cloud can perform error correction processing on the first track dot data set based on a preset error correction strategy to obtain a second track dot data set;

[0159] A verification unit 803 is configured to receive the second track dot data set sent by the cloud, and verify the second track dot data set based on the driving status data to obtain a verification result;

[0160] The writing unit 804 is configured to write the second track dotting data set into a preset track file when the verification result indicates that the verification is passed.

[0161] Optionally, in some embodiments, the verification unit 803 can be used to: generate a first driving trajectory line based on driving status data; generate a second driving trajectory line based on a second trajectory dot data set; determine the geometric similarity between the first driving trajectory line and the second driving trajectory line; compare the geometric similarity with a preset geometric similarity threshold to obtain a verification result, and the verification result indicates that the verification is passed when the geometric similarity is not less than the geometric similarity threshold; and the verification result indicates that the verification is failed when the geometric similarity is less than the geometric similarity threshold.

[0162] Optionally, in some embodiments, the verification unit 803 can be used to: determine multiple target time points within a preset time range; determine the vehicle's driving speed and driving direction at each of the multiple target time points based on driving status data; determine the vehicle position corresponding to each target time point based on the vehicle's driving speed and driving direction at each target time point; and generate a first driving trajectory line based on the vehicle position corresponding to each target time point.

[0163] Optionally, in some embodiments, the verification unit 803 may be configured to: determine position information corresponding to each second trajectory dot data in the second trajectory dot data set; and generate a second driving trajectory line based on the position information corresponding to each second trajectory dot data.

[0164] Optionally, in some embodiments, the verification unit 803 can be used to: calculate the geometric distance deviation value between the first driving trajectory line and the second driving trajectory line based on a preset geometric distance algorithm; normalize the geometric distance deviation value between the first driving trajectory line and the second driving trajectory line to obtain the geometric similarity between the first driving trajectory line and the second driving trajectory line.

[0165] Optionally, in some embodiments, the data processing device 800 may be configured to: when the verification result indicates that the verification has failed, control the terminal and / or the vehicle to push track marking abnormality information.

[0166] The effects that can be achieved by this embodiment can be found in the relevant embodiments of the above-mentioned data processing method, which will not be described in detail here.

[0167] Based on the above Figure 1 The following is a scene diagram of Figure 10 The data processing device provided in the embodiment of the present application is introduced in detail. It should be noted that, Figure 10 The data processing device in the present application is used to execute Figure 7 - Figure 8 For the convenience of explanation, only the part related to the embodiment of the present application is shown. For the specific technical details not disclosed, please refer to the present application. Figure 7 - Figure 8Specifically, the data processing device 900 is applied to the cloud, the cloud is connected to the terminal, and the terminal is deployed in the vehicle; the data processing device 900 may include a receiving unit 901, an error correction unit 902, and a sending unit 903, as follows:

[0168] The receiving unit 901 is configured to receive a first track dot data set of a vehicle within a preset time range sent by a terminal;

[0169] An error correction unit 902 is configured to perform error correction processing on the first track dot data set based on a preset error correction strategy to obtain a second track dot data set;

[0170] The sending unit 903 is configured to send the second trajectory dot data set to the terminal, so that the terminal can verify the second trajectory dot data set based on the driving status data of the vehicle within a preset time range to obtain a verification result, and write the second trajectory dot data set into a preset trajectory file if the verification result indicates that the verification is passed.

[0171] Optionally, in some embodiments, the error correction unit 902 can be used to: determine the target road section where each first trajectory dot data in the first trajectory dot data set is located in a preset high-precision map; determine the physical characteristic constraints and traffic rule constraints of the target road section based on the high-precision map; perform error correction processing on each first trajectory dot data based on the physical characteristic constraints and traffic rule constraints of the target road section to obtain the second trajectory dot data corresponding to each first trajectory dot data; and determine the second trajectory dot data set based on each second trajectory dot data.

[0172] Optionally, in some embodiments, the error correction unit 902 can be used to: based on the physical property constraints and traffic rule constraints of the target road section, and the predicted trajectory constraints of the associated traffic participants, perform error correction processing on each first trajectory dot data to obtain second trajectory dot data corresponding to each first trajectory dot data; wherein, the associated traffic participants are traffic participants other than vehicles in the target road section, and the predicted trajectory constraints of the associated traffic participants are determined based on the vehicle's perception data of the associated traffic participants.

[0173] The effects that can be achieved by this embodiment can be found in the relevant embodiments of the above-mentioned data processing method, which will not be described in detail here.

[0174] For the specific definition of the above-mentioned data processing device, please refer to the definition of the data processing method above, and will not be repeated here. Each unit in the above-mentioned data processing device can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned units can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned units.

[0175] In one embodiment, a computer device is provided. The internal structure diagram of the computer device can be as follows: Figure 11 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a data processing method as described above can be implemented. It includes: a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, any step in the data processing method as described above is implemented.

[0176] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, any step in the above data processing method can be implemented.

[0177] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0178] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0179] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0180] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0181] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0182] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. A data processing method, characterized in that: Applied to a terminal, the terminal is connected to the cloud for communication, and the terminal is deployed on a vehicle; the method includes: Acquire a first track dot data set of the vehicle within a preset time range, and acquire driving state data of the vehicle within the preset time range; Sending the first track dot data set to the cloud, so that the cloud performs error correction processing on the first track dot data set based on a preset error correction strategy to obtain a second track dot data set; receiving the second track dot data set sent by the cloud, and verifying the second track dot data set based on the driving status data to obtain a verification result; When the verification result indicates that the verification is passed, the second track dotting data set is written into a preset track file.

2. The method according to claim 1, characterized in that The verifying the second track dot data set based on the driving state data to obtain a verification result includes: generating a first driving trajectory line based on the driving state data; generating a second driving trajectory line based on the second trajectory dot data set; determining a geometric similarity between the first driving trajectory line and the second driving trajectory line; The geometric similarity is compared with a preset geometric similarity threshold to obtain a verification result, wherein the verification result indicates that the verification is passed if the geometric similarity is not less than the geometric similarity threshold, and the verification result indicates that the verification is failed if the geometric similarity is less than the geometric similarity threshold.

3. The method according to claim 2, characterized in that The step of generating a first driving trajectory line based on the driving state data includes: Determining multiple target time points within the preset time range; determining a driving speed and a driving direction of the vehicle at each of the plurality of target time points based on the driving state data; Determining the vehicle position corresponding to each target time point based on the vehicle's travel speed and travel direction at each target time point; A first driving trajectory line is generated based on the vehicle position corresponding to each target time point.

4. The method according to claim 2, characterized in that The step of generating a second driving trajectory line based on the second trajectory dot data set includes: Determining position information corresponding to each second track dot data in the second track dot data set; A second driving trajectory line is generated based on the position information corresponding to each of the second trajectory dot data.

5. The method according to claim 2, characterized in that The determining of the geometric similarity between the first driving trajectory line and the second driving trajectory line includes: Calculating a geometric distance deviation value between the first driving trajectory line and the second driving trajectory line based on a preset geometric distance algorithm; Normalizing the geometric distance deviation between the first driving trajectory line and the second driving trajectory line to obtain the geometric similarity between the first driving trajectory line and the second driving trajectory line.

6. The method according to claim 1, wherein After verifying the second track dot data set based on the driving state data to obtain a verification result, the method further includes: When the verification result indicates that the verification fails, the terminal and / or the vehicle are controlled to push track marking abnormality information.

7. A data processing method, characterized in that: Applied to the cloud, the cloud is in communication with a terminal, and the terminal is deployed in a vehicle; the method includes: receiving a first track dot data set of the vehicle within a preset time range sent by the terminal; Based on a preset error correction strategy, performing error correction processing on the first track dot data set to obtain a second track dot data set; The second trajectory dot data set is sent to the terminal, so that the terminal verifies the second trajectory dot data set based on the driving status data of the vehicle within the preset time range to obtain a verification result, and writes the second trajectory dot data set into a preset trajectory file if the verification result indicates that the verification is passed.

8. The method according to claim 7, characterized in that The step of performing error correction on the first track dot data set based on a preset error correction strategy to obtain a second track dot data set includes: In a preset high-precision map, determining a target road section where each first track dot data in the first track dot data set is located; Determining physical property constraints and traffic rule constraints of the target road section based on the high-precision map; Based on the physical property constraints and traffic rule constraints of the target road section, performing error correction processing on each of the first trajectory dotted data to obtain second trajectory dotted data corresponding to each of the first trajectory dotted data; A second track dotting data set is determined based on each of the second track dotting data.

9. The method according to claim 8, characterized in that The step of performing error correction processing on each of the first trajectory dotted data based on the physical property constraints and traffic rule constraints of the target road section to obtain second trajectory dotted data corresponding to each of the first trajectory dotted data includes: Based on the physical property constraints and traffic rule constraints of the target road section and the predicted trajectory constraints of the associated traffic participants, performing error correction processing on each of the first trajectory dotted data to obtain second trajectory dotted data corresponding to each of the first trajectory dotted data; The associated traffic participants are traffic participants other than the vehicle in the target road section, and the predicted trajectory constraints of the associated traffic participants are determined based on the vehicle's perception data of the associated traffic participants.

10. A computer device comprising: The method comprises a memory and a processor, wherein the memory stores a computer program, and is characterized in that the processor implements the steps of the method according to any one of claims 1 to 9 when executing the computer program.

Images