Vehicle end data return method, computer equipment and computer program product
By filtering and matching the positioning and perceptual data on the vehicle side, only the data located in the preset window is retained, which solves the problem of excessive network resources consumption in the data backload of the vehicle side, and improves the accuracy and difficulty of cloud mapping construction .
Patent Information
- Application Number
- CN202411569965.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-05
- Publication Date
- 2024-12-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the familiar mode, the vehicle side needs to pass a large amount of data back to the cloud, resulting in excessive consumption of network resources and affecting the accuracy and difficulty of cloud mapping construction.
By filtering and matching the positioning data and perceived data on the vehicle side, only the data located in the preset bit window and time window are retained, and the perceived timestamp is used to match the positioning timestamp, reducing the amount of data and improving the matching degree, uploading To the clouds.
It effectively reduces the amount of data transmitted to the cloud, reduces invalid data collection, improves the accuracy and difficulty of cloud mapping, and saves network resources.
Smart Images

Figure CN119094598A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of vehicle technology, and in particular to a vehicle-side data return method, a computer device, and a computer program product. Background Art
[0002] In the familiar road mode of intelligent driving of the vehicle, the user's high-frequency routes (such as commuting routes and familiar routes) can be optimized. For example, the user can manually drive on the high-frequency route several times to complete the collection and construction of the prior data of the route, and after the automated algorithm and manual analysis and approval, use this data to open the intelligent driving functions of the route for the user (such as automatic driving, map modeling, etc.).
[0003] In order to provide intelligent driving functions for high-frequency routes, a large amount of data needs to be collected from the vehicle and uploaded to the service cloud, which consumes a lot of network resources. Summary of the invention
[0004] The embodiments of the present disclosure at least provide a vehicle-side data transmission method, a computer device, and a computer program product.
[0005] In a first aspect, an embodiment of the present disclosure provides a vehicle-side data return method, including: After the positioning data is acquired, if the positioning position corresponding to the positioning data is within the current positioning window, the positioning data is retained; the positioning window corresponds to a preset distance length; After acquiring the perception data, determining whether the timestamp of the perception data is within the time window corresponding to the positioning window, and if so, retaining the perception data; the time window is determined based on the positioning timestamps corresponding to the starting positioning position and the ending positioning position of the positioning window respectively; For the retained perception data, determine target positioning data matching the perception data according to positioning data with at least one positioning timestamp closest to the perception timestamp of the perception data; The retained perception data and target positioning data matching the perception data are uploaded to the cloud; the perception data and target positioning data are used as reference data for controlling vehicle driving in the familiar road mode.
[0006] In an optional implementation manner, determining target positioning data matching the perception data according to positioning data under at least one positioning timestamp closest to the perception timestamp of the perception data includes: Acquire positioning data with two positioning timestamps closest to the sensing timestamp of the sensing data; According to the positioning data at the two preceding and succeeding positioning timestamps and the vehicle speed information recorded between the two preceding and succeeding positioning timestamps, the predicted positioning data at the perception timestamp is determined, and the predicted positioning data is used as the target positioning data matching the perception data.
[0007] In an optional implementation, the perception data includes marking line data on the road; Uploading the retained perception data and target positioning data matching the perception data to the cloud, including: Determining target marking line data located within a target distance range in front of the vehicle from the marking line data; The target marking line data and its corresponding target positioning data are uploaded to the cloud.
[0008] In an optional implementation manner, uploading the target marking line data and its corresponding target positioning data to the cloud includes: Determine parameter information of an equation expression corresponding to the marking line data according to coordinate data of each feature point corresponding to the marking line data; the equation expression is used to represent a conversion relationship between the coordinate data of any feature point and a timestamp corresponding to the feature point; The parameter information and the target positioning data are uploaded to the cloud.
[0009] In an optional implementation manner, before uploading the retained perception data to the cloud, the method further includes: When it is obtained that different perception data sent by multiple perception data sources include data for the same perception object, fusion processing is performed on the different perception data to obtain fused perception data for the perception object.
[0010] In an optional implementation manner, the fusing the different perception data to obtain the fused perception data for the perception object includes: For each of the plurality of perception data sources, extracting the to-be-fused data of the perception object in multiple attribute dimensions from the perception data corresponding to the perception data source; Performing a quality evaluation on each to-be-fused data corresponding to each perception data source to obtain a quality score corresponding to the to-be-fused data; Based on the quality scores of the data to be fused, determining the target data to be fused corresponding to each attribute dimension; The target data to be fused is fused to obtain fused perception data for the perception object.
[0011] In an optional implementation manner, the reference data for controlling the driving of the vehicle further includes lightweight map data; The method further comprises: After each vehicle travels a first preset distance, lightweight map data within a preset distance range centered on the vehicle is obtained; the lightweight map data is determined based on at least one simplified perception data; the accuracy level of the lightweight map data is lower than that of the high-definition map data; The lightweight map within the preset distance range is uploaded to the cloud.
[0012] In an optional implementation manner, the reference data for controlling the driving of the vehicle further includes image data; The method further comprises: After the vehicle travels a second preset distance each time, determining a target grid area where the vehicle is located according to the current location of the vehicle, and determining a preset number of grid areas around the target grid area as the center; Requesting the cloud to obtain historical image data corresponding to the target grid area and a preset number of surrounding grid areas; After the current image data to be uploaded is acquired, determining whether there is image data that overlaps with the image data to be uploaded from the most recently acquired historical image data; If so, the overlapping image data are filtered out and the non-overlapping image data are uploaded to the cloud.
[0013] In a second aspect, an optional implementation of the present disclosure further provides a computer device, a processor, and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the processor is used to execute the machine-readable instructions stored in the memory, and when the machine-readable instructions are executed by the processor, the machine-readable instructions perform the steps in the above-mentioned first aspect, or any possible implementation of the first aspect.
[0014] In a third aspect, an optional implementation of the present disclosure further provides a vehicle, comprising a control device, wherein the control device is used to execute the steps in the above-mentioned first aspect, or any possible implementation of the first aspect.
[0015] In a fourth aspect, an optional implementation of the present disclosure further provides a computer program product, which, when executed, executes the steps of the above-mentioned first aspect, or any possible implementation of the first aspect.
[0016] For a description of the effects of the above-mentioned computer equipment and computer program products, please refer to the description of the above-mentioned vehicle-side data feedback method, which will not be repeated here.
[0017] The vehicle-side data feedback method, computer device and computer program product provided by the embodiments of the present disclosure can, when acquiring positioning data, determine whether to retain the positioning data based on whether the positioning position of the positioning data is within the current positioning window; when acquiring perception data, determine whether to retain the perception data based on whether the timestamp of the perception data is within the time window corresponding to the positioning window; in this way, unnecessary perception data and positioning data can be filtered out, required data can be retained, and the amount of data transmitted to the cloud can be reduced; and through the perception timestamp and the positioning timestamp, the perception data can be matched with the positioning data, so that the cloud can use the perception data and its matched target positioning data as reference data for controlling vehicle driving in the familiar road mode.
[0018] In order to make the above-mentioned objectives, features and advantages of the present disclosure more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following is a brief introduction to the drawings required for use in the embodiments. The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used together with the specification to illustrate the technical solutions of the present disclosure. It should be understood that the following drawings only illustrate certain embodiments of the present disclosure and should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can also be obtained based on these drawings without creative work.
[0020] Figure 1 A flow chart of a vehicle-side data return method provided by an embodiment of the present disclosure is shown; Figure 2 A schematic diagram showing the perception data and target positioning data provided by an embodiment of the present disclosure is shown; Figure 3 A schematic diagram of the structure of a computer device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION
[0021] In order to make the purpose, technical scheme and advantages of the embodiments of the present disclosure clearer, the technical scheme in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The components of the embodiments of the present disclosure generally described and shown here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure is not intended to limit the scope of the present disclosure claimed for protection, but merely represents the selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present disclosure.
[0022] In addition, the terms "first", "second", etc. in the description and claims of the embodiments of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein.
[0023] The "multiple or several" mentioned in this article refers to two or more. "And / or" describes the association relationship of the 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. The character " / " generally indicates that the associated objects are in an "or" relationship.
[0024] Research has found that in the familiar road mode, the vehicle needs to transmit data (such as perception data, positioning data, image data, etc.) back to the cloud. The higher the frame rate of the returned data (the number of times data is transmitted per second), the easier it is for the cloud to build maps and the higher the accuracy, but the larger the data volume is, the more network resources are consumed. Therefore, how to reduce the network resources consumed by data transmission while keeping the difficulty of cloud-based mapping low has become an important issue in the familiar road mode.
[0025] Based on the above research, the present disclosure provides a vehicle-side data feedback method, computer equipment and computer program product. When the positioning data is obtained, it can be determined whether to retain the positioning data according to whether the positioning position of the positioning data is within the current positioning window; when the perception data is obtained, it can be determined whether to retain the perception data according to whether the timestamp of the perception data is within the time window corresponding to the positioning window; in this way, unnecessary perception data and positioning data can be filtered out, required data can be retained, and the amount of data transmitted to the cloud can be reduced; and through the perception timestamp and positioning timestamp, the perception data can be matched with the positioning data, so that the cloud can use the perception data and its matching target positioning data as reference data for controlling vehicle driving in familiar road mode.
[0026] The defects existing in the above solutions are the results obtained by the inventor after practice and careful research. Therefore, the discovery process of the above problems and the solutions proposed by the present disclosure for the above problems below should be the contributions made by the inventor to the present disclosure during the disclosure process.
[0027] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.
[0028] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.
[0029] To facilitate understanding of this embodiment, a vehicle-side data return method disclosed in an embodiment of the present disclosure is first introduced in detail. The executor of the vehicle-side data return method provided in the embodiment of the present disclosure is generally a terminal device or other processing device with certain computing capabilities, wherein the terminal device may be a user equipment (UE), a mobile device, a user terminal, a terminal, a vehicle-mounted device, a computer device, etc.; in some possible implementation methods, the vehicle-side data return method may be implemented by a processor calling computer-readable instructions stored in a memory.
[0030] The vehicle-side data feedback method provided by the embodiment of the present disclosure is described below by taking the execution subject as a terminal device as an example.
[0031] like Figure 1 As shown, it is a flow chart of a vehicle-side data return method provided by an embodiment of the present disclosure, which may include the following steps: S101. After obtaining positioning data, if the positioning position corresponding to the positioning data is within a current positioning window, retaining the positioning data; the positioning window corresponds to a preset distance length.
[0032] The terminal device may be a vehicle-mounted terminal, a mobile terminal, etc. This embodiment is described by taking a vehicle-mounted terminal as an example.
[0033] The above-mentioned vehicle-mounted terminal can be a domain controller in the vehicle. A domain controller is an integrated control unit in the automotive electronic and electrical architecture, which is used to centrally manage different functional domains of the vehicle. As the electronic control system of the vehicle becomes more and more complex, domain controllers are introduced to replace the traditional distributed electronic control unit (ECU) architecture, integrating multiple functions into a small number of high-performance computing units. Typically, domain controllers can include power domain controllers, body domain controllers, infotainment domain controllers, autonomous driving domain controllers, communication domain controllers, chassis domain controllers, etc.
[0034] In order to realize the intelligent driving function of the familiar road mode, the cloud usually needs to obtain perception data and positioning data to model the route. For example, the data sent back by the vehicle usually includes positioning data, perception data, map data, its own wheel speed, driving status, inertial data, etc. The above data can be transmitted to the terminal device through one or more perception data sources deployed in the vehicle, and the terminal device sends it to the cloud.
[0035] The above-mentioned perception data sources can usually be sensors or domain controllers deployed in the vehicle. A variety of sensors can usually be deployed in a vehicle, and these sensors can include cameras, global positioning systems, radars, lidars, ultrasonic sensors, wheel speed sensors, inertial measurement units, steering angle sensors, etc.
[0036] In the disclosed embodiment, the above positioning data may include vehicle location information received by a global positioning system. Normally, the global positioning system receives positioning information at a high frequency, such as 100 Hz, and each time the vehicle obtains positioning information, it will report it to the terminal device. However, positioning information needs to be used in conjunction with sensory information to complete modeling, but the frequency of collecting sensory information is usually low, such as 10 Hz. Therefore, the cloud does not require fine-grained positioning data for modeling, resulting in a large amount of invalid data in the positioning information returned by the vehicle to the cloud, wasting part of the network resources.
[0037] To this end, the embodiment of the present disclosure may be provided with a positioning window, the length of which may match the preset distance length, so that it is possible to determine whether the acquired positioning data is within the current positioning window to determine whether to retain the positioning data, and only retain the positioning data within the positioning window. The positioning window may adopt a sliding window method, and when the number of positioning data retained in the current positioning window meets a preset threshold (such as 1), the next positioning window may be entered, and the range of the next positioning window may be determined by sliding the preset distance length by the range of the current positioning window. In this way, the collection of "advanced sent" data (that is, the received positioning data exceeds the preset distance length corresponding to the positioning window) can be avoided, thereby reducing the collection of invalid data.
[0038] Exemplarily, the preset distance length can be set to 100m. When positioning data is obtained in the current positioning window, it can be determined whether the positioning data is in the current positioning window. If the current positioning window is 0 to 100m, and the currently obtained positioning data is 50m, which is in the current positioning window, the positioning data can be saved. At this time, if the number of retained positioning data meets the preset threshold, the next positioning window can be entered. The positioning window can adopt a sliding window method, and the next positioning window can be 100m to 200m. If new positioning data is obtained at this time, which is 51m, and the positioning data does not match the current positioning window of 100m to 200m, the positioning data can be not retained.
[0039] For another example, if the current positioning window is 0 to 100 meters, and the currently acquired positioning data is 150 meters, the positioning data will not be retained.
[0040] The above-mentioned preset distance length can be set according to the specific requirements of the cloud for the positioning window, or can also be set according to the collection frequency of the perception data.
[0041] S102. After acquiring the perception data, determine whether the timestamp of the perception data is within a time window corresponding to the positioning window. If so, retain the perception data. The time window is determined based on the positioning timestamps corresponding to the starting positioning position and the ending positioning position of the positioning window.
[0042] The above-mentioned perception data may include information such as lane lines, lane boundaries, sidewalks, traffic lights, obstacles, etc., which may be directly collected by sensors such as radars and cameras, or obtained through calculation and identification by other domain controllers using data collected by sensors.
[0043] In this step, the positioning data may carry a timestamp, or the timestamp corresponding to the positioning data may be determined when the positioning data is acquired. According to the timestamp of the positioning data, the time window corresponding to each positioning window may be determined.
[0044] Specifically, the time window may be determined based on the positioning timestamps corresponding to the starting positioning position and the ending positioning position of the positioning window. For example, the positioning timestamp corresponding to the starting positioning position of the positioning window may be determined as the starting time of the time window, and the positioning timestamp corresponding to the ending position of the positioning window may be determined as the ending time of the time window.
[0045] In this way, the time periods of the current positioning window and the perception window match each other, and the positioning data and perception data of this time period are retained, so that the time interval between the retained positioning data and the perception data is shorter and the matching degree is higher.
[0046] S103: For the retained perception data, determine target positioning data matching the perception data according to positioning data with at least one positioning timestamp closest to the perception timestamp of the perception data.
[0047] After the perception data and positioning data are acquired, the perception data and positioning data may be matched so that the cloud can use the perception data and positioning data smoothly. Specifically, the matching may be performed based on the timestamp of the perception data and the positioning timestamp of the positioning data.
[0048] For example, there are existing perception data A and positioning data B, and the positioning data B is the positioning data under at least one positioning timestamp that is closest to the perception timestamp of the perception data, and the time difference between the perception timestamp corresponding to the perception data A and the positioning timestamp of the positioning data B is less than or equal to the preset difference, then the positioning data B can be directly used as the target positioning data matching the perception data A.
[0049] In a possible implementation, if the above-mentioned time difference is greater than a preset difference, the positioning data under the two positioning timestamps closest to the perception time of the perception data can be obtained, and then, based on the vehicle speed information recorded between the two positioning timestamps, the predicted positioning data under the perception timestamp is determined, and the predicted positioning data is used as the target positioning data matching the perception data.
[0050] S104, uploading the retained perception data and target positioning data matching the perception data to the cloud; the perception data and target positioning data are used as reference data for controlling vehicle driving in the familiar road mode.
[0051] After matching the perception data with the positioning data, the perception data and the target positioning data that matches the perception data can be uploaded to the cloud. The cloud can use the perception data and the target positioning data as reference data for controlling vehicle driving in familiar road mode, modeling the route, and providing intelligent driving services.
[0052] See also Figure 2 , which is a schematic diagram of the perception data and target positioning data provided by an embodiment of the present disclosure. Figure 2 It includes perception information 21 and target positioning data 22 matching the perception information 21 .
[0053] The above-mentioned perception data may include road marking data. When uploading the data to the cloud, the target marking data located within the target distance range in front of the vehicle can be determined from the marking data, and the target marking data and its corresponding target positioning data can be uploaded to the cloud. In this way, on the one hand, repeated transmission of the markings behind the vehicle that have been uploaded can be avoided, and on the other hand, the uploaded marking data can be limited to the target distance range in front of the vehicle, further reducing the data volume of the marking data.
[0054] The above-mentioned marking lines can refer to the iconic lines on the road, which are used to indicate how vehicles and pedestrians should drive or pass on the road. In road traffic, marking lines can be divided into different types, such as lane lines, stop lines, crosswalks, etc., which play the role of indicating and regulating traffic flow. Roads can be modeled using marking line data.
[0055] In the vehicle-side protocol or data format, the marking line is usually represented by the three-dimensional coordinates of multiple coordinate points, and the coordinates can be represented by float type or double type. Preferably, the float type can be used to represent the number of coordinates, thereby reducing the storage occupancy of data.
[0056] Furthermore, the coordinates in the marking line data may be optimized, and the marking line data may be described using equation expressions, thereby depicting a plurality of discrete points as a curve.
[0057] Specifically, the parameter information of the equation corresponding to the marking line can be determined according to the coordinate data of each feature point corresponding to the marking line data, and the equation is used to express the conversion relationship between the coordinate data of any feature point and the timestamp corresponding to the feature point. In this way, by substituting the timestamp into the equation, the coordinate data of the feature point corresponding to the timestamp can be calculated.
[0058] For example, the marking line data usually has a sampling point (i.e., feature point) every 3m to 6m. For a 20m lane line, there will be 3 to 6 reference points, and each point is composed of three floats, x, y, and z, so a line has about 9-18 floats. When using equation expressions to represent the marking line data, for a second-order parametric equation, only the parameter information a, b, c, and the timestamp range tmin, tmax 5 floats are needed to express it.
[0059] In this way, uploading the parameter information and target positioning data to the cloud can further reduce the amount of data.
[0060] Perception data usually comes from different perception data sources, which are usually independent of each other. Perception data sent by different perception data sources may contain perception results of the same perception object. The perception object can be a real object in the perception data. For example, the perception data can contain information about perception objects such as lane lines, lane boundaries, sidewalks, traffic lights, obstacles, etc. When subscribing to perception data, terminal devices often encounter the problem of repeated subscription. For example, perception data A and perception data B both contain perception data for lane lines. The lane lines in perception data A have ids, but the lane line effect is not good; the lane lines in perception data B have good effects but no ids.
[0061] In order to collect better data and reduce the amount of data during data transmission, for each of the multiple perception data sources, the data to be fused of the perception object in multiple attribute dimensions can be extracted from the perception data corresponding to the perception data source. For example, when multiple perception data sources simultaneously contain perception data of a perception object, a specific lane line, the lane line can contain perception data of two attribute dimensions, namely, lane line geometry and lane line identification ID. The data to be fused in the attribute dimension of lane line geometry and the data to be fused in the attribute dimension of lane line identification can be extracted from the perception data.
[0062] After obtaining the data to be fused, the quality of each data to be fused corresponding to each perception data source can be evaluated to obtain a quality score corresponding to the data to be fused. The basis for the quality score can be one or more features of the data to be fused. Different features can be used to score different data to be fused. Exemplarily, the quality score of the lane line geometry can be determined based on at least one feature of the clarity of the lane line geometry, the number of reference points, and the curvature of the lane line geometry.
[0063] Afterwards, the target data to be fused corresponding to each attribute dimension can be determined based on the quality scores of the data to be fused; finally, the target data to be fused can be fused to obtain fused perception data for the perceived object. For example, for each attribute dimension, the data to be fused with the highest quality score can be selected from the data to be fused in the attribute dimension from each perception data source as the target data to be fused, and then the target data to be fused with the highest quality score under each attribute dimension can be combined together as the fused perception data of the perceived object.
[0064] For example, in the case of the existing perception data A and perception data B, since perception data A has an id, but the geometric features of the lane line are not clear; while the geometric features of the lane line of perception data B are clear, but the id is missing; therefore, a new lane line can be created: it uses the geometric features of the lane line in perception data B and the id in perception data A. When fusing the perception data, the perception data can also be interpolated to obtain the dataset topicA and dataset topicB corresponding to perception data A and perception data B, respectively. In order to use the geometric features of perception data B, a rectangular tree Rtree can also be constructed for B: Rtree is a spatial indexing technology used for spatial matching, suitable for storing and querying spatial objects such as rectangular areas, points, and lines, and can efficiently perform range queries and nearest neighbor queries. Afterwards, for each lane line in perception data A, its corresponding position in perception data B can be found, and the id in perception data A can be assigned to perception data B, thereby realizing the fusion of perception data.
[0065] Furthermore, the data collected by the Inertial Measurement Unit (IMU), the perception data of the vehicle information such as wheel speed, etc. (topic) can also be integrated into a new topic, and the purpose of saving resources can be achieved by controlling the sending frequency and the amount of information sent at a single time. For example, the receiving frequency of the IMU is 100h, while the frequency of the Global Navigation Satellite System (GNSS) is only 1hz. The two are not uniform in time. The distance sampling method can be used. For example, a new topic is sent every time the vehicle travels 1m. In this meter, the terminal device may receive a lot of IMU data, or it may not receive a frame of GNSS data. For this reason, a window can also be maintained to save various types of data between two returns, and the return is unified according to the distance threshold.
[0066] The reference data for controlling vehicle driving may also include map data. In order to reduce the amount of data transmission, lightweight map data may be selected for transmission. The lightweight map data may be determined based on at least one simplified perception data; the accuracy level of the lightweight map data is lower than that of the high-definition map data.
[0067] When transmitting lightweight map data, lightweight map data within a preset distance range centered on the vehicle can be obtained after the vehicle travels a first preset distance, and then the obtained lightweight map data within the preset distance range can be uploaded to the cloud, thereby reducing the amount of data transmission.
[0068] For example, the lightweight map data may be sent once every 150 m traveled by the vehicle, and the range of the sent lightweight map data may be within 800 m.
[0069] The reference data for controlling the driving of the vehicle may also include image data, which may be images of the road or environment taken by a camera on the vehicle, or point cloud images collected by a radar.
[0070] After the vehicle travels the second preset distance each time, the target grid area where the vehicle is located can be determined according to the current location of the vehicle, and a preset number of grid areas around the target grid area can be determined; thereafter, a request can be made to the cloud to obtain historical image data corresponding to the target grid area and the preset number of grid areas around it; after obtaining the current image data to be uploaded, determine from the most recently obtained historical image data whether there is image data that overlaps with the image data to be uploaded; if so, filter out the overlapping image data, and upload the non-overlapping image data to the cloud.
[0071] For example, the concept of a grid area tile can be defined. A tile can be understood as a square area with a side length of 1 km. The cloud divides the real world into multiple tiles. Every 5-10 meters, the vehicle can calculate a tile in which it is located based on the current pose, and then pull the image position information of the 8 tiles around it based on the current tile, a total of 9 tiles, and store them in RTree. Pulling 9 tiles from the cloud can prevent the situation where the vehicle suddenly runs out of data outside the current tile. In this way, you can know which sections of the road vehicles have sent back images. Finally, by synchronously updating the cloud image location database, you can continue to determine whether there is duplicate image data the next time you drive to a new tile.
[0072] In this way, by dividing the area into multiple grid areas and then requesting historical image data of the grid areas around the vehicle from the server, it is possible to determine whether the cloud has image data of the surrounding area stored, thereby deleting duplicate image data and further reducing the amount of transmitted data.
[0073] The vehicle-side data feedback method provided by the embodiment of the present disclosure can determine whether to retain the positioning data based on whether the positioning position of the positioning data is within the current positioning window when the positioning data is acquired; and can determine whether to retain the perception data based on whether the timestamp of the perception data is within the time window corresponding to the positioning window when the perception data is acquired. In this way, unnecessary perception data and positioning data can be filtered out, necessary data can be retained, and the amount of data transmitted to the cloud can be reduced; and through the perception timestamp and the positioning timestamp, the perception data can be matched with the positioning data, so that the cloud can use the perception data and its matched target positioning data as reference data for controlling vehicle driving in the familiar road mode.
[0074] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.
[0075] Based on the same technical concept, the embodiment of the present application also provides a computer device. Figure 3 FIG. 1 is a schematic diagram of a computer device provided in an embodiment of the present application, including: Processor 301, memory 302 and bus 303. The memory 302 stores machine-readable instructions executable by the processor 301, and the processor 301 is used to execute the machine-readable instructions stored in the memory 302. When the machine-readable instructions are executed by the processor 301, the processor 301 performs the following steps: After the positioning data is acquired, if the positioning position corresponding to the positioning data is within the current positioning window, the positioning data is retained; the positioning window corresponds to a preset distance length; After acquiring the perception data, determining whether the timestamp of the perception data is within the time window corresponding to the positioning window, and if so, retaining the perception data; the time window is determined based on the positioning timestamps corresponding to the starting positioning position and the ending positioning position of the positioning window respectively; For the retained perception data, determine target positioning data matching the perception data according to positioning data with at least one positioning timestamp closest to the perception timestamp of the perception data; The retained perception data and target positioning data matching the perception data are uploaded to the cloud; the perception data and target positioning data are used as reference data for controlling vehicle driving in the familiar road mode.
[0076] In an optional implementation manner, in the instruction executed by the processor 301, determining the target positioning data matching the perception data according to the positioning data under at least one positioning timestamp closest to the perception timestamp of the perception data includes: Acquire positioning data with two positioning timestamps closest to the sensing timestamp of the sensing data; According to the positioning data at the two preceding and succeeding positioning timestamps and the vehicle speed information recorded between the two preceding and succeeding positioning timestamps, the predicted positioning data at the perception timestamp is determined, and the predicted positioning data is used as the target positioning data matching the perception data.
[0077] In an optional implementation, in the instructions executed by the processor 301, the perception data includes marking line data on the road; Uploading the retained perception data and target positioning data matching the perception data to the cloud, including: Determining target marking line data located within a target distance range in front of the vehicle from the marking line data; The target marking line data and its corresponding target positioning data are uploaded to the cloud.
[0078] In an optional implementation manner, the instructions executed by the processor 301 to upload the target marking line data and its corresponding target positioning data to the cloud include: Determine parameter information of an equation expression corresponding to the marking line data according to coordinate data of each feature point corresponding to the marking line data; the equation expression is used to represent a conversion relationship between the coordinate data of any feature point and a timestamp corresponding to the feature point; The parameter information and the target positioning data are uploaded to the cloud.
[0079] In an optional implementation manner, the instructions executed by the processor 301, before uploading the retained sensing data to the cloud, further include: When it is obtained that different perception data sent by multiple perception data sources include data for the same perception object, fusion processing is performed on the different perception data to obtain fused perception data for the perception object.
[0080] In an optional implementation manner, in the instructions executed by the processor 301, the fusing the different perception data to obtain the fused perception data for the perception object includes: For each of the plurality of perception data sources, extracting the to-be-fused data of the perception object in multiple attribute dimensions from the perception data corresponding to the perception data source; Performing a quality evaluation on each to-be-fused data corresponding to each perception data source to obtain a quality score corresponding to the to-be-fused data; Based on the quality scores of the data to be fused, determining the target data to be fused corresponding to each attribute dimension; The target data to be fused is fused to obtain fused perception data for the perception object.
[0081] In an optional implementation, in the instructions executed by the processor 301, the reference data for controlling the driving of the vehicle further includes lightweight map data; The instructions executed by the processor 301 also include: After each vehicle travels a first preset distance, lightweight map data within a preset distance range centered on the vehicle is obtained; the lightweight map data is determined based on at least one simplified perception data; the accuracy level of the lightweight map data is lower than that of the high-definition map data; The lightweight map within the preset distance range is uploaded to the cloud.
[0082] In an optional implementation, in the instructions executed by the processor 301, the reference data for controlling the driving of the vehicle further includes image data; The instructions executed by the processor 301 also include: After the vehicle travels a second preset distance each time, determining a target grid area where the vehicle is located according to the current location of the vehicle, and determining a preset number of grid areas around the target grid area as the center; Requesting the cloud to obtain historical image data corresponding to the target grid area and a preset number of surrounding grid areas; After the current image data to be uploaded is acquired, determining whether there is image data that overlaps with the image data to be uploaded from the most recently acquired historical image data; If so, the overlapping image data are filtered out and the non-overlapping image data are uploaded to the cloud.
[0083] The above-mentioned memory 302 includes internal memory 3021 and external memory 3022; the memory 3021 here is also called internal memory, which is used to temporarily store the calculation data in the processor 301, as well as the data exchanged with the external memory 3022 such as a hard disk. The processor 301 exchanges data with the external memory 3022 through the internal memory 3021. When the computer device is running, the processor 301 communicates with the memory 302 through the bus 303, so that the processor 301 executes the execution instructions mentioned in the above-mentioned method embodiment.
[0084] The embodiment of the present disclosure also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the vehicle-side data return method described in the above method embodiment are executed. The storage medium can be a volatile or non-volatile computer-readable storage medium.
[0085] The embodiment of the present disclosure also provides a vehicle, including a control device, which is used to execute the steps of the vehicle-side data return method described in the above method embodiment. Since the principle of solving the problem by the vehicle in the embodiment of the present disclosure is similar to the above vehicle-side data return method in the embodiment of the present disclosure, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.
[0086] The disclosed embodiment also provides a computer program product, including a program code. The instructions included in the program code can be used to execute the steps of the vehicle-side data feedback method described in the above method embodiment. For details, please refer to the above method embodiment, which will not be repeated here.
[0087] The computer program product may be implemented in hardware, software or a combination thereof. In one optional embodiment, the computer program product is embodied as a computer storage medium, and in another optional embodiment, the computer program product is embodied as a software product, such as a software development kit (SDK).
[0088] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In the several embodiments provided in the present disclosure, it should be understood that the disclosed device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the device or unit can be electrical, mechanical or other forms.
[0089] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0090] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0091] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present disclosure, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0092] If the technical solution of this application involves personal information, the product using the technical solution of this application has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing the personal information. If the technical solution of this application involves sensitive personal information, the product using the technical solution of this application has obtained the individual's separate consent before processing the sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that he or she agrees to the collection of his or her personal information; or on the device that processes personal information, the personal information processing rules are notified by obvious signs / information, and the individual's authorization is obtained through pop-up information or by asking the individual to upload his or her personal information; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.
[0093] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The protection scope of the present disclosure is not limited thereto. Although the present disclosure is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed in the present disclosure, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.
Claims
1. A vehicle-side data transmission method, characterized in that: Applied to vehicles, including: After the positioning data is acquired, if the positioning position corresponding to the positioning data is within the current positioning window, the positioning data is retained; the positioning window corresponds to a preset distance length; After acquiring the perception data, determining whether the timestamp of the perception data is within the time window corresponding to the positioning window, and if so, retaining the perception data; the time window is determined based on the positioning timestamps corresponding to the starting positioning position and the ending positioning position of the positioning window respectively; For the retained perception data, determine target positioning data matching the perception data according to positioning data with at least one positioning timestamp closest to the perception timestamp of the perception data; The retained perception data and target positioning data matching the perception data are uploaded to the cloud; the perception data and target positioning data are used as reference data for controlling vehicle driving in the familiar road mode.
2. The method according to claim 1, characterized in that Determining target positioning data matching the perception data according to positioning data at at least one positioning timestamp closest to the perception timestamp of the perception data includes: Acquire positioning data with two positioning timestamps closest to the sensing timestamp of the sensing data; According to the positioning data at the two preceding and succeeding positioning timestamps and the vehicle speed information recorded between the two preceding and succeeding positioning timestamps, the predicted positioning data at the perception timestamp is determined, and the predicted positioning data is used as the target positioning data matching the perception data.
3. The method according to claim 1, characterized in that: The perception data includes road marking line data; Uploading the retained perception data and target positioning data matching the perception data to the cloud, including: Determining target marking line data located within a target distance range in front of the vehicle from the marking line data; The target marking line data and its corresponding target positioning data are uploaded to the cloud.
4. The method according to claim 3, characterized in that Uploading the target marking line data and its corresponding target positioning data to the cloud, including: Determine parameter information of an equation expression corresponding to the marking line data according to coordinate data of each feature point corresponding to the marking line data; the equation expression is used to represent a conversion relationship between the coordinate data of any feature point and a timestamp corresponding to the feature point; The parameter information and the target positioning data are uploaded to the cloud.
5. The method according to any one of claims 1 to 4, characterized in that Before uploading the retained sensing data to the cloud, the method further includes: When it is obtained that different perception data sent by multiple perception data sources include data for the same perception object, fusion processing is performed on the different perception data to obtain fused perception data for the perception object.
6. The method according to claim 5, characterized in that The fusing the different perception data to obtain fused perception data for the perception object includes: For each of the plurality of perception data sources, extracting the to-be-fused data of the perception object in multiple attribute dimensions from the perception data corresponding to the perception data source; Performing a quality evaluation on each to-be-fused data corresponding to each perception data source to obtain a quality score corresponding to the to-be-fused data; Based on the quality scores of the data to be fused, determining the target data to be fused corresponding to each attribute dimension; The target data to be fused is fused to obtain fused perception data for the perception object.
7. The method according to any one of claims 1 to 4, characterized in that The reference data for controlling the driving of the vehicle also includes lightweight map data; The method further comprises: After each vehicle travels a first preset distance, lightweight map data within a preset distance range centered on the vehicle is obtained; the lightweight map data is determined based on at least one simplified perception data; the accuracy level of the lightweight map data is lower than that of the high-definition map data; The lightweight map within the preset distance range is uploaded to the cloud.
8. The method according to any one of claims 1 to 4, characterized in that The reference data for controlling the driving of the vehicle also includes image data; The method further comprises: After the vehicle travels a second preset distance each time, determining a target grid area where the vehicle is located according to the current location of the vehicle, and determining a preset number of grid areas around the target grid area as the center; Requesting the cloud to obtain historical image data corresponding to the target grid area and a preset number of surrounding grid areas; After the current image data to be uploaded is acquired, determining whether there is image data that overlaps with the image data to be uploaded from the most recently acquired historical image data; If so, the overlapping image data are filtered out and the non-overlapping image data are uploaded to the cloud.
9. A computer device, characterized in that: include: A processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the processor is used to execute the machine-readable instructions stored in the memory. When the machine-readable instructions are executed by the processor, the processor executes the steps of the vehicle-side data feedback method as described in any one of claims 1 to 8.
10. A vehicle, characterized in that: It includes a control device, which is used to execute the steps of the vehicle-side data feedback method as described in any one of claims 1 to 8.
11. A computer program product, which, when executed by a computer device, executes the steps of the vehicle-side data feedback method as described in any one of claims 1 to 8.
Citation Information
Patent Citations
Monitoring system and method for riding violations of nonmotor vehicles
CN110085033A
Perception and planning collaboration framework for autonomous driving
CN110345955A
Automatic data labelling for autonomous driving vehicles
CN110377024A
Road end sensing method, device, road end equipment and system based on deep fusion
CN112562314A
Data display method and system, and storage medium
WO2024001573A1
