WO2022077949A1 - 一种数据处理的方法和装置 - Google Patents

一种数据处理的方法和装置 Download PDF

Info

Publication number
WO2022077949A1
WO2022077949A1 PCT/CN2021/101351 CN2021101351W WO2022077949A1 WO 2022077949 A1 WO2022077949 A1 WO 2022077949A1 CN 2021101351 W CN2021101351 W CN 2021101351W WO 2022077949 A1 WO2022077949 A1 WO 2022077949A1
Authority
WO
WIPO (PCT)
Prior art keywords
map
elements
waypoint
target
map data
Prior art date
Application number
PCT/CN2021/101351
Other languages
English (en)
French (fr)
Inventor
柴文楠
刘中元
李红军
黄亚
蒋少峰
赖健明
肖志光
Original Assignee
广州小鹏自动驾驶科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 广州小鹏自动驾驶科技有限公司 filed Critical 广州小鹏自动驾驶科技有限公司
Priority to EP21791232.8A priority Critical patent/EP4012342B1/en
Publication of WO2022077949A1 publication Critical patent/WO2022077949A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/38Electronic maps specially adapted for navigation; Updating thereof
    • G01C21/3804Creation or updating of map data
    • G01C21/3807Creation or updating of map data characterised by the type of data
    • G01C21/383Indoor data
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/005Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 with correlation of navigation data from several sources, e.g. map or contour matching
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/20Instruments for performing navigational calculations
    • G01C21/206Instruments for performing navigational calculations specially adapted for indoor navigation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/23Clustering techniques

Definitions

  • the present invention relates to the technical field of data processing, and in particular, to a method and device for data processing.
  • the map is an indispensable tool for the vehicle to help the vehicle to locate or navigate according to the data in the map.
  • the map of the target location In the process of generating the map of the target location, it can be generated according to the real-time collection of the map data of the location, as shown in Figure 1, if the target location is a multi-story floor, the collected map data can be distributed in different floors, then It is difficult to distinguish the floors where the map data collected in real time is located, making it difficult to apply the map in multiple floors for vehicle positioning or navigation.
  • a method of data processing comprising:
  • the map elements in the map data are clustered to obtain a map element cluster
  • the map elements in the map data are divided into floors.
  • the target map element includes a target waypoint element and a target landmark element
  • the target map element corresponding to the flat land mode is determined from the map data, including:
  • a target landmark element corresponding to the target waypoint element is determined.
  • the determining, according to the gradient information, the target waypoint element corresponding to the flat-land mode includes:
  • the target waypoint element corresponding to the flat-land mode is determined.
  • map elements in the map data are clustered to obtain map element clusters, including:
  • the map elements in the map data are clustered to obtain map element clusters.
  • performing floor division on map elements in the map data according to the map element clusters including:
  • floor division is performed on the map elements in the map element cluster.
  • the method before determining the target map element corresponding to the flat-land mode from the map data, the method further includes:
  • the map data is map data for a multi-storey parking lot.
  • a data processing device includes:
  • the map acquisition module is used to acquire map data
  • a map element determination module configured to determine the target map element corresponding to the flat-land mode from the map data
  • a map element cluster obtaining module is used to cluster the map elements in the map data according to the target map element to obtain a map element cluster
  • the floor division module is configured to perform floor division on the map elements in the map data according to the map element clusters.
  • a server includes a processor, a memory, and a computer program stored on the memory and capable of running on the processor, the computer program implementing the data processing method as described above when executed by the processor.
  • a computer-readable storage medium stores a computer program on the computer-readable storage medium, and when the computer program is executed by a processor, implements the above-mentioned data processing method.
  • the target map element corresponding to the flat land mode is determined, and according to the target map element, the map elements in the map data are clustered to obtain the map element Clustering, according to the map element clustering, the map elements in the map data are divided into floors, and the floor division based on the map elements is realized, so that the map data can be applied in multi-layer floors, and the map is more efficient. practicality.
  • FIG. 1 is a schematic diagram of a multi-layer floor map
  • FIG. 2 is a flowchart of steps of a data processing method provided by an embodiment of the present invention.
  • FIG. 3 is a flowchart of steps of another data processing method provided by an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of an example of a data processing method provided by an embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of a data processing apparatus according to an embodiment of the present invention.
  • FIG. 2 a flowchart of steps of a data processing method provided by an embodiment of the present invention is shown, which may specifically include the following steps:
  • Step 201 obtaining map data
  • the map data may be map data for multi-storey parking lots
  • the multi-storey parking lot may be an above-ground parking lot or an underground parking lot
  • the map data may be map data that has undergone multiple fusions
  • the map data may include one or more maps element.
  • the vehicle can collect map data for the area where the vehicle is located in real time, such as collecting map elements in the parking lot area and uploading it to the server, and then the server can receive the map data collected by different vehicles in the same area.
  • the map data collected by each vehicle and the map data preset in the server are fused to obtain map data that has undergone multiple fusions.
  • Step 202 from the map data, determine the target map element corresponding to the flat mode
  • the map elements in the map data can be divided into flat mode and ramp mode.
  • the map element corresponding to the flat mode can indicate that the map element is actually on a flat road
  • the map element corresponding to the ramp mode can indicate the map.
  • the element is actually on a ramp road
  • the map element can include waypoint elements and road sign elements
  • the target map element can include target waypoint elements and target road sign elements
  • the target waypoint element can be a flat type waypoint element
  • the target signpost element may be a flat type signpost element.
  • the waypoint element may be a point of a road in the map data.
  • the road may be divided according to a preset distance, and then multiple roads and the endpoints of the multiple roads may be obtained.
  • the road sign element may include speed bumps in the map data, Elements such as the entry point, the exit point, the entry point, the exit point, the entrance, the parking space, etc.
  • the height information of the map elements in the map data can be determined, such as the height information of the waypoint elements and the height information of the road sign elements, and then the map elements of the flat mode and the map elements of the ramp mode can be determined according to the height information. .
  • the vehicle when the vehicle collects the map data of the area where the vehicle is located, it can collect the map element of the area where the vehicle is located, as well as the location information and height information of the map element, and then any map element can be determined from the map data. location and altitude information.
  • step 202 the following steps may be further included:
  • the entrance road sign element may be the entrance and exit of a regional road, such as the entrance and exit of a parking lot
  • the second height information may be height information of the entrance road sign element
  • the entry road sign element can be determined by determining the map element with the most fusion times in the map data.
  • the location information and height information of the entrance road sign element can be determined. Since the entrance road sign element should be the entrance of the road in the area where the vehicle is located, that is, the entrance road sign element can be the starting point, the height information of the entrance road sign element can be taken as Benchmark, determine the relative height information of all map elements in the map data and the entrance road sign element, that is, the height information of all map elements in the map data can be the height information relative to the entrance road sign element, and then can perform in step 202 According to the height information Steps to determine map elements for flat mode and map elements for ramp mode.
  • Step 203 according to the target map element, clustering the map elements in the map data to obtain a map element cluster;
  • the map element cluster may be a set of map elements in which the map elements in the map data are classified according to height information.
  • clustering can be performed according to the height information of the map elements corresponding to the flat land mode. For example, map elements with height information of 1 to 3 meters can be classified into one category, which is a cluster of map elements. A, the map elements with height information of 3 to 6 meters are divided into another category, which is the map element cluster B, and then one or more map element sets can be obtained, that is, one or more map element clusters can be obtained. cluster.
  • the clustering of map elements in the map data in step 203 is performed.
  • the step if the map elements on the ramp road are considered, and the map elements in the map data may be continuous, it is difficult to cluster the map elements in the map data to obtain the map element clusters.
  • the map elements of the flat mode and the map elements of the ramp mode can be determined according to the height information, and further clustering can be performed according to the height information of the map elements corresponding to the flat mode to obtain the map element clusters.
  • clustering is performed only according to the height information of the map elements corresponding to the flat land mode, which can not only reduce the calculation amount of clustering, but also obtain the clustering of map elements on the same floor. class cluster.
  • Step 204 according to the map element clusters, perform floor division on the map elements in the map data.
  • the map elements in the map data can be divided into floors according to the height information corresponding to different map element clusters.
  • the height information corresponding to the map element cluster A may be 1 to 3 meters, and the height information corresponding to the map element cluster B may be 3 to 6 meters, and then the height information corresponding to the map element cluster A can be determined.
  • the floor is the 1st floor
  • the map elements in the map element cluster A can be the map elements of the 1st floor
  • the floor corresponding to the map element cluster B is the 2nd floor
  • the map elements in the map element cluster B can be Map elements for the 2nd floor.
  • the target map element corresponding to the flat land mode is determined, and according to the target map element, the map elements in the map data are clustered to obtain the map element Clustering, according to the map element clustering, the map elements in the map data are divided into floors, and the floor division based on the map elements is realized, so that the map data can be applied in multi-layer floors, and the map is more efficient. practicality.
  • FIG. 3 a flowchart of steps of another data processing method provided by an embodiment of the present invention is shown, which may specifically include the following steps:
  • Step 301 obtaining map data
  • Step 302 from the map data, determine a plurality of waypoint elements
  • the map data can include multiple map elements, and the map elements can include waypoint elements and road sign elements, and the waypoint elements can be points of the road in the map data, the roads can be divided according to the preset distance, Then, the multi-segment roads and the endpoints of the multi-segment roads can be obtained.
  • Step 303 determining the gradient information corresponding to the multiple waypoint elements
  • the gradient information may be angle information of multiple waypoint elements relative to the horizontal plane.
  • height information of the waypoint elements and distance information between the plurality of waypoint elements may be determined from the map data.
  • the mileage information of the vehicle can be collected, and then the mileage information corresponding to the waypoint element can be determined, and the distance information between multiple waypoint elements can be determined according to the mileage information.
  • the mileage information corresponding to the waypoint element A can be is 1001 meters
  • the mileage information corresponding to the waypoint element B may be 1003 meters
  • the distance information between the waypoint element A and the waypoint element B may be 2 meters.
  • a selection window with a preset range can be determined, for example, a selection window within a range of 5 meters can be determined, and then the path point elements within the range can be determined through the selection window.
  • the path point element A can be used as the starting point, Further, according to the selection window, a plurality of waypoint elements in the direction of the road and the distance from the waypoint element A within 5 meters can be determined.
  • a certain waypoint element may be a ramp road and the turning point of the flat road, and when selecting through the selection window, only the waypoint element on the ramp road is selected, and then the slope information corresponding to the waypoint element is determined as the slope information corresponding to the ramp road, but it should actually be a flat road. Corresponding slope information.
  • the slope information corresponding to each waypoint element can be generated according to the height information of the waypoint element and the distance information between the multiple waypoint elements.
  • a fitted straight line can be generated according to the height information of each waypoint element and the distance information between multiple waypoint elements, and then the angle information between the fitted straight line and a preset horizontal plane can be determined.
  • the fitting straight line may be a straight line including as many waypoint elements as possible, or may be a straight line that minimizes the distance between the waypoint elements and the straight line.
  • the waypoint element A and the waypoint element B can be determined according to the selection window, the height information of the waypoint element A can be determined to be 1 meter, the height information of the waypoint element B can be determined to be 2 meters, and the waypoint element A and the waypoint element can be determined. If the distance between elements B is 2 meters, it can be determined that the slope information corresponding to the selection window is 30 degrees, that is, the angle information between the fitting straight line determined by the selection window and the preset horizontal plane can be 30 degrees.
  • Step 304 according to the gradient information, determine the target waypoint element corresponding to the flat mode
  • the slope information After the slope information is determined, it can be determined whether the slope information is greater than a preset angle threshold, such as 2 degrees. When the slope information is greater than the preset angle threshold, it can be determined that the selection window is the selection window of the slope mode, and then it can be determined that the selection window is in the selection window.
  • the waypoint element of is the waypoint element corresponding to the ramp mode, that is, the waypoint element can be the waypoint element of the ramp type, and the waypoint element of the ramp type can be the waypoint on the ramp road in practice.
  • the selection window is the selection window of the flat ground mode, and then it can be determined that the path point element in the selection window is the path point element corresponding to the flat ground mode, that is, the path point element can be A waypoint element of the flat type, which can be a waypoint that is actually on a flat road.
  • a path point element can be selected by multiple selection windows, and the modes of different selection windows can be different, the modes of multiple selection windows can be determined separately, and then the number of selection windows in different modes can be determined according to the number of selection windows in different modes. Determines the type of waypoint element.
  • the waypoint element can be selected by 5 selection windows, among which, the mode corresponding to 4 selection windows can be flat mode, and the mode corresponding to 1 selection window can be ramp mode, and 4 is greater than 1, that is, the selection of flat mode There are more windows than the selection window of the ramp mode, so that the waypoint element can be determined to be a level-type waypoint element.
  • step 304 may include the following sub-steps:
  • Sub-step 11 determine the mode coefficients corresponding to the multiple waypoint elements
  • the mode coefficient may be a coefficient for selecting a window mode, and the value of the coefficient may be determined from map data.
  • the gradient information After the gradient information is determined, it can be determined whether the gradient information is greater than the preset angle threshold, and when the gradient information is greater than the preset angle threshold, it can be determined that the selection window is the selection window of the ramp mode, and the mode coefficient corresponding to the ramp mode is determined, When the gradient information is less than or equal to the preset angle threshold, it may be determined that the selection window is the selection window of the flat mode, and the mode coefficient corresponding to the ramp mode may be determined.
  • the mode coefficient of the path point element selected by the selection window may be the mode coefficient of the selection window.
  • the map data may include road information of the area where the vehicle is located, such as the degree of unevenness, and then the mode coefficient corresponding to the ramp mode and the mode coefficient corresponding to the flat mode may be determined according to the road information, for example, the corresponding mode coefficient of the ramp mode
  • the mode coefficient can be 1, and the mode coefficient corresponding to the flat ground mode can be 0.
  • Sub-step 12 determine the target waypoint element corresponding to the flat-land mode.
  • the mode coefficient After the mode coefficient is determined, it can be determined whether the waypoint element is a flat-type waypoint element according to the value of the mode coefficient.
  • the corresponding mode coefficients can also be different, and then the mode coefficients of the waypoint elements can be counted and calculated.
  • the average value of the mode coefficients can determine whether the average value of the mode coefficients is greater than the preset coefficient threshold. When the average value of the mode coefficients is greater than the preset coefficient threshold, it can be determined that the waypoint element is a ramp-type waypoint element. When the average value of is less than or equal to the preset coefficient threshold, it can be determined that the waypoint element is a waypoint element of the flat land type.
  • the preset coefficient threshold may be 0.5
  • the mode coefficient corresponding to the flat mode may be 0,
  • the mode coefficient corresponding to the ramp mode may be 0.9
  • the waypoint element may be selected by 3 selection windows, of which 2 selection windows correspond to
  • the mode can be flat mode
  • the mode corresponding to 1 selection window can be ramp mode, then it can be determined that the average value of the mode coefficient is 0.3, which is less than the preset coefficient threshold, and then it can be determined that the waypoint element is a flat-type waypoint element.
  • Step 305 determining the target signpost element corresponding to the target waypoint element
  • the road sign element corresponding to the waypoint element can be determined. Since the waypoint element is a waypoint element of the flat land type, that is, the waypoint element can be a waypoint actually on a flat road, and then It can be determined that the road sign element corresponding to the waypoint element is a road sign element of the flat land type.
  • waypoint elements may correspond to one or more road sign elements, or may not correspond to road sign elements.
  • road sign elements such as parking spaces and lane lines can be collected, and then the road sign elements can be determined.
  • the waypoint element corresponds to the waypoint element.
  • Step 306 according to the target map element, clustering the map elements in the map data to obtain a map element cluster
  • clustering can be performed according to the height information of the waypoint elements and signpost elements.
  • class which is the map element cluster A
  • the waypoint elements and road sign elements with height information of 3 to 6 meters are divided into another class, which is the map element cluster B, and then one or more map element sets can be obtained. That is, one or more map element clusters can be obtained.
  • map element clusters corresponding to all the waypoint elements and road sign elements it is necessary to perform clustering according to the waypoint elements and road sign elements corresponding to all the flat land patterns, which requires a large amount of computation, low efficiency, and high hardware requirements. .
  • the waypoint elements can be sparse, that is, a specific waypoint element can be selected from all the waypoint elements according to a preset sparse distance.
  • the sparse distance can be 5 meters, and then One waypoint element can be selected every 5 meters, and the signpost element corresponding to the selected waypoint element can be determined, and the map element cluster can be obtained according to the waypoint element and the signpost element selected according to the sparse distance.
  • the waypoint elements selected according to the preset sparse distance belong to the same map element cluster, and since the waypoint elements are continuous in the road direction, the distance between the selected waypoint elements can be determined.
  • the waypoint elements should also be the same map element cluster, and then only the selected waypoint elements can be clustered to obtain the map element clusters corresponding to all waypoint elements and road sign elements, that is, the waypoint elements are sparse. It can not only reduce the amount of calculation and improve the efficiency, but also obtain the map element clusters corresponding to all the waypoint elements and road sign elements, thereby reducing the requirements for hardware.
  • Step 307 according to the map element clusters, perform floor division on the map elements in the map data.
  • the map elements in the map data are clustered to obtain a map element cluster, and clustering is performed according to the map element.
  • the class cluster divides the map elements in the map data into floors, and realizes the floor division of elements based on the flat ground mode, so that the map data can be applied in multi-storey floors, and the practicability of the map is improved.
  • FIG. 4 a flowchart of steps of another data processing method provided by an embodiment of the present invention is shown, which may specifically include the following steps:
  • Step 401 obtaining map data
  • Step 402 from the map data, determine the target map element corresponding to the flat mode
  • Step 403 determining the first height information of the target map element
  • the first height information may be height information of the target map element, or may be the relative height information of the target map element relative to the entry road sign element.
  • the height information corresponding to the target map element can be determined from the map data, and the height information corresponding to the entry road sign element can also be determined from the map data.
  • the altitude information determines relative altitude information.
  • Step 404 according to the height information, clustering map elements in the map data to obtain map element clusters;
  • clustering can be performed according to the height information of the map elements corresponding to the flat mode.
  • map elements with height information of 1 to 3 meters can be divided into one category, which is map element cluster A, and map elements with height information of 3 to 6 meters can be divided into another category, which is map element cluster B , and then one or more map element sets can be obtained, that is, one or more map element clusters can be obtained.
  • a height range can be set, for example, the height range can be 1.5 meters, and then a plurality of map elements can be clustered according to the height range.
  • the height information of map element a can be 1 meter
  • the height information of map element b can be 2 meters
  • the height information of map element c can be 5 meters
  • the height information of map element d can be 5 meters.
  • the height information can be 6 meters, according to the height range, it can be determined that map element a and map element b are of the same type, and map element c and map element d are determined to be of the same type, and then one or more map element sets can be obtained.
  • DBSCAN Density-Based Spatial Clustering of Applications with Noise, density-based clustering algorithm
  • the scanning distance parameter in DBSCAN can be set to 1.5 meters
  • the number of samples can be set to 10
  • a set of map elements with 10 map elements and a height range of 1.5 meters can be obtained.
  • Step 405 determining the average height information of the map elements in the map element cluster
  • map element clusters After the map element clusters are determined, all map elements in the map element clusters can be determined, and the height information corresponding to the map elements can be determined, and then the map element clusters can be calculated according to the sum of the height information corresponding to all the map elements. Average height information for map elements in the cluster.
  • map element cluster A may include map element a, map element b, and map element c, and the height information of map element a may be 1 meter, the height information of map element b may be 2 meters, and the height information of map element c may be 2 meters.
  • the information can be 3 meters, and then it can be determined that the average height information of the map elements in the map element cluster A is 2 meters.
  • Step 406 determining the floor information corresponding to the average height information
  • the map element clusters can be sorted according to the average height information, and then the floors where the map element clusters are located can be determined according to the arrangement order of the average height information.
  • the average height information of map element cluster A may be 0.5 meters
  • the average height information of map element cluster B may be 6 meters
  • the average height information of map element cluster A may be 3.5 meters. Sorting by the size of the average height information, it can be obtained that the map element cluster A ⁇ map element cluster C ⁇ map element cluster B, and the average height information is greater than 0, then the corresponding map element cluster A can be determined.
  • the floor information is the first floor
  • the floor information corresponding to the map element cluster B is the third floor
  • the floor information corresponding to the map element cluster C is the second floor.
  • the average height information of the map element cluster is less than 0, it can be determined that the floor information corresponding to the map element cluster should be an underground floor.
  • the average height information of the map element cluster A can be 0.5 meters
  • the average height information of the map element cluster B can be -3 meters
  • the floor information corresponding to the map element cluster B is the negative first. layer.
  • Step 407 According to the floor information, perform floor division on the map elements in the map element cluster.
  • the map elements in the map data can be divided into floors according to the floor information.
  • the floor information corresponding to the map element cluster A may be the first floor
  • the floor information corresponding to the map element cluster B may be the second floor
  • the map elements in the map element cluster A can be is the map element of the first floor
  • the map element in the map element cluster B may be the map element of the second floor.
  • a target map element corresponding to the flat land mode is determined, first height information of the target map element is determined, and according to the height information, the map The map elements in the data are clustered to obtain a map element cluster, the average height information of the map elements in the map element cluster is determined, the floor information corresponding to the average height information is determined, and according to the floor information, the The map elements in the map element cluster are divided into floors, and the floor division is realized based on the height of the map elements, so that the map data can be applied in multi-storey floors, and the practicability of the map is improved.
  • semantic map data for the target scene can be obtained, wherein the semantic map data can include waypoint elements and road sign elements, and the road sign elements can include entry sign elements;
  • height calibration can be performed on the map data in the semantic map data based on the height information of the entrance road sign element
  • the map elements can be identified according to the calibrated height information, and then the map elements corresponding to the flat mode and the map elements corresponding to the ramp mode can be determined;
  • the map elements corresponding to the flat land mode can be sparsely selected according to the sparse distance, and then the map elements required for floor division can be selected;
  • clustering can be performed according to the calibrated height information, and then the map element clusters of different heights can be obtained;
  • the map element clusters of different heights can be sorted according to the size of the height information, and then the map elements can be divided into floors according to the sorted result.
  • FIG. 6 a schematic structural diagram of a data processing apparatus provided by an embodiment of the present invention is shown, which may specifically include the following modules:
  • a map acquisition module 601 used for acquiring map data
  • a map element determination module 602 configured to determine the target map element corresponding to the flat land mode from the map data
  • a map element cluster obtaining module 603, configured to perform clustering on the map elements in the map data according to the target map element to obtain map element clusters;
  • the floor division module 604 is configured to perform floor division on the map elements in the map data according to the map element clusters.
  • the target map element includes a target waypoint element and a target landmark element
  • the map element determination module 602 includes:
  • a waypoint element determination sub-module for determining a plurality of waypoint elements from the map data
  • a gradient information determination submodule configured to determine the gradient information corresponding to the plurality of waypoint elements
  • a target waypoint element determination sub-module used for determining the target waypoint element corresponding to the flat mode according to the slope information
  • the target landmark element determination sub-module is used for determining the target landmark element corresponding to the target waypoint element.
  • the target waypoint element determination submodule includes:
  • a mode coefficient determination unit configured to determine mode coefficients corresponding to the plurality of waypoint elements according to the gradient information
  • the corresponding element determination unit is configured to determine the target waypoint element corresponding to the flat-land mode according to the mode coefficient.
  • the map element cluster obtaining module 603 includes:
  • a first height information determination submodule configured to determine the first height information of the target map element
  • the map element clustering sub-module is used for clustering the map elements in the map data according to the height information to obtain the map element clusters.
  • the floor division module 604 includes:
  • an average height information determination submodule used for determining the average height information of map elements in the map element cluster
  • a floor information determination submodule configured to determine the floor information corresponding to the average height information
  • the map element floor division sub-module is configured to perform floor division on the map elements in the map element cluster according to the floor information.
  • the device further includes:
  • an entrance road sign element determination module configured to determine an entrance road sign element from the map data
  • a height calibration module configured to acquire second height information corresponding to the entry road sign element, and use the second height information as a reference to perform height calibration on map elements in the map data.
  • the target map element corresponding to the flat land mode is determined, and according to the target map element, the map elements in the map data are clustered to obtain the map element Clustering, according to the map element clustering, the map elements in the map data are divided into floors, and the floor division based on the map elements is realized, so that the map data can be applied in multi-layer floors, and the map is more efficient. practicality.
  • An embodiment of the present invention also provides a server, which may include a processor, a memory, and a computer program stored in the memory and capable of running on the processor.
  • the computer program is executed by the processor to implement the above data processing method.
  • An embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the above data processing method is implemented.
  • embodiments of the present invention may be provided as a method, an apparatus, or a computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
  • computer-usable storage media including, but not limited to, disk storage, CD-ROM, optical storage, etc.
  • Embodiments of the present invention are described with reference to flowcharts and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the present invention. It will be understood that each flow and/or block in the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing terminal equipment to produce a machine that causes the instructions to be executed by the processor of the computer or other programmable data processing terminal equipment Means are created for implementing the functions specified in the flow or flows of the flowcharts and/or the blocks or blocks of the block diagrams.
  • These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing terminal equipment to operate in a particular manner, such that the instructions stored in the computer readable memory result in an article of manufacture comprising instruction means, the The instruction means implement the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

Abstract

一种数据处理的方法和装置,该方法包括:获取地图数据(301),从地图数据中,确定多个路径点元素(302),确定多个路径点元素对应的坡度信息(303),根据坡度信息,确定平地模式对应的目标路径点元素(304),确定目标路径点元素对应的目标路标元素(305),根据目标地图元素,对地图数据中地图元素进行聚类,得到地图元素聚类簇(306),按照地图元素聚类簇,对地图数据中地图元素进行楼层划分(307)。通过实施例,实现了基于地图元素进行楼层划分,以使得地图数据能在多层楼层中应用,提高了地图的实用性。

Description

一种数据处理的方法和装置
本发明要求在2020年10月15日提交中国专利局、申请号202011105412.3、发明名称为“一种数据处理的方法和装置”的中国专利申请的优先权,其全部内容通过引用结合在本发明中。
技术领域
本发明涉及数据处理技术领域,特别是涉及一种数据处理的方法和装置。
背景技术
地图是车辆在行驶的过程中必不可少的工具,以帮助车辆可以根据地图中的数据进行定位或导航。
而在生成目标位置的地图的过程中,可以根据实时采集该位置的地图数据进行生成,如图1所示,倘若目标位置为多层楼层,采集的地图数据可以分布在不同的楼层中,则难以区分实时采集的地图数据所在的楼层,从而导致难以在多层楼层中应用该地图,以进行车辆的定位或导航。
发明内容
鉴于上述问题,提出了以便提供克服上述问题或者至少部分地解决上述问题的一种数据处理的方法和装置,包括:
一种数据处理的方法,所述方法包括:
获取地图数据;
从所述地图数据中,确定平地模式对应的目标地图元素;
根据所述目标地图元素,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇;
按照所述地图元素聚类簇,对所述地图数据中地图元素进行楼层划分。
可选地,所述目标地图元素包括目标路径点元素和目标路标元素,所述从所述地图数据中,确定平地模式对应的目标地图元素,包括:
从所述地图数据中,确定多个路径点元素;
确定所述多个路径点元素对应的坡度信息;
根据所述坡度信息,确定平地模式对应的目标路径点元素;
确定所述目标路径点元素对应的目标路标元素。
可选地,所述根据所述坡度信息,确定平地模式对应的目标路径点元素,包括:
根据所述坡度信息,确定所述多个路径点元素对应的模式系数;
按照所述模式系数,确定平地模式对应的目标路径点元素。
可选地,所述根据所述目标地图元素,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇,包括:
确定所述目标地图元素的第一高度信息;
根据所述高度信息,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇。
可选地,所述按照所述地图元素聚类簇,对所述地图数据中地图元素进行楼层划分,包括:
确定所述地图元素聚类簇中地图元素的平均高度信息;
确定所述平均高度信息对应的楼层信息;
按照所述楼层信息,对所述地图元素聚类簇中地图元素进行楼层划分。
可选地,在所述从所述地图数据中,确定平地模式对应的目标地图元素之前,还包括:
从所述地图数据中,确定入口路标元素;
获取所述入口路标元素对应的第二高度信息,并以所述第二高度信息为基准,对所述地图数据中地图元素进行高度校准。
可选地,所述地图数据为针对多层停车场的地图数据。
一种数据处理的装置,所述装置包括:
地图获取模块,用于获取地图数据;
地图元素确定模块,用于从所述地图数据中,确定平地模式对应的目标地图元素;
地图元素聚类簇得到模块,用于根据所述目标地图元素,对所述地图数 据中地图元素进行聚类,得到地图元素聚类簇;
楼层划分模块,用于按照所述地图元素聚类簇,对所述地图数据中地图元素进行楼层划分。
一种服务器,包括处理器、存储器及存储在所述存储器上并能够在所述处理器上运行的计算机程序,所述计算机程序被所述处理器执行时实现如上所述的数据处理的方法。
一种计算机可读存储介质,所述计算机可读存储介质上存储计算机程序,所述计算机程序被处理器执行时实现如上所述的数据处理的方法。
本发明实施例具有以下优点:
在本发明实施例中,通过获取地图数据,从所述地图数据中,确定平地模式对应的目标地图元素,根据所述目标地图元素,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇,按照所述地图元素聚类簇,对所述地图数据中地图元素进行楼层划分,实现了基于地图元素进行楼层划分,以使得地图数据能在多层楼层中应用,提高了地图的实用性。
上述说明仅是本发明技术方案的概述,为了能够更清楚了解本发明的技术手段,而可依照说明书的内容予以实施,并且为了让本发明的上述和其它目的、特征和优点能够更明显易懂,以下特举本发明的具体实施方式。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作一简单地介绍,显而易见地,下面描述中的附图是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是一种多层楼层地图的示意图;
图2是本发明一实施例提供的一种数据处理的方法的步骤流程图;
图3是本发明一实施例提供的另一种数据处理的方法的步骤流程图;
图4是本发明一实施例提供的又一种数据处理的方法的步骤流程图;
图5是本发明一实施例提供的一种数据处理的方法的实例示意图;
图6是本发明一实施例提供的一种数据处理的装置的结构示意图。
具体实施例
为使本发明实施例的目的、技术方案和优点更加清楚,下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
参照图2,示出了本发明一实施例提供的一种数据处理的方法的步骤流程图,具体可以包括如下步骤:
步骤201,获取地图数据;
其中,地图数据可以为针对多层停车场的地图数据,多层停车场可以为地上停车场或地下停车场,地图数据可以为经过多次融合的地图数据,地图数据可以包括一个或多个地图元素。
在实际应用中,车辆可以实时采集针对车辆所在区域的地图数据,如采集停车场区域中的地图元素,并上传至服务器中,进而服务器可以接收不同车辆在同一区域采集的地图数据,以根据多个车辆采集的地图数据以及预置在服务器中的地图数据进行地图融合,可以得到经过多次融合的地图数据。
步骤202,从所述地图数据中,确定平地模式对应的目标地图元素;
其中,地图数据中的地图元素可以分为平地模式和坡道模式,平地模式对应的地图元素可以表示该地图元素在实际中是处于平地道路上的,坡道模式对应的地图元素可以表示该地图元素在实际中是处于坡道道路上的,地图元素可以包括路径点元素和路标元素,目标地图元素可以包括目标路径点元素和目标路标元素,目标路径点元素可以为平地类型的路径点元素,目标路标元素可以为平地类型的路标元素。
作为一示例,路径点元素可以为地图数据中的道路的点,例如,可以按照预设的距离划分道路,进而可以得到多段道路以及多段道路的端点,路标元素可以包括地图数据中的减速带、入弯点、出弯点、入坡点、出坡点、入口、车位等元素。
在获取地图数据后,可以确定地图数据中地图元素的高度信息,如确定 路径点元素的高度信息和路标元素的高度信息,进而可以根据高度信息确定平地模式的地图元素和坡道模式的地图元素。
在实际应用中,车辆在采集车辆所在的区域的地图数据时,可以采集车辆所在的区域的地图元素,以及采集该地图元素的位置信息和高度信息,进而可以从地图数据中,确定任意地图元素的位置信息和高度信息。
在本发明一实施例中,在步骤202之前,还可以包括如下步骤:
从地图数据中,确定入口路标元素,获取入口路标元素对应的第二高度信息,并以第二高度信息为基准,对地图数据中地图元素进行高度校准。
其中,入口路标元素可以为区域道路的出入口,如停车场的出入口,第二高度信息可以为入口路标元素的高度信息。
在实际应用中,由于不同的车辆进入区域道路时,都会经过该区域道路的出入口,即每个车辆采集的地图数据应当都有该出入口对应的地图元素,进而每次地图融合时都会融合该出入口对应的地图元素,则可以通过确定地图数据中融合次数最多的地图元素,来确定入口路标元素。
在确定入口路标元素后,可以确定入口路标元素的位置信息和高度信息,由于入口路标元素应当为车辆所在区域道路的入口,即入口路标元素可以是起点,则可以以入口路标元素的高度信息为基准,确定地图数据中所有地图元素与入口路标元素的相对高度信息,即地图数据中所有的地图元素的高度信息可以为相对于入口路标元素的高度信息,进而可以在步骤202中执行根据高度信息确定平地模式的地图元素和坡道模式的地图元素的步骤。
步骤203,根据所述目标地图元素,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇;
其中,地图元素聚类簇可以为地图数据中的地图元素针对高度信息进行分类的地图元素集合。
在确定平地模式对应的地图元素后,可以根据平地模式对应的地图元素的高度信息进行聚类,例如,可以将高度信息为1到3米的地图元素分为一类,为地图元素聚类簇A,高度信息为3到6米的地图元素分为另一类,为地图元素聚类簇B,进而可以得到一个或多个地图元素集合,也即是可以得到一个或多个地图元素聚类簇。
在实际应用中,由于地图数据中的地图元素实际上可以是处于平地道路上的元素或者是处于坡道道路上的元素,因此,在执行步骤203中的对地图数据中地图元素进行聚类的步骤时,倘若考虑到处于坡道道路上的地图元素,而地图数据中的地图元素可以是连续的,则难以对地图数据中地图元素进行聚类,以得到地图元素聚类簇。
而在本发明实施例中,可以根据高度信息确定平地模式的地图元素和坡道模式的地图元素,进而可以根据平地模式对应的地图元素的高度信息进行聚类,以得到地图元素聚类簇。
而且,同一楼层中的地图元素应当是平地类型的地图元素,则仅根据平地模式对应的地图元素的高度信息进行聚类,既可以减少聚类的计算量,还可以得到同一楼层的地图元素聚类簇。
步骤204,按照所述地图元素聚类簇,对所述地图数据中地图元素进行楼层划分。
在得到地图元素聚类簇后,可以按照不同的地图元素聚类簇所对应的高度信息,对地图数据中的地图元素进行楼层划分。
例如,地图元素聚类簇A所对应的高度信息可以为1到3米,地图元素聚类簇B所对应的高度信息可以为3到6米,进而可以确定地图元素聚类簇A所对应的楼层为1层,地图元素聚类簇A中的地图元素可以为1层楼层的地图元素,地图元素聚类簇B所对应的楼层为2层,地图元素聚类簇B中的地图元素可以为2层楼层的地图元素。
在本发明实施例中,通过获取地图数据,从所述地图数据中,确定平地模式对应的目标地图元素,根据所述目标地图元素,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇,按照所述地图元素聚类簇,对所述地图数据中地图元素进行楼层划分,实现了基于地图元素进行楼层划分,以使得地图数据能在多层楼层中应用,提高了地图的实用性。
参照图3,示出了本发明一实施例提供的另一种数据处理的方法的步骤流程图,具体可以包括如下步骤:
步骤301,获取地图数据;
步骤302,从所述地图数据中,确定多个路径点元素;
在实际应用中,由于地图数据可以包括多个地图元素,且地图元素可以包括路径点元素和路标元素,路径点元素可以为地图数据中的道路的点,则可以按照预设的距离划分道路,进而可以得到多段道路以及多段道路的端点。
步骤303,确定所述多个路径点元素对应的坡度信息;
其中,坡度信息可以为多个路径点元素相对于水平平面的角度信息。
在确定多个路径点元素后,可以从地图数据中确定路径点元素的高度信息,以及确定多个路径点元素之间的距离信息。
在实际应用中,可以采集车辆的里程信息,进而可以确定路径点元素对应的里程信息,可以根据里程信息确定多个路径点元素之间的距离信息,例如,路径点元素A对应的里程信息可以为1001米,路径点元素B对应的里程信息可以为1003米,则路径点元素A与路径点元素B之间的距离信息可以为2米。
在确定距离信息后,可以确定具有预设范围的选取窗口,如确定5米范围内的选取窗口,进而可以通过选取窗口确定范围内的路径点元素,例如,可以以路径点元素A为起点,进而可以根据选取窗口,确定在道路方向上且与路径点元素A的距离在5米内的多个路径点元素。
在本发明一实施例中,倘若路径点元素不被重复选取,即一个路径点元素只对应一个选取窗口,则容易在确定坡度信息时出现误差,例如,某个路径点元素可以为坡道道路与平地道路的转折点,而通过选取窗口选取时,仅选取坡道道路上的路径点元素,进而确定该路径点元素对应的坡度信息为坡道道路对应的坡度信息,而实际上应当是平地道路对应的坡度信息。
因此,可以在以路径点元素A为起点进行选取后,继续以路径点元素A相邻的路径点元素B为起点,确定在道路方向上且与路径点元素B的距离在预设范围内的多个路径点元素,直至确定以最后一个路径点元素为起点所选取的路径点元素。
在根据选取窗口确定多个路径点元素后,可以根据路径点元素的高度信息以及多个路径点元素之间的距离信息生成每个路径点元素对应的坡度信 息。
在实际应用中,可以根据每个路径点元素的高度信息和多个路径点元素之间的距离信息生成一拟合直线,进而可以确定拟合直线与预置的水平平面之间的角度信息。
其中,拟合直线可以为包括尽可能多的路径点元素的直线,也可以是使得路径点元素相对于该直线的距离最小的直线。
例如,根据选取窗口可以确定路径点元素A和路径点元素B,可以确定路径点元素A的高度信息为1米,路径点元素B的高度信息为2米,以及确定路径点元素A与路径点元素B之间的距离为2米,则可以确定该选取窗口对应的坡度信息为30度,即由该选取窗口确定的拟合直线与预置的水平平面之间的角度信息可以为30度。
步骤304,根据所述坡度信息,确定平地模式对应的目标路径点元素;
在确定坡度信息后,可以判断坡度信息是否大于预设角度阈值,如2度,在坡度信息大于预设角度阈值时,可以确定该选取窗口为坡道模式的选取窗口,进而可以确定选取窗口中的路径点元素为坡道模式对应的路径点元素,即该路径点元素可以为坡道类型的路径点元素,坡道类型的路径点元素可以是在实际中处于坡道道路上的路径点。
在坡度信息小于或等于预设角度阈值时,可以确定该选取窗口为平地模式的选取窗口,进而可以确定选取窗口中的路径点元素为平地模式对应的路径点元素,即该路径点元素可以为平地类型的路径点元素,平地类型的路径点元素可以是在实际中处于平地道路上的路径点。
在实际应用中,由于一个路径点元素可以被多个选取窗口选取,且不同的选取窗口的模式可以不同,则可以分别确定多个选取窗口的模式,进而可以根据不同模式的选取窗口的个数确定路径点元素的类型。
例如,路径点元素可以被5个选取窗口选取,其中,4个选取窗口对应的模式可以为平地模式,1个选取窗口对应的模式可以为坡道模式,且4大于1,即平地模式的选取窗口多于坡道模式的选取窗口,进而可以确定该路径点元素为平地类型的路径点元素。
在本发明一实施例中,步骤304可以包括如下子步骤:
子步骤11,根据所述坡度信息,确定所述多个路径点元素对应的模式系数;
其中,模式系数可以为选取窗口模式的系数,该系数的数值可以从地图数据中确定。
在确定坡度信息后,可以判断坡度信息是否大于预设角度阈值,在坡度信息大于预设角度阈值时,可以确定该选取窗口为坡道模式的选取窗口,以及确定坡道模式对应的模式系数,在坡度信息小于或等于预设角度阈值时,可以确定该选取窗口为平地模式的选取窗口,以及确定坡道模式对应的模式系数。
其中,选取窗口所选取的路径点元素的模式系数可以为选取窗口的模式系数。
在实际应用中,地图数据中可以包括车辆所在区域的道路信息,如凹凸程度,进而可以根据道路信息确定坡道模式对应的模式系数,以及平地模式对应的模式系数,例如,坡道模式对应的模式系数可以为1,平地模式对应的模式系数可以为0。
子步骤12,按照所述模式系数,确定平地模式对应的目标路径点元素。
在确定模式系数后,可以按照模式系数的数值确定路径点元素是否为平地类型的路径点元素。
在实际应用中,由于一个路径点元素可以被多个选取窗口选取,且不同的选取窗口的模式可以不同,则对应的模式系数也可以不同,进而可以统计路径点元素的模式系数,并求出模式系数的平均值,可以判断模式系数的平均值是否大于预设系数阈值,在模式系数的平均值大于预设系数阈值时,可以确定路径点元素为坡道类型的路径点元素,在模式系数的平均值小于或等于预设系数阈值时,可以确定路径点元素为平地类型的路径点元素。
例如,预设系数阈值可以为0.5,平地模式对应的模式系数可以为0,坡道模式对应的模式系数可以为0.9,且路径点元素可以被3个选取窗口选取,其中,2个选取窗口对应的模式可以为平地模式,1个选取窗口对应的模式可以为坡道模式,则可以确定模式系数的平均值为0.3,小于预设系数阈值,进而可以确定该路径点元素为平地类型的路径点元素。
步骤305,确定所述目标路径点元素对应的目标路标元素;
在确定平地类型的路径点元素后,可以确定与路径点元素对应的路标元素,由于路径点元素是平地类型的路径点元素,即该路径点元素可以是实际上在平地道路的路径点,进而可以确定与该路径点元素对应的路标元素为平地类型的路标元素。
在实际应用中,路径点元素可以对应一个或多个路标元素,也可以不对应路标元素,例如,在垂直道路的方向上,可以采集有车位、车道线等路标元素,进而可以确定该道路中的路径点元素对应的路标元素。
步骤306,根据所述目标地图元素,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇;
在确定平地模式对应的路径点元素和路标元素后,可以根据路径点元素和路标元素的高度信息进行聚类,例如,可以将高度信息为1到3米的路径点元素和路标元素分为一类,为地图元素聚类簇A,高度信息为3到6米的路径点元素和路标元素分为另一类,为地图元素聚类簇B,进而可以得到一个或多个地图元素集合,也即是可以得到一个或多个地图元素聚类簇。
在实际应用中,由于平地模式对应的路径点元素和路标元素可以有多个,进而根据全部平地模式对应的路径点元素和路标元素进行聚类,可以得到全部路径点元素和路标元素对应的地图元素聚类簇。
然而,得到全部路径点元素和路标元素对应的地图元素聚类簇,需要根据全部平地模式对应的路径点元素和路标元素进行聚类,计算量较大,效率较低,对硬件的要求较高。
在本发明一实施例中,可以将路径点元素稀疏化,也即是可以按照预设的稀疏距离从全部的路径点元素中选取特定的路径点元素,例如,稀疏距离可以为5米,进而可以每隔5米选取一个路径点元素,以及确定与选取的路径点元素对应的路标元素,可以根据按照稀疏距离选取的路径点元素和路标元素得到地图元素聚类簇。
在实际应用中,倘若按照预设的稀疏距离选取的路径点元素为同一个地图元素聚类簇,且由于路径点元素在道路方向上是连续的,则可以确定选取的路径点元素之间的路径点元素也应当为同一个地图元素聚类簇,进而可以 仅通过对选取的路径点元素进行聚类,得到全部路径点元素和路标元素对应的地图元素聚类簇,即将路径点元素稀疏化既能减少计算量,提高效率,还能得到全部路径点元素和路标元素对应的地图元素聚类簇,从而降低了对硬件的要求。
步骤307,按照所述地图元素聚类簇,对所述地图数据中地图元素进行楼层划分。
在本发明实施例中,通过获取地图数据,从所述地图数据中,确定多个路径点元素,确定所述多个路径点元素对应的坡度信息,根据所述坡度信息,确定平地模式对应的目标路径点元素,确定所述目标路径点元素对应的目标路标元素,根据所述目标地图元素,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇,按照所述地图元素聚类簇,对所述地图数据中地图元素进行楼层划分,实现了基于平地模式的元素进行楼层划分,以使得地图数据能在多层楼层中应用,提高了地图的实用性。
参照图4,示出了本发明一实施例提供的又一种数据处理的方法的步骤流程图,具体可以包括如下步骤:
步骤401,获取地图数据;
步骤402,从所述地图数据中,确定平地模式对应的目标地图元素;
步骤403,确定所述目标地图元素的第一高度信息;
其中,第一高度信息可以为目标地图元素的高度信息,也可以为目标地图元素相对于入口路标元素的相对高度信息。
在确定目标地图元素后,可以从地图数据中确定目标地图元素对应的高度信息,也可以从地图数据中确定入口路标元素对应的高度信息,进而可以根据目标地图元素的高度信息和入口路标元素的高度信息确定相对高度信息。
步骤404,根据所述高度信息,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇;
在确定高度信息后,可以根据平地模式对应的地图元素的高度信息进行聚类。
例如,可以将高度信息为1到3米的地图元素分为一类,为地图元素聚类簇A,高度信息为3到6米的地图元素分为另一类,为地图元素聚类簇B,进而可以得到一个或多个地图元素集合,也即是可以得到一个或多个地图元素聚类簇。
在本发明一实施例中,可以设置高度范围,如高度范围可以为1.5米,进而可以根据高度范围对多个地图元素进行聚类。
例如,地图数据中可以有4个目标地图元素,且地图元素a的高度信息可以为1米,地图元素b的高度信息可以为2米,地图元素c的高度信息可以为5米,地图元素d的高度信息可以为6米,则可以根据高度范围,确定地图元素a与地图元素b为同一类,确定地图元素c与地图元素d为同一类,进而可以得到一个或多个地图元素集合。
在实际应用中,可以使用DBSCAN(Density-Based Spatial Clustering of Applications with Noise,基于密度的聚类算法)进行聚类,可以将DBSCAN中的扫描距离参数设定为1.5米,样本数设定为10个,进而可以得到地图元素为10个、高度范围为1.5米的多个地图元素集合。
步骤405,确定所述地图元素聚类簇中地图元素的平均高度信息;
在确定地图元素聚类簇后,可以确定地图元素聚类簇中的全部地图元素,以及确定地图元素对应的高度信息,进而可以根据全部地图元素对应的高度信息的总和,计算出地图元素聚类簇中地图元素的平均高度信息。
例如,地图元素聚类簇A可以包括地图元素a、地图元素b以及地图元素c,且地图元素a的高度信息可以为1米,地图元素b的高度信息可以为2米,地图元素c的高度信息可以为3米,进而可以确定地图元素聚类簇A中地图元素的平均高度信息为2米。
步骤406,确定所述平均高度信息对应的楼层信息;
在确定平均高度信息后,可以根据平均高度信息对地图元素聚类簇进行排序,进而可以根据平均高度信息的排列顺序确定地图元素聚类簇所在的楼层。
例如,地图元素聚类簇A的平均高度信息可以为0.5米,地图元素聚类簇B的平均高度信息可以为6米,地图元素聚类簇A的平均高度信息可以 为3.5米,进而可以根据平均高度信息的大小进行排序,可以得到地图元素聚类簇A<地图元素聚类簇C<地图元素聚类簇B,且平均高度信息均大于0,则可以确定地图元素聚类簇A对应的楼层信息为第一层,地图元素聚类簇B对应的楼层信息为第三层,地图元素聚类簇C对应的楼层信息为第二层。
在实际应用中,若地图元素聚类簇的平均高度信息小于0,则可以确定该地图元素聚类簇对应的楼层信息应该为地下楼层,例如,地图元素聚类簇A的平均高度信息可以为0.5米,地图元素聚类簇B的平均高度信息可以为-3米,则可以确定地图元素聚类簇A对应的楼层信息为第一层,地图元素聚类簇B对应的楼层信息为第负一层。
步骤407,按照所述楼层信息,对所述地图元素聚类簇中地图元素进行楼层划分。
在得到对应的楼层信息后,可以按照楼层信息对地图数据中的地图元素进行楼层划分。
例如,地图元素聚类簇A所对应的楼层信息可以为第一层,地图元素聚类簇B所对应的楼层信息可以为第二层,进而可以确定地图元素聚类簇A中的地图元素可以为第一层楼层的地图元素,地图元素聚类簇B中的地图元素可以为第二层楼层的地图元素。
在本发明实施例中,通过获取地图数据,从所述地图数据中,确定平地模式对应的目标地图元素,确定所述目标地图元素的第一高度信息,根据所述高度信息,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇,确定所述地图元素聚类簇中地图元素的平均高度信息,确定所述平均高度信息对应的楼层信息,按照所述楼层信息,对所述地图元素聚类簇中地图元素进行楼层划分,实现了基于地图元素的高度进行楼层划分,以使得地图数据能在多层楼层中应用,提高了地图的实用性。
以下结合图5对本发明实施例进行示例性说明:
1、在实际应用中,可以获取针对目标场景的语义地图数据(Venue-map),其中,语义地图数据可以包括路径点元素和路标元素,路标元素可以包括入口路标元素;
2、在获取语义地图数据后,可以以入口路标元素的高度信息为基准,对语义地图数据中的地图数据进行高度校准;
3、在高度校准后,可以根据校准后的高度信息对地图元素进行识别,进而可以确定平地模式对应的地图元素和坡道模式对应的地图元素;
4、在确定平地模式对应的地图元素和坡道模式对应的地图元素后,可以按照稀疏距离对平地模式对应的地图元素进行稀疏化选取,进而可以选取进行楼层划分所需的地图元素;
5、在选取进行楼层划分所需的地图元素后,可以根据校准后的高度信息进行聚类,进而可以得到不同高度的地图元素聚类簇;
6、在得到不同高度的地图元素聚类簇后,可以根据高度信息的大小对地图元素聚类簇进行排序,进而可以根据排序后的结果对地图元素进行楼层划分。
需要说明的是,对于方法实施例,为了简单描述,故将其都表述为一系列的动作组合,但是本领域技术人员应该知悉,本发明实施例并不受所描述的动作顺序的限制,因为依据本发明实施例,某些步骤可以采用其他顺序或者同时进行。其次,本领域技术人员也应该知悉,说明书中所描述的实施例均属于优选实施例,所涉及的动作并不一定是本发明实施例所必须的。
参照图6,示出了本发明一实施例提供的一种数据处理的装置的结构示意图,具体可以包括如下模块:
地图获取模块601,用于获取地图数据;
地图元素确定模块602,用于从所述地图数据中,确定平地模式对应的目标地图元素;
地图元素聚类簇得到模块603,用于根据所述目标地图元素,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇;
楼层划分模块604,用于按照所述地图元素聚类簇,对所述地图数据中地图元素进行楼层划分。
在本发明一实施例中,所述目标地图元素包括目标路径点元素和目标路 标元素,所述地图元素确定模块602,包括:
路径点元素确定子模块,用于从所述地图数据中,确定多个路径点元素;
坡度信息确定子模块,用于确定所述多个路径点元素对应的坡度信息;
目标路径点元素确定子模块,用于根据所述坡度信息,确定平地模式对应的目标路径点元素;
目标路标元素确定子模块,用于确定所述目标路径点元素对应的目标路标元素。
在本发明一实施例中,所述目标路径点元素确定子模块,包括:
模式系数确定单元,用于根据所述坡度信息,确定所述多个路径点元素对应的模式系数;
对应元素确定单元,用于按照所述模式系数,确定平地模式对应的目标路径点元素。
在本发明一实施例中,所述地图元素聚类簇得到模块603,包括:
第一高度信息确定子模块,用于确定所述目标地图元素的第一高度信息;
地图元素聚类子模块,用于根据所述高度信息,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇。
在本发明一实施例中,所述楼层划分模块604,包括:
平均高度信息确定子模块,用于确定所述地图元素聚类簇中地图元素的平均高度信息;
楼层信息确定子模块,用于确定所述平均高度信息对应的楼层信息;
地图元素楼层划分子模块,用于按照所述楼层信息,对所述地图元素聚类簇中地图元素进行楼层划分。
在本发明一实施例中,所述装置还包括:
入口路标元素确定模块,用于从所述地图数据中,确定入口路标元素;
高度校准模块,用于获取所述入口路标元素对应的第二高度信息,并以所述第二高度信息为基准,对所述地图数据中地图元素进行高度校准。
在本发明实施例中,通过获取地图数据,从所述地图数据中,确定平地模式对应的目标地图元素,根据所述目标地图元素,对所述地图数据中地图 元素进行聚类,得到地图元素聚类簇,按照所述地图元素聚类簇,对所述地图数据中地图元素进行楼层划分,实现了基于地图元素进行楼层划分,以使得地图数据能在多层楼层中应用,提高了地图的实用性。
本发明一实施例还提供了一种服务器,可以包括处理器、存储器及存储在存储器上并能够在处理器上运行的计算机程序,计算机程序被处理器执行时实现如上数据处理的方法。
本发明一实施例还提供了一种计算机可读存储介质,计算机可读存储介质上存储计算机程序,计算机程序被处理器执行时实现如上数据处理的方法。
对于装置实施例而言,由于其与方法实施例基本相似,所以描述的比较简单,相关之处参见方法实施例的部分说明即可。
本说明书中的各个实施例均采用递进的方式描述,每个实施例重点说明的都是与其他实施例的不同之处,各个实施例之间相同相似的部分互相参见即可。
本领域内的技术人员应明白,本发明实施例可提供为方法、装置、或计算机程序产品。因此,本发明实施例可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且,本发明实施例可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本发明实施例是参照根据本发明实施例的方法、终端设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理终端设备的处理器以产生 一个机器,使得通过计算机或其他可编程数据处理终端设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理终端设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理终端设备上,使得在计算机或其他可编程终端设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程终端设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
尽管已描述了本发明实施例的优选实施例,但本领域内的技术人员一旦得知了基本创造性概念,则可对这些实施例做出另外的变更和修改。所以,所附权利要求意欲解释为包括优选实施例以及落入本发明实施例范围的所有变更和修改。
最后,还需要说明的是,在本文中,诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来,而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者终端设备不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者终端设备所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括所述要素的过程、方法、物品或者终端设备中还存在另外的相同要素。
以上对所提供的一种数据处理的方法和装置,进行了详细介绍,本文中应用了具体个例对本发明的原理及实施方式进行了阐述,以上实施例的说明只是用于帮助理解本发明的方法及其核心思想;同时,对于本领域的一般技 术人员,依据本发明的思想,在具体实施方式及应用范围上均会有改变之处,综上所述,本说明书内容不应理解为对本发明的限制。

Claims (12)

  1. 一种数据处理的方法,其特征在于,所述方法包括:
    获取地图数据;
    从所述地图数据中,确定平地模式对应的目标地图元素;
    根据所述目标地图元素,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇;
    按照所述地图元素聚类簇,对所述地图数据中地图元素进行楼层划分。
  2. 根据权利要求1所述的方法,其特征在于,所述目标地图元素包括目标路径点元素和目标路标元素,所述从所述地图数据中,确定平地模式对应的目标地图元素,包括:
    从所述地图数据中,确定多个路径点元素;
    确定所述多个路径点元素对应的坡度信息;
    根据所述坡度信息,确定平地模式对应的目标路径点元素;
    确定所述目标路径点元素对应的目标路标元素。
  3. 根据权利要求2所述的方法,其特征在于,所述根据所述坡度信息,确定平地模式对应的目标路径点元素,包括:
    根据所述坡度信息,确定所述多个路径点元素对应的模式系数;
    按照所述模式系数,确定平地模式对应的目标路径点元素。
  4. 根据权利要求1或2或3所述的方法,其特征在于,所述根据所述目标地图元素,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇,包括:
    确定所述目标地图元素的第一高度信息;
    根据所述高度信息,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇。
  5. 根据权利要求4所述的方法,其特征在于,所述按照所述地图元素聚类簇,对所述地图数据中地图元素进行楼层划分,包括:
    确定所述地图元素聚类簇中地图元素的平均高度信息;
    确定所述平均高度信息对应的楼层信息;
    按照所述楼层信息,对所述地图元素聚类簇中地图元素进行楼层划分。
  6. 根据权利要求1所述的方法,其特征在于,在所述从所述地图数据中,确定平地模式对应的目标地图元素之前,还包括:
    从所述地图数据中,确定入口路标元素;
    获取所述入口路标元素对应的第二高度信息,并以所述第二高度信息为基准,对所述地图数据中地图元素进行高度校准。
  7. 根据权利要求1所述的方法,其特征在于,所述地图数据为针对多层停车场的地图数据。
  8. 一种数据处理的装置,其特征在于,所述装置包括:
    地图获取模块,用于获取地图数据;
    地图元素确定模块,用于从所述地图数据中,确定平地模式对应的目标地图元素;
    地图元素聚类簇得到模块,用于根据所述目标地图元素,对所述地图数据中地图元素进行聚类,得到地图元素聚类簇;
    楼层划分模块,用于按照所述地图元素聚类簇,对所述地图数据中地图元素进行楼层划分。
  9. 根据权利要求8所述的装置,其特征在于,所述地图元素确定模块,包括:
    路径点元素确定子模块,用于从所述地图数据中,确定多个路径点元素;
    坡度信息确定子模块,用于确定所述多个路径点元素对应的坡度信息;
    目标路径点元素确定子模块,用于根据所述坡度信息,确定平地模式对应的目标路径点元素;
    目标路标元素确定子模块,用于确定所述目标路径点元素对应的目标路标元素。
  10. 根据权利要求9所述的装置,其特征在于,所述目标路径点元素确定子模块,包括:
    模式系数确定单元,用于根据所述坡度信息,确定所述多个路径点元素对应的模式系数;
    对应元素确定单元,用于按照所述模式系数,确定平地模式对应的目标 路径点元素。
  11. 一种服务器,其特征在于,包括处理器、存储器及存储在所述存储器上并能够在所述处理器上运行的计算机程序,所述计算机程序被所述处理器执行时实现如权利要求1至7中任一项所述的数据处理的方法。
  12. 一种计算机可读存储介质,其特征在于,所述计算机可读存储介质上存储计算机程序,所述计算机程序被处理器执行时实现如权利要求1至7中任一项所述的数据处理的方法。
PCT/CN2021/101351 2020-10-15 2021-06-21 一种数据处理的方法和装置 WO2022077949A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP21791232.8A EP4012342B1 (en) 2020-10-15 2021-06-21 Data processing method and device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011105412.3A CN112229411B (zh) 2020-10-15 2020-10-15 一种数据处理的方法和装置
CN202011105412.3 2020-10-15

Publications (1)

Publication Number Publication Date
WO2022077949A1 true WO2022077949A1 (zh) 2022-04-21

Family

ID=74117764

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/101351 WO2022077949A1 (zh) 2020-10-15 2021-06-21 一种数据处理的方法和装置

Country Status (3)

Country Link
EP (1) EP4012342B1 (zh)
CN (1) CN112229411B (zh)
WO (1) WO2022077949A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112229411B (zh) * 2020-10-15 2021-12-07 广州小鹏自动驾驶科技有限公司 一种数据处理的方法和装置
CN112927552B (zh) * 2021-01-20 2022-03-11 广州小鹏自动驾驶科技有限公司 一种车位检测的方法和装置
CN113532450B (zh) * 2021-06-29 2024-04-30 广州小鹏汽车科技有限公司 一种虚拟停车地图数据处理方法和系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130297198A1 (en) * 2010-12-20 2013-11-07 Tomtom Belgium N.V. Method for generating map data
CN110914639A (zh) * 2017-07-10 2020-03-24 奥迪股份公司 产生和更新至少一个建筑物的至少一个空间的拓扑地图的数据生成方法
US20200292318A1 (en) * 2019-03-14 2020-09-17 Here Global B.V. Multi-level altitude map
CN112229411A (zh) * 2020-10-15 2021-01-15 广州小鹏自动驾驶科技有限公司 一种数据处理的方法和装置

Family Cites Families (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006234418A (ja) * 2005-02-22 2006-09-07 Toyota Motor Corp 車両位置案内装置及びシステム、並びに自動遠隔操作装置及びシステム
CN103152697A (zh) * 2013-03-25 2013-06-12 南京大学 一种利用智能手机Wi-Fi功能实现楼层自动定位的方法
CN204242370U (zh) * 2014-07-19 2015-04-01 安徽理工大学 安装在高楼中的高楼火灾受困人员搜救系统
US9478134B2 (en) * 2014-10-10 2016-10-25 General Motors Llc Method of determining an attribute of a parking structure
CN106441305B (zh) * 2016-10-09 2019-11-08 北京超图软件股份有限公司 室内跨楼层的路径规划方法和装置
CN110226186B (zh) * 2016-12-26 2023-08-04 宝马股份公司 表示地图元素的方法和装置以及定位的方法和装置
US10694325B2 (en) * 2016-12-31 2020-06-23 Google Llc Determining position of a device in three-dimensional space and corresponding calibration techniques
CN106842272B (zh) * 2017-03-01 2020-04-10 深圳市浩科电子有限公司 停车场内生成停车轨迹的方法、系统及装置
US9838850B2 (en) * 2017-05-12 2017-12-05 Mapsted Corp. Systems and methods for determining indoor location and floor of a mobile device
JP6969962B2 (ja) * 2017-10-05 2021-11-24 トヨタ自動車株式会社 車両の運転支援及び/又は走行制御のための地図情報提供システム
CN108332761B (zh) * 2018-02-07 2022-10-28 驭势(上海)汽车科技有限公司 一种使用及创建路网地图信息的方法与设备
CN108717796B (zh) * 2018-03-21 2021-05-07 厦门雅迅网络股份有限公司 一种车辆的地理坡道自学习方法和系统
CN110657804B (zh) * 2018-06-29 2023-10-20 微软技术许可有限责任公司 室内位置服务
CN108966131A (zh) * 2018-07-23 2018-12-07 广州都市圈网络科技有限公司 基于室内定位的融合楼层推断方法
US11361247B2 (en) * 2018-10-01 2022-06-14 International Business Machines Corporation Spatial device clustering-based emergency response floor identification
US10631129B1 (en) * 2018-10-01 2020-04-21 International Business Machines Corporation First responder feedback-based emergency response floor identification
US20200132473A1 (en) * 2018-10-26 2020-04-30 Ford Global Technologies, Llc Systems and methods for determining vehicle location in parking structures
EP3693699B1 (en) * 2019-02-05 2021-12-22 HERE Global B.V. Determining of absolute altitudes of floors of a structure
CN110175654B (zh) * 2019-05-29 2021-06-01 广州小鹏自动驾驶科技有限公司 一种轨迹路标的更新方法及系统
CN110415549A (zh) * 2019-07-26 2019-11-05 戴姆勒股份公司 用于在停车场中控制车辆和移动终端的方法、装置和系统
CN110455300B (zh) * 2019-09-03 2021-02-19 广州小鹏汽车科技有限公司 导航方法、导航显示方法、装置、车辆及机器可读介质
CN110517531B (zh) * 2019-09-05 2021-08-17 武汉中海庭数据技术有限公司 一种基于高精度地图数据的多层停车场定位方法
CN110926473B (zh) * 2019-11-18 2022-08-16 北京三快在线科技有限公司 识别楼层的方法、装置、电子设备及存储介质
CN110888866B (zh) * 2019-12-05 2023-01-17 秒针信息技术有限公司 数据扩充方法、装置、数据处理设备及存储介质
CN110986985B (zh) * 2019-12-17 2022-07-12 广州小鹏汽车科技有限公司 车辆行程推送方法、装置、介质、控制终端及汽车
CN111735429B (zh) * 2020-05-11 2021-11-26 腾讯科技(深圳)有限公司 楼层识别方法、装置、计算机设备和存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130297198A1 (en) * 2010-12-20 2013-11-07 Tomtom Belgium N.V. Method for generating map data
CN110914639A (zh) * 2017-07-10 2020-03-24 奥迪股份公司 产生和更新至少一个建筑物的至少一个空间的拓扑地图的数据生成方法
US20200292318A1 (en) * 2019-03-14 2020-09-17 Here Global B.V. Multi-level altitude map
CN112229411A (zh) * 2020-10-15 2021-01-15 广州小鹏自动驾驶科技有限公司 一种数据处理的方法和装置

Also Published As

Publication number Publication date
EP4012342A4 (en) 2022-10-19
EP4012342B1 (en) 2024-02-07
EP4012342C0 (en) 2024-02-07
CN112229411B (zh) 2021-12-07
EP4012342A1 (en) 2022-06-15
CN112229411A (zh) 2021-01-15

Similar Documents

Publication Publication Date Title
WO2022077949A1 (zh) 一种数据处理的方法和装置
US10302438B2 (en) Probe based identification and validation of roundabout junctions
CN108267747B (zh) 基于激光点云的道路特征提取方法和装置
CN107784084B (zh) 基于车辆定位数据的路网生成方法及系统
CN110379163B (zh) 一种基于轨迹数据的车辆异常减速区域检测方法及系统
KR20200121274A (ko) 전자 지도를 업데이트하기 위한 방법, 장치 및 컴퓨터 판독 가능한 저장 매체
CN111542860A (zh) 用于自主车辆的高清地图的标志和车道创建
CN108961758B (zh) 一种基于梯度提升决策树的路口展宽车道探测方法
US20130322694A1 (en) Energy Efficient Routing Using An Impedance Factor
CN108845569A (zh) 生成三维高清道路图水平弯道行车线的半自动点云方法
Uduwaragoda et al. Generating lane level road data from vehicle trajectories using kernel density estimation
CN104677363A (zh) 一种道路生成方法和装置
CN112154446B (zh) 立体车道线确定方法、装置和电子设备
CN111062958B (zh) 一种城市道路要素提取方法
CN104462155A (zh) 一种兴趣点父子关系的建立方法及装置
EP4120123A1 (en) Scan line-based road point cloud extraction method
CN110555432B (zh) 一种兴趣点处理方法、装置、设备和介质
CN110362640B (zh) 一种基于电子地图数据的任务分配方法及装置
Axer et al. Level of service estimation based on low-frequency floating car data
CN114842450A (zh) 一种可行驶区域检测方法、装置及设备
CN101776457A (zh) 一种导航方法和装置
CN110827537B (zh) 一种潮汐车道的设置方法、装置及设备
CN105574313A (zh) 一种道路等级的确定方法和装置
CN113822190B (zh) 一种厂区路网数据拟合方法、装置、电子设备和存储介质
WO2018151629A1 (ru) Способ и система автоматического построения трехмерных моделей городов

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2021791232

Country of ref document: EP

Effective date: 20211029

NENP Non-entry into the national phase

Ref country code: DE