WO2022088723A1 - 一种数据处理方法及装置 - Google Patents

一种数据处理方法及装置 Download PDF

Info

Publication number
WO2022088723A1
WO2022088723A1 PCT/CN2021/102869 CN2021102869W WO2022088723A1 WO 2022088723 A1 WO2022088723 A1 WO 2022088723A1 CN 2021102869 W CN2021102869 W CN 2021102869W WO 2022088723 A1 WO2022088723 A1 WO 2022088723A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
point cloud
collection
cloud data
radar device
Prior art date
Application number
PCT/CN2021/102869
Other languages
English (en)
French (fr)
Inventor
赵明
王兆圣
刘余钱
Original Assignee
上海商汤临港智能科技有限公司
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 上海商汤临港智能科技有限公司 filed Critical 上海商汤临港智能科技有限公司
Priority to KR1020227007210A priority Critical patent/KR20220058901A/ko
Priority to JP2022514853A priority patent/JP2023503767A/ja
Publication of WO2022088723A1 publication Critical patent/WO2022088723A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S17/00Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/88Lidar systems specially adapted for specific applications
    • G01S17/93Lidar systems specially adapted for specific applications for anti-collision purposes
    • G01S17/931Lidar systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S17/00Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/86Combinations of lidar systems with systems other than lidar, radar or sonar, e.g. with direction finders
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S17/00Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/88Lidar systems specially adapted for specific applications
    • G01S17/89Lidar systems specially adapted for specific applications for mapping or imaging
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/0002Inspection of images, e.g. flaw detection
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/70Determining position or orientation of objects or cameras
    • G06T7/73Determining position or orientation of objects or cameras using feature-based methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10028Range image; Depth image; 3D point clouds
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30248Vehicle exterior or interior
    • G06T2207/30252Vehicle exterior; Vicinity of vehicle

Definitions

  • the present disclosure relates to the technical field of information processing, and in particular, to a data processing method and apparatus.
  • autonomous driving technology mainly collects radar point cloud data through lidar, and then determines the position information of lidar based on positioning devices (such as Global Positioning System (GPS), integrated inertial navigation system, etc.), and then analyzes the point cloud data.
  • positioning devices such as Global Positioning System (GPS), integrated inertial navigation system, etc.
  • Information fusion with position information is performed to determine the positional relationship between the automatic driving device and obstacles, so as to achieve obstacle avoidance.
  • the position information of the laser radar is determined based on the positioning device
  • the position information of the positioning device is generally determined first, and then the position information of the positioning device is converted into the position information of the laser radar based on the external parameter data. It is used to represent the relative positional relationship between the positioning device and the lidar.
  • the embodiments of the present disclosure provide at least one data processing method and apparatus.
  • an embodiment of the present disclosure provides a data processing method, including: acquiring positioning data collected by a positioning device and point cloud data collected by a radar device, where the positioning device and the radar device are deployed on the same vehicle; The positioning data and the point cloud data determine the position data of the radar device when the point cloud data is collected; the ground parameter information representing the ground is determined based on the point cloud data, and the location data is determined based on the position data.
  • the initial pose data of the radar device is adjusted; based on the ground parameter information and the initial pose data, the external parameter data representing the relative positional relationship between the positioning device and the radar device is adjusted.
  • the ground parameter information representing the ground and the initial pose data of the radar device can be determined according to the positioning data collected by the positioning device and the radar data collected by the radar device, and then based on the ground parameter information and the initial pose data of the radar device , adjust the external parameter data. Since the ground parameter information and the initial pose data of the radar device can be obtained through data collection or calculation, in the process of adjusting the external parameter data, there is no need to manually use other equipment to adjust the position of the positioning device and the radar device. Calibration improves the adjustment efficiency of external parameter data and saves labor costs.
  • the point cloud data includes coordinate information of a plurality of radar scanning points; the determining, based on the point cloud data, ground parameter information representing the ground includes: performing a plane mapping based on the point cloud data. Fitting to obtain fitting plane information; based on the coordinate information of the radar scanning points in the point cloud data, determine the distance between each radar scanning point and the plane indicated by the fitting plane information; The distance between the radar scanning point and the plane indicated by the fitting plane information, screen the point cloud data, and return to the step of performing plane fitting based on the point cloud data based on the screened point cloud data, until reaching Iterative conditions are preset, and the ground parameter information is determined based on the finally screened point cloud data.
  • the point cloud data is screened multiple times by means of iterative fitting, and then the ground parameter information is determined based on the finally screened point cloud data, which improves the accuracy of the ground parameter information.
  • the following data screening processes before performing plane fitting based on the point cloud data, perform at least one of the following data screening processes: perform downsampling processing on the point cloud data, so that radar scans in the point cloud data are performed.
  • the distribution density of the points conforms to the preset condition; the radar scanning points whose corresponding coordinate information is located within the target coordinate range are screened out, wherein the target coordinate range is determined according to the preset installation height of the radar device.
  • data screening of point cloud data can improve the calculation accuracy of ground parameter information on the one hand, and reduce the amount of calculation in the plane fitting process and improve the calculation efficiency on the other hand.
  • the position data includes height data; the method further includes: based on the height data in the position data and the corresponding relationship between the position data and the point cloud data, the point cloud The data is divided into multiple collection intervals.
  • dividing the point cloud data into multiple collection intervals based on the height data in the position data and the correspondence between the position data and the point cloud data includes: The height data is filtered to determine at least one extreme point in the height data retained after filtering; the point cloud data is divided into multiple collection intervals by taking the collection time point corresponding to each extreme point as the dividing point of the collection interval.
  • the determining the ground parameter information representing the ground based on the point cloud data, and determining the initial pose data of the radar device based on the position data include: for each collection interval, based on point cloud data in the collection interval, determine the ground parameter information corresponding to the collection interval, and determine the radar device corresponding to the collection interval based on the position data of the radar device when collecting the point cloud data in the collection interval the initial pose data; the adjusting the external parameter data representing the relative positional relationship between the positioning device and the radar device based on the ground parameter information and the initial pose data, including: based on the The ground parameter information and the initial pose data corresponding to each collection interval are adjusted to the external parameter data representing the relative positional relationship between the positioning device and the radar device.
  • the consideration of ground height information is added, by dividing the point cloud data into a plurality of collection intervals, and determining the ground parameter information corresponding to the different collection intervals, and then based on the ground parameter information corresponding to the different collection intervals, and For the initial pose data in the collection interval, when adjusting the external parameter data, the adjusted external parameter is more accurate.
  • the external parameters representing the relative positional relationship between the positioning device and the radar device are compared.
  • the data adjustment includes: determining the optimized pose data corresponding to each collection interval based on the ground parameter information and the initial pose data corresponding to each collection interval; The pose data and the optimized pose data are used to adjust the external parameter data.
  • determining the optimized pose data corresponding to each collection interval includes: for any collection interval, Based on the ground parameter information and the initial pose data corresponding to the collection interval, determine the pose data that minimizes the value of the objective function, and use the pose data that minimizes the value of the objective function as the optimization corresponding to the collection interval
  • the objective function is the sum of the absolute value of the difference between the ground parameter information before and after the pose data optimization and the absolute value of the difference between the pose data before and after the pose data optimization.
  • the adjustment of the external parameter data based on the initial pose data corresponding to each collection interval and the optimized pose data includes: The initial pose data and the optimized pose data are used to determine the first average pose change in the collection interval; and the multiple collections are determined based on the first average pose change in each collection interval. The second average pose change amount corresponding to the interval; based on the second average pose change amount, the external parameter data is adjusted.
  • the optimized pose data corresponding to the initial pose data determined according to the ground parameter information in different collection intervals is more accurate, and then the first average pose variation and When the external parameter data is adjusted based on the second average pose change, the adjusted external parameter data is more accurate.
  • the adjusting the external parameter data based on the second average pose change includes: comparing the second average pose change with the extrinsic data before adjustment. The product is determined as the adjusted extrinsic data.
  • the positioning data includes position data collected at multiple first collection time points
  • the point cloud data includes point cloud data collected at multiple second collection time points
  • the determining, based on the positioning data and the point cloud data, the position data of the radar device when collecting the point cloud data includes: for each of the second collection time points, at the plurality of first time points.
  • the position data of the radar device at each second collection time point can be determined, thereby avoiding the interference caused by the difference in the data collection frequency of the radar device and the positioning device.
  • an embodiment of the present disclosure further provides a data processing device, comprising: an acquisition module configured to acquire positioning data collected by a positioning device and point cloud data collected by a radar device; wherein the positioning device and the radar device is deployed on the same vehicle; a first determination module is used to determine, based on the positioning data and the point cloud data, the position data of the radar device when the positioning device collects the positioning data; a second determination module, for determining ground parameter information representing the ground based on the point cloud data, and determining initial pose data of the radar device based on the position data; an adjustment module for determining based on the ground parameter information and the initial position Attitude data is used to adjust the external parameter data representing the relative positional relationship between the positioning device and the radar device.
  • embodiments of the present disclosure further provide a computer device, including: a processor, a memory, and a bus, where the memory stores machine-readable instructions executable by the processor, and when the computer device runs, the processing A bus communicates between the processor and the memory, and when the machine-readable instructions are executed by the processor, the first aspect or the steps in any possible implementation manner of the first aspect are performed.
  • an embodiment of the present disclosure further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is run by a processor, the steps in the first aspect or any of the possible implementation manners are executed.
  • FIG. 1 shows a schematic flowchart of a data processing method provided by an embodiment of the present disclosure
  • FIG. 2 shows a schematic diagram of a comparison between a first collection time point and a second collection time point provided by an embodiment of the present disclosure
  • FIG. 3 is a schematic diagram illustrating a comparison between another first acquisition time point and a second acquisition time point provided by an embodiment of the present disclosure
  • FIG. 4 shows a schematic flowchart of a method for determining ground parameter information representing the ground based on point cloud data provided by an embodiment of the present disclosure
  • FIG. 5 shows a schematic diagram of a method for determining an extreme point of altitude data provided by an embodiment of the present disclosure; a schematic diagram of a method for determining an extreme point of data;
  • FIG. 6 shows a schematic flowchart of a method for adjusting external parameter data provided by an embodiment of the present disclosure
  • FIG. 7 shows a schematic diagram of the architecture of a data processing apparatus provided by an embodiment of the present disclosure
  • FIG. 8 shows a schematic structural diagram of a computer device provided by an embodiment of the present disclosure.
  • the current mapping and positioning algorithms mostly use a multi-sensor fusion scheme, that is, the observations in the respective coordinate systems of multiple sensors are unified into the same coordinate system, while The bridge that unifies the observations in the respective coordinate systems of multiple sensors into the same coordinate system is the external parameter data between the sensors. Therefore, the quality of external parameter calibration has a great influence on the performance of multi-sensor fusion algorithm.
  • a mapping and positioning scheme that integrates positioning devices (such as integrated inertial navigation systems) and radar devices is common.
  • positioning devices such as integrated inertial navigation systems
  • radar devices For example, in the field of autonomous driving, a mapping and positioning scheme that integrates positioning devices (such as integrated inertial navigation systems) and radar devices is common.
  • a series of equipment will be used to calibrate the positions of the positioning device and the radar device on the autonomous vehicle, so as to determine the external parameter data representing the relative positional relationship between the positioning device and the radar device.
  • the relative position between the positioning device and the radar device will change during the use of the autonomous vehicle, the external parameter data needs to be updated regularly.
  • the external parameter data can also be estimated by a hand-eye calibration algorithm.
  • the self-driving vehicle can be controlled to run a certain trajectory, and the positioning data collected by the positioning device on the self-driving vehicle and the point cloud data collected by the radar device can be recorded during the operation of the self-driving vehicle, and then the positioning device and the radar device can be calculated respectively.
  • Pose trajectory and then solve the external parameter data based on the hand-eye calibration algorithm.
  • the measurement accuracy of the positioning device and the radar device is relatively high, and it can only provide variable parameters in three degrees of freedom.
  • the external parameter data are parameters in six degrees of freedom, that is, the translation distance along the X-axis, Y-axis and Z-axis of the Cartesian coordinate system, and the rotation angle along the X-axis, Y-axis and Z-axis respectively, we can It is represented by (X, Y, Z, Roll, Pitch, Yaw), so the error of the external parameter data determined by this method is relatively large.
  • the present disclosure provides a data processing method, which can determine the ground parameter information representing the ground and the initial pose data of the radar device according to the positioning data collected by the positioning device and the radar data collected by the radar device, Then, based on the ground parameter information and the initial pose data of the radar device, the external parameter data is adjusted. Since the ground parameter information and the initial pose data of the radar device can be obtained through data collection or calculation, in the process of adjusting the external parameter data, there is no need to manually calibrate the positions of the positioning device and the radar device with the help of other equipment, which improves the external parameters. The adjustment efficiency of data saves labor costs; in addition, in this scheme, the external parameter data is adjusted based on the pose change. Since the pose data is data with six degrees of freedom, when adjusting the external parameter data, you can All six quantities of the external parameter data are adjusted, thereby improving the adjustment accuracy of the external parameter data.
  • the devices include, for example, terminal devices or servers or other processing devices, and the terminal devices may be user equipment (User Equipment, UE), mobile devices, user terminals, terminals, cellular phones, cordless phones, and personal digital assistants (Personal Digital Assistant, PDA) Wait.
  • the data processing method may be processed by a processor installed on the automatic driving device.
  • the data processing method provided by the embodiment of the present disclosure will be described below by taking the execution body as the processor installed on the automatic driving device as an example. It should be noted that the positioning device and the radar device are respectively connected to the processor, and the connection methods include: It is not limited to wired connection and wireless connection, wherein, the wireless connection mode may be, for example, Bluetooth connection, wireless local area network connection, or the like.
  • an embodiment of the present disclosure provides a data processing method, which includes the following steps:
  • Step 101 Acquire positioning data collected by a positioning device and point cloud data collected by a radar device; wherein the positioning device and the radar device are deployed on the same vehicle.
  • Step 102 Based on the positioning data and the point cloud data, determine the position data of the radar device when collecting the point cloud data.
  • Step 103 Determine ground parameter information representing the ground based on the point cloud data, and determine initial pose data of the radar device based on the position data.
  • Step 104 Based on the ground parameter information and the initial pose data, adjust the external parameter data representing the relative positional relationship between the positioning device and the radar device.
  • the ground parameter information representing the ground and the initial pose data of the radar device can be determined according to the positioning data collected by the positioning device and the radar data collected by the radar device, and then based on the ground parameter information and the initial pose data of the radar device , to adjust the external parameter data. Since the ground parameter information and the initial pose data of the radar device can be obtained through data collection or calculation, in the process of adjusting the external parameter data, there is no need to manually calibrate the positions of the positioning device and the radar device with the help of other equipment, which improves the external parameters. Data adjustment efficiency saves labor costs.
  • Both the positioning device and the radar device may be installed on the same vehicle, and the installation positions of the positioning device and the radar device are different.
  • the positioning device may be, for example, a Global Positioning System (Global Positioning System, GPS), or a combined inertial navigation system, which is a combination of GPS and inertial sensors.
  • step 102 For step 102:
  • the data collection frequencies of the radar device and the positioning device are different.
  • the positioning device collects the position data, it can collect the position data according to the collection frequency set by the GPS satellite within the data collection time; when the radar device collects the point cloud data, it can emit a radio beam according to the set frequency, and the set frequency frequency to collect point cloud data.
  • the collection frequency set by the GPS satellite is higher than the set frequency corresponding to the radar device.
  • the positioning data collected by the positioning device includes a plurality of first collection time points and position data collected at each first collection time point, wherein the position data collected at each first collection time point includes latitude and longitude data and altitude data,
  • the height data is the height relative to the sea level;
  • the point cloud data includes a plurality of second collection time points and point cloud data collected at each second collection time point, wherein the points collected at each second collection time point
  • the cloud data includes coordinate information of multiple radar scan points.
  • the positioning data output by the positioning device is the positioning data adjusted based on the external parameter data representing the relative positional relationship between the positioning device and the radar device, that is, the positioning data output by the positioning device is the positioning data of the radar device.
  • the location data corresponding to the point cloud data is collected at the collection time point.
  • the position data corresponding to the point cloud data collected by the radar device at the second collection time point may also be referred to as “the position data corresponding to the radar device when collecting the cloud data at the second collection time point”, or further abbreviated as "Position data of the radar device at this second acquisition time point”.
  • each second collection time point it may be first detected whether there is a first collection time point corresponding to the second collection time point; if the second collection time point has a corresponding first collection time point, the The position data of the first collection time point corresponding to the second collection time point is determined as the position data of the radar device when collecting the point cloud data of the second collection time point; if there is no corresponding first collection time point at the second collection time point At a collection time point, the position data of the radar device at the second collection time point is calculated based on the time interval between the first collection time point and the second collection time point.
  • the two data with the shortest time interval from the second collection time point may be respectively determined first.
  • the first collection time point, the two first collection time points include a first collection time point before the second collection time point and a first collection time point after the second collection time point, and then based on the two The position data corresponding to the first collection time points are determined, and the position data of the radar device at the second collection time point are determined.
  • the average value of the position data of the two first collection time points with the shortest interval from the second time point may be determined as the position data of the radar device at the second collection time point.
  • the points a 1 , b 1 , c 1 , d 1 , e 1 , f 1 , and g 1 above the horizontal line in FIG. 2 all represent the first acquisition time point.
  • the collection time points have corresponding position data
  • a 2 , b 2 , c 2 , d 2 , and e 2 at the bottom of the horizontal direction all represent the second collection time points, and each second collection time point has corresponding point cloud data.
  • the first collection time points a 1 , d 1 , and g 1 all have corresponding second collection time points, and the specific corresponding relationship is a 1 ⁇ a 2 , d 1 ⁇ c 2 , g 1 ⁇ e 2 , then the The position data corresponding to the first collection time point a 1 is determined as the position data of the radar device at the second collection time point a 2 , and the position data corresponding to the first collection time point d 1 is determined as the radar device at the second collection time point c 2 , the position data corresponding to the first collection time point g1 is determined as the position data of the radar device at the second collection time point e2 .
  • a second collection time point without a corresponding first collection time point such as b 2
  • the mean value of the position data corresponding to the time points b 1 and c 1 respectively is determined as the position data of the radar device at the second collection time point b 2 .
  • the position data of the radar device at the second acquisition time point d 2 can be calculated.
  • the mean value of each item of data included in the position data may be calculated.
  • the location data includes longitude, latitude, and altitude
  • the mean value of longitude, mean value of latitude, and mean value of altitude can be calculated respectively, and then the mean value of longitude, mean value of latitude, and mean value of altitude can be used as the mean value of the mean value of the position data .
  • the radar device may perform the determination based on the distance between the two first collection time points and the second collection time point.
  • point a and point b are the first collection time point
  • point c is the second collection time point
  • when determining the position data of the radar device at the second collection time point c it can be determined according to The distance between the first collection time point a and the second collection time point c, and the distance between the first collection time point b and the second collection time point c, respectively determine the weights corresponding to the position data of the first collection time point a
  • the weight corresponding to the position data at the first collection time point b, and then the position data at the first collection time point a and the position data at the first collection time point b are weighted and summed according to their corresponding weights to obtain the radar device at the first 2. Collect position data at time point c.
  • the position data of the radar device at each second collection time point can be determined, thereby avoiding the interference caused by the difference in the data collection frequency of the radar device and the positioning device.
  • the ground parameter information that characterizes the ground includes normal vector and intercept. Specifically, when determining the ground parameter information representing the ground based on the point cloud data, the method shown in FIG. 4 can be referred to, including the following steps:
  • Step 401 Perform plane fitting based on the point cloud data to obtain fitting plane information.
  • the plane fitting when the plane fitting is performed based on the point cloud data, the plane fitting can be performed based on the random sample consensus algorithm (Random sample consensus, RANSAC). Specifically, a preset ratio of point cloud data may be randomly selected from the point cloud data, and plane fitting may be performed based on the randomly selected point cloud data.
  • RANSAC random sample consensus
  • Step 402 based on the coordinate information of the radar scanning points in the point cloud data, determine the distance between each radar scanning point and the plane indicated by the fitting plane information.
  • Step 403 Perform point cloud data screening based on the determined distance between each radar scanning point and the plane indicated by the fitting plane information.
  • the radar scanning points that are not on the ground are filtered out. Specifically, when screening the point cloud data based on the determined distance between each radar scanning point and the plane indicated by the fitting information, the distance from the plane indicated by the fitting information may exceed a preset value. The radar scan points of the distance range are filtered out.
  • Step 404 judging whether a preset iteration condition is satisfied.
  • step 405 is performed;
  • step 401 If the preset iterative conditions are not met, based on the filtered point cloud data, return to step 401 to continue screening the point cloud data.
  • the preset iteration condition may be that the preset number of iterations is reached, or when the point cloud data is screened in step 403, the distances between all the radar scanning points and the plane indicated by the fitting plane information are all within the preset distance. within the range.
  • Step 405 Determine ground parameter information based on the finally screened point cloud data.
  • the point cloud data is screened multiple times by means of iterative fitting, and then the ground parameter information is determined based on the finally screened point cloud data, which improves the accuracy of the ground parameter information.
  • ground parameter information is determined based on the last screened point cloud data
  • plane fitting may be performed again based on the last screened point cloud data, and then ground parameter information of the fitted plane is determined, wherein the ground parameter information includes normal vector and intercept distance.
  • At least one of the following data screening processes may also be performed:
  • the point cloud data may be down-sampled by a voxel filtering method.
  • the preset installation height of the radar device is the height of the radar device from the ground.
  • the position of the radar device can be used as the starting point of the Z axis, and then the target coordinate range can be determined according to the preset installation height of the radar device.
  • the target coordinate range can be used to indicate the height range of the radar scanning point located on the ground in the Z-axis.
  • the pose data of the radar device includes not only the coordinates of the radar device in the established world coordinate system, but also at least one of a pitch angle, a yaw angle, and a roll angle.
  • the automatic driving device may not drive on the same ground plane, therefore, the ground parameter information between different ground planes may be different. Therefore, in another embodiment of the present disclosure, the The height data in the position data, and the correspondence between the position data and the point cloud data, divide the point cloud data into multiple collection intervals, and each collection interval includes multiple second collection time points. It is considered that the vehicle is driving on the same ground plane.
  • the height data may be filtered based on the height values of the height data at each second collection time point , and then determine at least one extreme point in the height data retained after filtering processing, and use the acquisition time point corresponding to each extreme point as the dividing point of the acquisition interval, and divide the point cloud data into multiple acquisition intervals.
  • the height data collected by the automatic driving device during driving may be affected by obstacles on the ground, such as bumps caused by obstacles, etc.
  • the height data collected by the automatic driving device can be filtered first. Specifically, the height value whose fluctuation frequency is higher than the set frequency or whose fluctuation range is smaller than the set range can be filtered out.
  • the abscissa is the second collection time point
  • the ordinate is the height data in the position data corresponding to each second collection time point
  • the point in the line graph is the extreme point
  • every two The second collection time point between the extreme points is a collection interval
  • each collection interval includes multiple second collection time points, and similarly includes the corresponding point cloud data, for example, including multiple frames of point cloud images .
  • the ground parameter information corresponding to the collection interval can be determined based on the point cloud data in the collection interval, and based on the point cloud collected in the collection interval
  • the data is the position data of the radar device, and the initial pose data of the radar device corresponding to the collection interval is determined.
  • the external parameter data representing the relative positional relationship between the positioning device and the radar device is adjusted.
  • the external parameter data shown in FIG. 6 may be referred to.
  • the adjustment method includes the following steps:
  • Step 601 Determine optimized pose data corresponding to each collection interval based on the ground parameter information and the initial pose data corresponding to each collection interval.
  • the optimized pose data corresponding to each collection interval when determining the optimized pose data corresponding to each collection interval based on the ground parameter information and the initial pose data corresponding to each collection interval, for any collection interval, it may be based on the collection interval.
  • Corresponding ground parameter information and initial pose data determine the optimized pose data that minimizes the value of the objective function, and use the pose data that minimizes the value of the objective function as the optimized pose data corresponding to the collection interval data, wherein the objective function is the sum of the absolute value of the difference between the ground parameter information before and after the pose data optimization and the absolute value of the difference between the pose data before and after the pose data optimization.
  • the objective function can be as shown in the following formula:
  • N represents the number of point cloud images included in the point cloud data in the collection interval, that is, the number of second collection time points in the collection interval.
  • a group of point cloud data will be collected, and this group of point cloud data can form a frame of point cloud image.
  • the second collection time point corresponds to multiple sets of point cloud data
  • each set of point cloud data corresponds to a frame of point cloud image
  • the initial pose data corresponding to the set of point cloud data is calculated, and the Entering the above formula, and solving the formula by the least square method, the optimized pose data corresponding to the initial pose data of each frame of point cloud image can be obtained.
  • the optimized ground parameter information can also be obtained, but the optimized ground parameter information has nothing to do with adjusting the external parameter data.
  • the optimized pose data corresponding to the first frame of point cloud images can be calculated respectively
  • the optimized pose data corresponding to the second frame of point cloud images The optimized pose data corresponding to the third frame of point cloud images and optimized ground parameter information
  • each element in the vector represents the coordinates x, y, z, pitch angle, yaw angle, and roll angle of the radar device in the world coordinate system, respectively.
  • each element in the vector represents the value and intercept of the normal vector, and the normal vector is a three-dimensional vector.
  • Step 602 Adjust the external parameter data based on the initial pose data corresponding to each collection interval and the optimized pose data.
  • the initial pose data corresponding to each collection interval and the optimized pose data can be adjusted first. pose data, and determine the first average pose change in the collection interval; then determine the second average pose change corresponding to multiple collection intervals based on the first average pose change in each collection interval; and then based on the first average pose change in each collection interval
  • the second average pose change is to adjust the external parameter data.
  • the frame of point cloud image when determining the first average pose change in any collection interval, for each frame of point cloud image corresponding to the point cloud data in the collection interval, the frame of point cloud image may be determined
  • the first average pose change is calculated based on the pose change between the corresponding initial pose data and the optimized pose data, and then based on the pose change of the multi-frame point cloud images corresponding to the point cloud data.
  • the latitude, longitude, altitude, latitude and longitude in the initial pose data can be calculated respectively.
  • the amount of change corresponding to the latitude, longitude and altitude in the optimized pose data can be calculated respectively.
  • the product between the second average pose change and the extrinsic data before adjustment may be determined as the adjusted external parameter.
  • external parameter data when the external parameter data is adjusted based on the second average pose change, the product between the second average pose change and the extrinsic data before adjustment may be determined as the adjusted external parameter.
  • the ground parameter information and the initial pose data of the radar device can be determined according to the positioning data collected by the positioning device and the radar data collected by the radar device, and then based on the ground parameter information and the initial pose data of the radar device, the external
  • the external In the process of adjusting the external parameter data, there is no need to use other equipment to calibrate the relative position of the positioning device and the radar device, which improves the accuracy and adjustment efficiency of the external parameter data, and then determines the pose data of the radar device. Higher efficiency and precision.
  • the writing order of each step does not mean a strict execution order but constitutes any limitation on the implementation process, and the specific execution order of each step should be based on its function and possible Internal logic is determined.
  • the embodiment of the present disclosure also provides a data processing apparatus corresponding to the data processing method. Reference may be made to the implementation of the method, and repeated descriptions will not be repeated.
  • an embodiment of the present disclosure provides a data processing apparatus, the apparatus includes: an acquisition module 701, a first determination module 702, a second determination module 703, and an adjustment module 704; wherein,
  • An acquisition module 701 configured to acquire positioning data collected by a positioning device and point cloud data collected by a radar device; wherein the positioning device and the radar device are deployed on the same vehicle;
  • a first determining module 702 configured to determine, based on the positioning data and the point cloud data, the position data of the radar device when the point cloud data is collected;
  • a second determining module 703, configured to determine ground parameter information representing the ground based on the point cloud data, and determine initial pose data of the radar device based on the position data;
  • the adjustment module 704 is configured to adjust the external parameter data representing the relative positional relationship between the positioning device and the radar device based on the ground parameter information and the initial pose data.
  • the point cloud data includes coordinate information of multiple radar scanning points.
  • the second determining module 703 is configured to: perform plane fitting based on the point cloud data to obtain fitting plane information; Coordinate information of radar scanning points in the point cloud data, determine the distance between each radar scanning point and the plane indicated by the fitting plane information; based on each radar scanning point determined and indicated by the fitting plane information According to the distance between the planes, screen the point cloud data, and based on the screened point cloud data, return to the step of plane fitting based on the point cloud data, until the preset iteration conditions are reached, based on the final screened point cloud data.
  • the ground parameter information is determined.
  • the second determining module 703 is further configured to perform at least one of the following data screening processes before performing plane fitting based on the point cloud data: performing a downgrade on the point cloud data Sampling processing, so that the distribution density of radar scanning points in the point cloud data conforms to preset conditions; screening out the radar scanning points whose corresponding coordinate information is located within the target coordinate range, wherein the target coordinate range is based on the radar scanning point.
  • the preset installation height of the device is determined.
  • the location data includes altitude data.
  • the second determining module 703 is further configured to: divide the point cloud data into multiple collection intervals based on the height data in the position data and the correspondence between the position data and the point cloud data.
  • the second determining module 703 divides the point cloud data into a plurality of pieces based on the height data in the position data and the correspondence between the position data and the point cloud data During the collection interval, it is used to: filter the height data, and determine at least one extreme point in the height data retained after filtering; take the collection time point corresponding to each extreme point as the dividing point of the collection interval, and divide the The point cloud data is divided into multiple collection intervals.
  • the second determining module 703 when determining the ground parameter information representing the ground based on the point cloud data, and determining the initial pose data of the radar device based on the position data, Used for: for each collection interval, based on the point cloud data in the collection interval, determine the ground parameter information corresponding to the collection interval, and based on the position data of the radar device when collecting the point cloud data in the collection interval, determine the The initial pose data of the radar device corresponding to the collection interval is collected.
  • the adjustment module 704 adjusts the external parameter data representing the relative positional relationship between the positioning device and the radar device based on the ground parameter information and the initial pose data, It is used for: adjusting the external parameter data representing the relative positional relationship between the positioning device and the radar device based on the ground parameter information and the initial pose data corresponding to each collection interval.
  • the adjustment module 704 based on the ground parameter information and the initial pose data corresponding to each collection interval, adjusts the relationship between the positioning device and the radar device.
  • the external parameter data of the relative position relationship is adjusted, it is used to: determine the optimized pose data corresponding to each collection interval based on the ground parameter information and the initial pose data corresponding to each collection interval; The initial pose data and the optimized pose data corresponding to each collection interval are adjusted to the external parameter data.
  • the adjustment module 704 when determining the optimized pose data corresponding to each collection interval based on the ground parameter information and the initial pose data corresponding to each collection interval , for: for any collection interval, based on the ground parameter information corresponding to the collection interval and the initial pose data, determine the pose data that minimizes the value of the objective function, and will make the value of the objective function the smallest.
  • the pose data is used as the optimized pose data corresponding to the collection interval, wherein the objective function is the absolute value of the difference between the ground parameter information before and after the pose data optimization is performed, and the pose data before and after the pose data optimization is performed. The sum of the absolute values of the differences.
  • the adjustment module 704 when adjusting the external parameter data based on the initial pose data corresponding to each collection interval and the optimized pose data, is used to adjust the external parameter data. : Based on the initial pose data corresponding to each collection interval and the optimized pose data, determine the first average pose change in the collection interval; based on the first average pose change in each collection interval The second average pose change amount corresponding to the multiple collection intervals is determined; based on the second average pose change amount, the external parameter data is adjusted.
  • the adjustment module 704 when adjusting the external parameter data based on the second average pose change amount, is configured to: adjust the second average pose change amount The product of the external parameter data before adjustment is determined as the external parameter data after adjustment.
  • the positioning data includes position data collected at multiple first collection time points
  • the point cloud data includes point cloud data collected at multiple second collection time points ; determining the position data of the radar device when collecting the point cloud data based on the positioning data and the point cloud data.
  • the first determining module 702 is configured to: for each One of the second collection time points, in the case that there is a target first collection time point that overlaps with the second collection time point in the plurality of first collection time points, based on the corresponding first collection time point of the target determine the position data of the radar device at the second collection time point; in the case that the target first collection time point does not exist in the plurality of first collection time points, based on the plurality of first collection time points Position data corresponding to two adjacent first collection time points with the shortest time interval from the second collection time point in a collection time point, determine the position data of the radar device at the second collection time point, and the second collection time point.
  • the acquisition time point is located between the two adjacent first acquisition time points.
  • the functions or templates included in the apparatus provided by the embodiments of the present disclosure may be used to execute the methods described in the above method embodiments.
  • the functions or templates included in the apparatus provided by the embodiments of the present disclosure may be used to execute the methods described in the above method embodiments.
  • the computer device includes a processor 801 , a memory 802 , and a bus 803 .
  • the memory 802 is used to store the execution instructions, including the memory 8021 and the external memory 8022; the memory 8021 here is also called the internal memory, and is used to temporarily store the operation data in the processor 801 and the data exchanged with the external memory 8022 such as the hard disk,
  • the processor 801 exchanges data with the external memory 8022 through the memory 8021 .
  • the processor 801 communicates with the memory 802 through the bus 803, so that the processor 801 executes the following instructions: acquiring the positioning data collected by the positioning device and the point cloud data collected by the radar device; wherein, the positioning The device and the radar device are deployed on the same vehicle; based on the positioning data and the point cloud data, the position data of the radar device when collecting the point cloud data is determined; the ground representing the ground is determined based on the point cloud data parameter information, and determining the initial pose data of the radar device based on the position data; based on the ground parameter information and the initial pose data, to characterize the relative relationship between the positioning device and the radar device The external parameter data of the positional relationship is adjusted.
  • the specific processing process performed by the processor 801 may refer to the descriptions in the foregoing method embodiments, which will not be described herein again.
  • Embodiments of the present disclosure further provide a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is run by a processor, the steps of the data processing method described in the foregoing method embodiments are executed.
  • the storage medium may be a volatile or non-volatile computer-readable storage medium.
  • the computer program product of the data processing method provided by the embodiments of the present disclosure includes a computer-readable storage medium storing program codes, and the instructions included in the program codes can be used to execute the steps of the data processing methods described in the above method embodiments. .
  • the computer program product can be specifically implemented by hardware, software or a combination thereof.
  • 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 (Software Development Kit, SDK), etc. Wait.
  • the units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.
  • each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
  • the functions, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a processor-executable non-volatile computer-readable storage medium.
  • the computer software products are stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present disclosure.
  • the aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program codes .

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Electromagnetism (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Traffic Control Systems (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

本公开提供了一种数据处理方法及装置,包括:获取定位装置采集的定位数据和雷达装置采集的点云数据;其中,所述定位装置和所述雷达装置部署在同一车辆上;基于所述定位数据和所述点云数据,确定所述雷达装置在采集所述点云数据时的位置数据;基于所述点云数据确定表征地面的地面参数信息,以及基于所述位置数据确定所述雷达装置的初始位姿数据;基于所述地面参数信息、以及所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整。

Description

一种数据处理方法及装置
相关申请的交叉引用
本专利申请要求于2020年10月30日提交的、申请号为202011190058.9、发明名称为“一种数据处理方法及装置”的中国专利申请的优先权,该申请以引用的方式并入本文中。
技术领域
本公开涉及信息处理技术领域,具体而言,涉及一种数据处理方法及装置。
背景技术
随着信息技术的发展,自动驾驶技术也逐渐得到了广泛的应用。目前,自动驾驶技术主要通过激光雷达采集雷达点云数据,然后基于定位装置(如全球定位系统(Global Positioning System,GPS)、组合惯性导航系统等)确定激光雷达的位置信息,再对点云数据与位置信息进行信息融合,确定自动驾驶装置与障碍物之间的位置关系,从而实现避障。其中,在基于定位装置确定激光雷达的位置信息时,一般是先通过确定定位装置的位置信息,然后基于外参数据将定位装置的位置信息转换为激光雷达的位置信息,其中,外参数据用于表示定位装置和激光雷达之间的相对位置关系。
相关技术中,在确定外参数据时,需要借助一系列设备对定位装置和激光雷达的位置进行标定。但是自动驾驶装置在使用过程中,定位装置和激光雷达的位置可能会发生变化,因此外参数据需要每隔预设时间更新一次。每次更新都需要人工借助测量设备重新标定定位装置和激光雷达的位置,再重新确定外参数据,这种方法效率较低。
发明内容
本公开实施例至少提供一种数据处理方法及装置。
第一方面,本公开实施例提供了一种数据处理方法,包括:获取定位装置采集的定位数据和雷达装置采集的点云数据,所述定位装置和所述雷达装置部署在同一车辆上;基于所述定位数据和所述点云数据,确定所述雷达装置在采集所述点云数据时的位置数据;基于所述点云数据确定表征地面的地面参数信息,以及基于所述位置数据确定所述雷达装置的初始位姿数据;基于所述地面参数信息、及所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整。
通过上述实施方式,可以根据定位装置采集的定位数据以及雷达装置采集的雷达数据,确定表征地面的地面参数信息以及雷达装置的初始位姿数据,然后基于地面参数信息以及雷达装置的初始位姿数据,对外参数据进行调整,由于地面参数信息和雷达装置的初始位姿数据都可以通过数据采集或计算得到,在外参数据的调整过程中,无需由人力借助其他设备对定位装置和雷达装置的位置进行标定,提高了外参数据的调整效率,节省了人力成本。
一种可能的实施方式中,所述点云数据中包括多个雷达扫描点的坐标信息;所述基于所述点云数据确定表征地面的地面参数信息,包括:基于所述点云数据进行平面拟合,得到拟合平面信息;基于所述点云数据中雷达扫描点的坐标信息,确定每一个雷达扫描点与所述拟合平面信息所指示的平面之间的距离;基于确定的每一个雷达扫描点与所述拟合平面信息所指示的平面之间的距离,进行点云数据筛选,并基于筛选后的点云数据,返回基于所述点云数据进行平面拟合的步骤,直到达到预设迭代条件,基于最终筛选的点云数据确定所述地面参数信息。
上述实施方式中,通过迭代拟合的方式,对点云数据进行多次筛选,然后再基于最后筛选后的点云数据确定地面参数信息,提高了地面参数信息的精度。
一种可能的实施方式中,在基于所述点云数据进行平面拟合之前,执行以下至少一种数据筛选过程:对所述点云数据进行降采样处理,使得所述点云数据中雷达扫描点的分布密度符合预设条件;筛选出对应的坐标信息位于所述目标坐标范围内的雷达扫描点,其中所述目标坐标范围是根据所述雷达装置的预设安装高度确定的。
在基于点云数据进行平面拟合之前,先对点云数据进行数据筛选,一方面可以提高地面参数信息的计算精度,另一方面可以降低平面拟合过程中的计算量,提高计算效率。
一种可能的实施方式中,所述位置数据包括高度数据;所述方法还包括:基于所述位置数据中的高度数据,以及位置数据和点云数据之间的对应关系,将所述点云数据划分至多个采集区间。
一种可能的实施方式中,所述基于所述位置数据中的高度数据,以及位置数据和点云数据之间的对应关系,将所述点云数据划分至多个采集区间,包括:对所述高度数据进行过滤,确定过滤后保留的高度数据中的至少一个极值点;以各个极值点对应的采集时间点作为采集区间的切分点,将所述点云数据划分至多个采集区间。
一种可能的实施方式中,所述基于所述点云数据确定表征地面的地面参数信息,以 及基于所述位置数据确定所述雷达装置的初始位姿数据,包括:针对每个采集区间,基于该采集区间内的点云数据,确定该采集区间对应的地面参数信息,以及基于所述雷达装置在采集该采集区间内的点云数据时的位置数据,确定该采集区间对应的所述雷达装置的初始位姿数据;所述基于所述地面参数信息、以及所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整,包括:基于每个采集区间对应的所述地面参数信息和所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整。
上述实施方式中,增加了对于地面高度信息的考虑,通过将点云数据划分至多个采集区间,并确定不同采集区间分别对应的地面参数信息,再基于不同采集区间分别对应的地面参数信息,以及该采集区间内的初始位姿数据,对外参数据进行调整时,调整的外参更加精确。
一种可能的实施方式中,所述基于每个采集区间对应的所述地面参数信息和所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整,包括:基于每个采集区间对应的所述地面参数信息和所述初始位姿数据,确定每个采集区间对应的优化后的位姿数据;基于每个采集区间对应的所述初始位姿数据、以及优化后的位姿数据,对所述外参数据进行调整。
一种可能的实施方式中,基于每个采集区间对应的所述地面参数信息和所述初始位姿数据,确定每个采集区间对应的优化后的位姿数据,包括:针对任一个采集区间,基于该采集区间对应的所述地面参数信息、以及所述初始位姿数据,确定使得目标函数的值最小的位姿数据,将使得目标函数的值最小的位姿数据作为该采集区间对应的优化后的位姿数据,所述目标函数为进行位姿数据优化前后的地面参数信息之差的绝对值,与进行位姿数据优化前后的位姿数据之差的绝对值之和。
一种可能的实施方式中,所述基于每个采集区间对应的所述初始位姿数据、以及优化后的位姿数据,对所述外参数据进行调整,包括:基于每个采集区间对应的初始位姿数据,及优化后的位姿数据,确定该采集区间内的第一平均位姿变化量;基于每个采集区间内的所述第一平均位姿变化量,确定所述多个采集区间对应的第二平均位姿变化量;基于所述第二平均位姿变化量,对所述外参数据进行调整。
由于不同采集区间内的地面参数信息的不同,因此根据不同采集区间的地面参数信息所确定的初始位姿数据对应的优化后的位姿数据更加精确,进而在计算第一平均位姿变化量和第二平均位姿变化量,并基于第二平均位姿变化量对外参数据进行调整时,调 整后的外参数据更加精确。
一种可能的实施方式中,所述基于所述第二平均位姿变化量,对所述外参数据进行调整,包括:将所述第二平均位姿变化量与调整前的外参数据的乘积确定为调整后的外参数据。
一种可能的实施方式中,所述定位数据中包含在多个第一采集时间点分别采集的位置数据,所述点云数据中包含在多个第二采集时间点分别采集的点云数据;所述基于所述定位数据和所述点云数据,确定所述雷达装置在采集所述点云数据时的位置数据,包括:针对每一个所述第二采集时间点,在所述多个第一采集时间点中存在与该第二采集时间点重叠的目标第一采集时间点的情况下,基于所述目标第一采集时间点对应的位置数据确定所述雷达装置在该第二采集时间点的位置数据;在所述多个第一采集时间点中不存在所述目标第一采集时间点的情况下,基于所述多个第一采集时间点中与该第二采集时间点相隔时间最短的两个相邻第一采集时间点分别对应的位置数据,确定所述雷达装置在该第二采集时间点的位置数据,该第二采集时间点位于所述两个相邻第一采集时间点之间。
通过上述实施方式,可以确定雷达装置在每个第二采集时间点的位置数据,避免了雷达装置和定位装置的数据采集频率的不同所带来的干扰。
第二方面,本公开实施例还提供一种数据处理装置,包括:获取模块,用于获取定位装置采集的定位数据和雷达装置采集的点云数据;其中,所述定位装置和所述雷达装置部署在同一车辆上;第一确定模块,用于基于所述定位数据和所述点云数据,确定所述雷达装置在所述定位装置采集所述定位数据时的位置数据;第二确定模块,用于基于所述点云数据确定表征地面的地面参数信息,以及基于所述位置数据确定所述雷达装置的初始位姿数据;调整模块,用于基于所述地面参数信息、以及所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整。
第三方面,本公开实施例还提供一种计算机设备,包括:处理器、存储器和总线,所述存储器存储有所述处理器可执行的机器可读指令,当计算机设备运行时,所述处理器与所述存储器之间通过总线通信,所述机器可读指令被所述处理器执行时执行上述第一方面,或第一方面中任一种可能的实施方式中的步骤。
第四方面,本公开实施例还提供一种计算机可读存储介质,其上存储有计算机程序,该计算机程序被处理器运行时执行上述第一方面或其中任一可能的实施方式中的步骤。
关于上述数据处理装置、计算机设备、及计算机可读存储介质的效果描述参见上述数据处理方法的说明,这里不再赘述。为使本公开的上述目的、特征和优点能更明显易懂,下文特举较佳实施例,并配合所附附图,作详细说明如下。
附图说明
为了更清楚地说明本公开实施例的技术方案,下面将对实施例中所需要使用的附图作简单地介绍。这些附图示出了符合本公开的实施例,并与说明书一起用于说明本公开的技术方案。应当理解,以下附图仅示出了本公开的某些实施例,因此不应被看作是对范围的限定,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相关的附图。
图1示出了本公开实施例所提供的数据处理方法的流程示意图;
图2示出了本公开实施例所提供的一种第一采集时间点与第二采集时间点的对照示意图;
图3示出了本公开实施例所提供的另一种第一采集时间点与第二采集时间点的对照示意图;
图4示出了本公开实施例所提供的一种基于点云数据确定表征地面的地面参数信息的方法流程示意图;
图5示出了本公开实施例所提供的一种高度数据极值点确定方法示意图数据极值点确定方法示意图;
图6示出了本公开实施例所提供的一种外参数据调整方法的流程示意图;
图7示出了本公开实施例所提供的一种数据处理装置的架构示意图;
图8示出了本公开实施例所提供的一种计算机设备的结构示意图。
具体实施方式
为使本公开实施例的目的、技术方案和优点更加清楚,下面将结合本公开实施例中附图,对本公开实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本公开一部分实施例,而不是全部的实施例。通常在此处附图中描述和示出的本公开实施例的组件可以以各种不同的配置来布置和设计。因此,以下对在附图中提供的本公开的实施例的详细描述并非旨在限制要求保护的本公开的范围,而是仅仅表示本公开的选定实施例。基于本公开的实施例,本领域技术人员在没有做出创造性劳动的前提下 所获得的所有其他实施例,都属于本公开保护的范围。
由于单个传感器或多或少存在一些自身的局限性,因此目前的建图定位算法多采用多传感器融合的方案,即,将多个传感器各自坐标系下的观测统一到同一个坐标系下,而将多个传感器各自坐标系下的观测统一到同一个坐标系下的桥梁正是传感器之间的外参数据。因此,外参标定的好坏对多传感器融合算法的性能有着非常大的影响。
例如,在自动驾驶领域,比较常见的是定位装置(例如组合惯性导航系统)与雷达装置融合的建图与定位方案。一般在自动驾驶车辆出厂前,会借助一系列设备对自动驾驶车辆上的定位装置和雷达装置的位置进行标定,从而确定表征定位装置与雷达装置之间的相对位置关系的外参数据。然而由于自动驾驶车辆在使用过程中,定位装置和雷达装置之间的相对位置会发生变化,因此需要对外参数据进行定期更新。
相关技术中,即使在对外参数据进行更新时,也采用自动驾驶车辆出厂时的外参数据的确定方法,步骤较繁琐,效率较低。
在另外一种方法中,还可以通过手眼标定算法对外参数据进行估计。具体的,可以控制自动驾驶车辆运行一段轨迹,并记录下自动驾驶车辆在运行过程中其上的定位装置采集的定位数据以及雷达装置采集的点云数据,然后分别计算出定位装置和雷达装置的位姿轨迹,再基于手眼标定算法进行外参数据的求解。
然而这种方法中,对定位装置和雷达装置的测量精度要求比较高,且仅能提供三个自由度上的变化参量。由于外参数据为六个自由度上的参量,即,沿直角坐标系的X轴、Y轴和Z轴方向平移的距离,以及分别沿X轴、Y轴和Z轴方向旋转的角度,可以通过(X,Y,Z,Roll,Pitch,Yaw)进行表示,因此,通过这种方法确定出的外参数据误差较大。
针对以上方案所存在的缺陷,本公开提供了一种数据处理方法,可以根据定位装置采集的定位数据以及雷达装置采集的雷达数据,确定表征地面的地面参数信息以及雷达装置的初始位姿数据,然后基于地面参数信息以及雷达装置的初始位姿数据,对外参数据进行调整。由于地面参数信息和雷达装置的初始位姿数据都可以通过数据采集或计算得到,在外参数据的调整过程中,无需由人力借助其他设备对定位装置和雷达装置的位置进行标定,提高了外参数据的调整效率,节省了人力成本;另外本方案中是基于位姿变化量对外参数据进行调整,由于位姿数据为六个自由度上的数据,因此,在对外参数据进行调整时,可以对外参数据的六个量都进行调整,从而提高了外参数据的调整精度。
以上方案所存在的缺陷,均是发明人在经过实践并仔细研究后得出的,因此,上述问题的发现过程以及下文中本公开针对上述问题所提出的解决方案,都应该是发明人在本公开过程中对本公开做出的贡献。
下面将结合本公开中附图,对本公开中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本公开一部分实施例,而不是全部的实施例。通常在此处附图中描述和示出的本公开的组件可以以各种不同的配置来布置和设计。因此,以下对在附图中提供的本公开的实施例的详细描述并非旨在限制要求保护的本公开的范围,而是仅仅表示本公开的选定实施例。基于本公开的实施例,本领域技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例,都属于本公开保护的范围。
应注意到:相似的标号和字母在下面的附图中表示类似项,因此,一旦某一项在一个附图中被定义,则在随后的附图中不需要对其进行进一步定义和解释。
为便于对本实施例进行理解,首先对本公开实施例所公开的一种数据处理方法进行详细介绍,本公开实施例所提供的数据处理方法的执行主体一般为具有一定计算能力的计算机设备,该计算机设备例如包括:终端设备或服务器或其它处理设备,终端设备可以为用户设备(User Equipment,UE)、移动设备、用户终端、终端、蜂窝电话、无绳电话、个人数字处理(Personal Digital Assistant,PDA)等。在一些可能的实现方式中,该数据处理方法可以通过安装在自动驾驶装置上的处理器进行处理。
下面以执行主体为安装在自动驾驶装置上的处理器为例对本公开实施例提供的数据处理方法加以说明,其中需要说明的是,定位装置和雷达装置分别与处理器连接,其连接方式包括但不仅限于有线连接、无线连接,其中,无线连接方式例如可以为蓝牙连接、无线局域网连接等。
参见图1所示,本公开实施例提供一种数据处理方法,包括以下几个步骤:
步骤101、获取定位装置采集的定位数据和雷达装置采集的点云数据;其中,所述定位装置和所述雷达装置部署在同一车辆上。
步骤102、基于所述定位数据和所述点云数据,确定所述雷达装置在采集所述点云数据时的位置数据。
步骤103、基于所述点云数据,确定表征地面的地面参数信息,以及基于所述位置数据确定所述雷达装置的初始位姿数据。
步骤104、基于所述地面参数信息、以及所述初始位姿数据,对表征所述定位装置 与所述雷达装置之间的相对位置关系的外参数据进行调整。
通过上述实施方法,可以根据定位装置采集的定位数据以及雷达装置采集的雷达数据,确定表征地面的地面参数信息以及雷达装置的初始位姿数据,然后基于地面参数信息以及雷达装置的初始位姿数据,对外参数据进行调整。由于地面参数信息和雷达装置的初始位姿数据都可以通过数据采集或计算得到,在外参数据的调整过程中,无需由人力借助其他设备对定位装置和雷达装置的位置进行标定,提高了外参数据的调整效率,节省了人力成本。
以下是针对步骤101至步骤104的详细说明。
针对步骤101:
定位装置和雷达装置都可以是安装在同一车辆上,且定位装置和雷达装置的安装位置不同。定位装置例如可以为全球定位系统(Global Positioning System,GPS),或者组合惯性导航系统,组合惯性导航系统为GPS与惯性传感器组合的系统。
针对步骤102:
实际应用中,在同一时间段内,雷达装置和定位装置的数据采集频率是不同的。定位装置在采集位置数据时,可以在数据采集时间内,按照GPS卫星设置的采集频率对位置数据进行采集;雷达装置在采集点云数据时,可以按照设定频率发射无线电光束,以该设定频率进行点云数据的采集。实际应用中,GPS卫星设置的采集频率比雷达装置对应的设定频率高。
定位装置采集的定位数据中包括多个第一采集时间点和在每个第一采集时间点采集的位置数据,其中,每个第一采集时间点采集的位置数据包括经纬度数据、以及高度数据,所述高度数据为相对于海平面的高度;点云数据中包含多个第二采集时间点和在每个第二采集时间点采集的点云数据,其中,每个第二采集时间点采集的点云数据包括多个雷达扫描点的坐标信息。
需要说明的是,定位装置输出的定位数据为基于表征定位装置和雷达装置之间的相对位置关系的外参数据进行调整后的定位数据,即定位装置输出的定位数据为雷达装置的定位数据。
由于雷达装置采集雷达数据的频率与定位装置采集定位数据的频率不同,因此需要基于定位数据和雷达数据,确定雷达装置在定位装置采集定位数据时的位置数据,即确定雷达装置在每个第二采集时间点采集点云数据时对应的位置数据。其中,雷达装置在 第二采集时间点采集点云数据时对应的位置数据,以下也可称为“雷达装置在采集该第二采集时间点的云数据时对应的位置数据”,或进一步简称为“雷达装置在该第二采集时间点的位置数据”。
具体的,针对每一个第二采集时间点,可以先检测是否存在与该第二采集时间点对应的第一采集时间点;若该第二采集时间点有对应的第一采集时间点,则将该第二采集时间点对应的第一采集时间点的位置数据,确定为雷达装置在采集该第二采集时间点的点云数据时的位置数据;若在该第二采集时间点没有对应的第一采集时间点,则基于第一采集时间点与第二采集时间点之间的时间间隔,计算雷达装置在该第二采集时间点的位置数据。
在一种可能的实施方式中,在计算雷达装置在第二采集时间点的位置数据时,针对每一个第二采集时间点,可以先分别确定与该第二采集时间点相隔时间最短的两个第一采集时间点,两个第一采集时间点包括在该第二采集时间点之前的一个第一采集时间点和在该第二采集时间点之后的一个第一采集时间点,然后基于两个第一采集时间点分别对应的位置数据,确定雷达装置在该第二采集时间点的位置数据。
具体的,在基于两个第一采集时间点分别对应的位置数据,确定雷达装置在该第二采集时间点的位置数据时,如第一采集时间点和第二采集时间点是均匀分布的,则可以将与第二时间点间隔时间最短的两个第一采集时间点的位置数据的均值确定为雷达装置在该第二采集时间点的位置数据。
示例性的,如图2所示,图2中横线上方的点a 1、b 1、c 1、d 1、e 1、f 1、g 1均表示第一采集时间点,每个第一采集时间点都有对应的位置数据,横向下方的a 2、b 2、c 2、d 2、e 2均表示第二采集时间点,每个第二采集时间点都有对应的点云数据。其中,第一采集时间点a 1、d 1、g 1均有对应的第二采集时间点,具体的对应关系为a 1→a 2、d 1→c 2、g 1→e 2,则将第一采集时间点a 1对应的位置数据确定为雷达装置在第二采集时间点a 2的位置数据,将第一采集时间点d 1对应的位置数据确定为雷达装置在第二采集时间点c 2的位置数据,将第一采集时间点g 1对应的位置数据确定为雷达装置在第二采集时间点e 2的位置数据。而对于没有对应的第一采集时间点的第二采集时间点,如b 2,则选择与第二采集时间点b 2距离最近的第一采集时间点b 1和c 1,然后将第一采集时间点b 1和c 1分别对应的位置数据的均值,确定为雷达装置在第二采集时间点b 2的位置数据。以此类推,可计算雷达装置在第二采集时间点d 2的位置数据。
其中,在计算两个第一采集时间点的位置数据的均值时,可以将位置数据所包含的 每一项数据求均值。例如,位置数据包括经度、纬度、以及高度,则可以分别计算经度的均值、纬度的均值、以及高度的均值,然后将经度的均值、纬度的均值、以及高度的均值作为求均值之后的位置数据。
在基于两个第一采集时间点分别对应的位置数据,确定雷达装置在该第二采集时间点的位置数据时,若第一采集时间点和第二采集时间点不是均匀分布的,则在计算雷达装置在没有对应的第一采集时间点的第二采集时间点的位置数据时,可以基于两个第一采集时间点与第二采集时间点之间的间距进行确定。
示例性的,如图3所示,点a和点b为第一采集时间点,点c为第二采集时间点,则在确定雷达装置在第二采集时间点c的位置数据时,可以根据第一采集时间点a与第二采集时间点c之间的距离、第一采集时间点b与第二采集时间点c之间的距离,分别确定第一采集时间点a的位置数据对应的权重和第一采集时间点b的位置数据对应的权重,然后将第一采集时间点a的位置数据和第一采集时间点b的位置数据按照各自对应的权重进行加权求和,得到雷达装置在第二采集时间点c的位置数据。
通过上述实施方式,可以确定雷达装置在每个第二采集时间点的位置数据,避免了雷达装置和定位装置的数据采集频率的不同所带来的干扰。
针对步骤103:
表征地面的地面参数信息包括法向量和截距。具体在基于点云数据确定表征地面的地面参数信息时,可以参照图4所示的方法,包括以下几个步骤:
步骤401、基于所述点云数据进行平面拟合,得到拟合平面信息。
其中,在基于点云数据进行平面拟合时,可以基于随机抽样一致算法(Random sample consensus,RANSAC)进行平面拟合。具体的,可以从点云数据中随机选取预设比例的点云数据,并基于随机选取的点云数据进行平面拟合。
步骤402、基于所述点云数据中雷达扫描点的坐标信息,确定每一个雷达扫描点与所述拟合平面信息所指示的平面之间的距离。
步骤403、基于确定的每一个雷达扫描点与所述拟合平面信息所指示的平面之间的距离,进行点云数据筛选。
确定地面参数信息时,需要根据点云数据拟合出地面,因此,在进行点云数据筛选时,是将不在地面上的雷达扫描点过滤掉。具体的,在基于确定的每一个雷达扫描点与 拟合平面信息所指示的平面之间的距离,进行点云数据筛选时,可以将与拟合信息所指示的平面之间的距离超过预设距离范围的雷达扫描点过滤掉。
步骤404、判断是否满足预设迭代条件。
若满足预设迭代条件,则执行步骤405;
若不满足预设迭代条件,则基于筛选后的点云数据,返回执行步骤401,继续进行点云数据的筛选。
其中,预设迭代条件可以是达到预设的迭代次数,或者在步骤403中进行点云数据筛选时,所有的雷达扫描点与拟合平面信息所指示的平面之间的距离均在预设距离范围内。
步骤405、基于最终筛选的点云数据确定地面参数信息。
上述方式中,通过迭代拟合的方式,对点云数据进行多次筛选,然后再基于最终筛选的点云数据确定地面参数信息,提高了地面参数信息的精度。
在基于最后筛选的点云数据确定地面参数信息时,可以是基于最后筛选的点云数据再次进行平面拟合,然后确定拟合后的平面的地面参数信息,其中地面参数信息包括法向量和截距。
在一种可能的实施方式中,为了提高数据筛选的效率,在基于点云数据进行平面拟合之前,还可以执行以下至少一种数据筛选过程:
1、对点云数据进行降采样处理,使得点云数据中雷达扫描点的分布密度符合预设条件。
其中,由于点云数据的密度会对平面拟合的效率产生影响,因此可以通过对点云数据进行降采样处理,从而提高点云数据筛选的效率,进而提高地面参数信息的确定效率。示例性的,可以通过体素滤波法对点云数据进行降采样处理。
2、根据雷达装置的预设安装高度,确定目标坐标范围,并筛选出对应的坐标信息位于目标坐标范围内的雷达扫描点。
雷达装置的预设安装高度为雷达装置距离地面的高度,在建立世界坐标系时,可以以雷达装置所在的位置点为Z轴的起点,然后根据雷达装置的预设安装高度,确定目标坐标范围,该目标坐标范围可以用于表示位于地面的雷达扫描点在Z轴的高度范围,通过筛选出对应的坐标信息位于目标坐标范围内的雷达扫描点,可以对点云数据进行粗略 的筛选,进而可以提高点云数据筛选的效率,进而提高地面参数信息的确定效率。
雷达装置的位姿数据除了包括雷达装置在建立的世界坐标系中的坐标,还包括俯仰角、航偏角、以及翻滚角中的至少一种。在基于雷达装置在每个第二采集时间点的位置数据,确定雷达装置的初始位姿数据时,由于位置数据中包括经纬度和高度,因此,雷达装置在三维空间中的位置是确定的,因此对三维空间建立世界坐标系之后,雷达装置在世界坐标系中的位姿数据也是确定的。
考虑到在数据采集过程中,自动驾驶装置可能并不行驶在同一地平面中,因此,不同地平面之间的地面参数信息可能并不相同,因此,本公开另外一实施例中,还可以根据位置数据中的高度数据,以及位置数据和点云数据之间的对应关系,将点云数据划分至多个采集区间,每个采集区间包括多个第二采集时间点,在每个采集区间内可以看作车辆行驶在同一地平面上。
在一种可能的实施方式中,在根据位置数据中的高度数据,将点云数据划分至多个采集区间时,可以基于高度数据在各个第二采集时间点的高度值,对高度数据进行过滤处理,然后确定经过过滤处理后保留的高度数据中的至少一个极值点,以各个极值点对应的采集时间点作为采集区间的切分点,将点云数据划分至多个采集区间。
考虑到自动驾驶装置在行驶过程中所采集的高度数据,可能会受到地面障碍物的影响,例如障碍物引起的颠簸等,因此,可以先对自动驾驶装置采集的高度数据进行过滤处理。具体的,可以将高度值的波动频率高于设定频率或波动幅度小于设定幅度的高度值过滤掉。
示例性的,如图5所示,横坐标为第二采集时间点,纵坐标为每个第二采集时间点对应的位置数据中的高度数据,折线图中的点为极值点,每两个极值点之间的第二采集时间点为一个采集区间,每个采集区间中,包括多个第二采集时间点,同理也包括对应的点云数据,例如,包括多帧点云图像。
在将点云数据划分至多个采集区间之后,针对每个采集区间,可以基于该采集区间内的点云数据,确定该采集区间对应的地面参数信息,以及基于在采集该采集区间内的点云数据时雷达装置的位置数据,确定该采集区间对应的所述雷达装置的初始位姿数据。
在确定每个采集区间的地面参数信息、以及该采集区间对应的雷达装置的初始位姿数据之后,可以基于每个所述采集区间分别对应的所述地面参数信息和所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整。
在一种可能的实施方式中,在基于每个所述采集区间分别对应的所述地面参数信息和所述初始位姿数据,对外参数据进行调整时,可以参照图6所示的外参数据调整方法,包括以下几个步骤:
步骤601、基于每个采集区间对应的所述地面参数信息和所述初始位姿数据,确定每个采集区间对应的优化后的位姿数据。
具体的,在基于每个采集区间对应的所述地面参数信息和所述初始位姿数据,确定每个采集区间对应的优化后的位姿数据时,针对任一个采集区间,可以基于该采集区间对应的地面参数信息、以及初始位姿数据,确定使得目标函数的值最小的优化后的位姿数据,并将使得目标函数的值最小的位姿数据作为该采集区间对应的优化后的位姿数据,其中,目标函数为进行位姿数据优化前后的地面参数信息之差的绝对值,与进行位姿数据优化前后的位姿数据之差的绝对值之和。
示例性的,目标函数可以为如下公式所示:
Figure PCTCN2021102869-appb-000001
其中,
Figure PCTCN2021102869-appb-000002
表示第i帧点云图像的初始位姿数据,
Figure PCTCN2021102869-appb-000003
表示第i帧点云图像的初始位姿数据对应的待计算的优化后的位姿数据,
Figure PCTCN2021102869-appb-000004
表示计算优化前的地面参数信息,即基于在该采集区间内的点云数据确定出的地面参数信息,
Figure PCTCN2021102869-appb-000005
表示优化后的地面参数信息,N表示在该采集区间内的点云数据中的所包含的点云图像的个数,即在该采集区间内的第二采集时间点的个数。
针对任一采集区间,在该采集区间的每一个第二采集时间点,都会采集一组点云数据,该组点云数据可以构成一帧点云图像,在数据采集过程中,包括多个第二采集时间点,对应多组点云数据,每一组点云数据都对应一帧点云图像,对于每一组点云数据,都计算该组点云数据对应的初始位姿数据,并带入上述公式中,通过最小二乘法求解该公式,可以得到每一帧点云图像的初始位姿数据对应的优化后的位姿数据。
另外,通过求解上述方程,还可以得到优化后的地面参数信息,但优化后的地面参数信息与调整外参数据无关。
示例性的,若N=3,则上述公式可以展开为:
Figure PCTCN2021102869-appb-000006
通过最小二乘法,可以分别计算出第一帧点云图像对应的优化后的位姿数据
Figure PCTCN2021102869-appb-000007
第二帧点云图像对应的优化后的位姿数据
Figure PCTCN2021102869-appb-000008
第三帧点云图像对应的优化后的位姿数据
Figure PCTCN2021102869-appb-000009
以及优化后的地面参数信息
Figure PCTCN2021102869-appb-000010
需要说明的是,
Figure PCTCN2021102869-appb-000011
Figure PCTCN2021102869-appb-000012
可以是六维向量,向量中各元素分别表示雷达装置在世界坐标系中的坐标x、y、z、俯仰角、航偏角、以及翻滚角,
Figure PCTCN2021102869-appb-000013
Figure PCTCN2021102869-appb-000014
可以是四维向量,向量中各元素分别表示法向量的值和截距,法向量为三维向量。
步骤602、基于每个采集区间对应的所述初始位姿数据、以及优化后的位姿数据,对所述外参数据进行调整。
具体的,在基于每个采集区间对应的初始位姿数据、以及优化后的位姿数据,对外参数据进行调整时,可以先基于每个采集区间对应的初始位姿数据,以及优化后的位姿数据,确定该采集区间内的第一平均位姿变化量;然后基于每个采集区间内的第一平均位姿变化量,确定多个采集区间对应的第二平均位姿变化量;再基于第二平均位姿变化量,对外参数据进行调整。
在一种可能的实施方式中,在确定任一采集区间的第一平均位姿变化量时,针对该采集区间内的点云数据对应的每一帧点云图像,可以确定该帧点云图像对应的初始位姿数据与优化后的位姿数据之间的位姿变化量,然后基于点云数据对应的多帧点云图像的位姿变化量,计算所述第一平均位姿变化量。
针对任一帧点云图像,在计算该帧点云图像对应的初始位姿数据与优化后的位姿数据之间的位姿变化量时,可以分别计算初始位姿数据中的经纬度和高度、与优化后的位姿数据中的经纬度和高度对应的变化量。
在一种可能的实施方式中,在基于第二平均位姿变化量,对外参数据进行调整时,可以将第二平均位姿变化量与调整前的外参数据之间的乘积确定为调整后的外参数据。
通过上述方法,可以根据定位装置采集的定位数据以及雷达装置采集的雷达数据,确定地面参数信息、以及雷达装置的初始位姿数据,然后基于地面参数信息、以及雷达装置的初始位姿数据,对外参数据进行调整,在外参数据的调整过程中,无需借助其他设备对定位装置和雷达装置的相对位置进行标定,提高了外参数据的精度和调整效率,进而在确定雷达装置的位姿数据时效率和精度更高。
本领域技术人员可以理解,在具体实施方式的上述方法中,各步骤的撰写顺序并不意味着严格的执行顺序而对实施过程构成任何限定,各步骤的具体执行顺序应当以其功能和可能的内在逻辑确定。
基于同一发明构思,本公开实施例中还提供了与数据处理方法对应的数据处理装置,由于本公开实施例中的装置解决问题的原理与本公开实施例上述数据处理方法相似,因此装置的实施可以参见方法的实施,重复之处不再赘述。
参照图7所示,本公开实施例提供一种数据处理装置,所述装置包括:获取模块701、第一确定模块702、第二确定模块703以及调整模块704;其中,
获取模块701,用于获取定位装置采集的定位数据和雷达装置采集的点云数据;其中,所述定位装置和所述雷达装置部署在同一车辆上;
第一确定模块702,用于基于所述定位数据和所述点云数据,确定所述雷达装置在采集点云数据时的位置数据;
第二确定模块703,用于基于所述点云数据确定表征地面的地面参数信息,以及基于所述位置数据确定所述雷达装置的初始位姿数据;
调整模块704,用于基于所述地面参数信息、以及所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整。
在一种可能的实施方式中,所述点云数据中包括多个雷达扫描点的坐标信息。相应地,所述第二确定模块703,在基于所述点云数据确定表征地面的地面参数信息时,用于:基于所述点云数据进行平面拟合,得到拟合平面信息;基于所述点云数据中雷达扫描点的坐标信息,确定每一个雷达扫描点与所述拟合平面信息所指示的平面之间的距离;基于确定的每一个雷达扫描点与所述拟合平面信息所指示的平面之间的距离,进行 点云数据筛选,并基于筛选后的点云数据,返回基于所述点云数据进行平面拟合的步骤,直到达到预设迭代条件,基于最终筛选的点云数据确定所述地面参数信息。
在一种可能的实施方式中,所述第二确定模块703,还用于在基于所述点云数据进行平面拟合之前,执行以下至少一种数据筛选过程:对所述点云数据进行降采样处理,使得所述点云数据中雷达扫描点的分布密度符合预设条件;筛选出对应的坐标信息位于所述目标坐标范围内的雷达扫描点,其中所述目标坐标范围是根据所述雷达装置的预设安装高度确定的。
在一种可能的实施方式中,所述位置数据包括高度数据。相应地,所述第二确定模块703,还用于:基于所述位置数据中的高度数据,以及位置数据和点云数据之间的对应关系,将所述点云数据划分至多个采集区间。
在一种可能的实施方式中,所述第二确定模块703,在基于所述位置数据中的高度数据,以及位置数据和点云数据之间的对应关系,将所述点云数据划分至多个采集区间时,用于:对所述高度数据进行过滤,确定过滤后保留的高度数据中的至少一个极值点;以各个极值点对应的采集时间点作为采集区间的切分点,将所述点云数据划分至多个采集区间。
在一种可能的实施方式中,所述第二确定模块703,在基于所述点云数据确定表征地面的地面参数信息,以及基于所述位置数据确定所述雷达装置的初始位姿数据时,用于:针对每个采集区间,基于该采集区间内的点云数据,确定该采集区间对应的地面参数信息,以及基于在采集该采集区间内的点云数据时雷达装置的位置数据,确定该采集区间对应的所述雷达装置的初始位姿数据。相应地,所述调整模块704,在基于所述地面参数信息、以及所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整时,用于:基于每个采集区间对应的所述地面参数信息和所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整。
在一种可能的实施方式中,所述调整模块704,在基于每个采集区间对应的所述地面参数信息和所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整时,用于:基于每个采集区间对应的所述地面参数信息和所述初始位姿数据,确定每个采集区间对应的优化后的位姿数据;基于每个采集区间对应的所述初始位姿数据、以及优化后的位姿数据,对所述外参数据进行调整。
在一种可能的实施方式中,所述调整模块704,在基于每个采集区间对应的所述地面参数信息和所述初始位姿数据,确定每个采集区间对应的优化后的位姿数据时,用于:针对任一个采集区间,基于该采集区间对应的所述地面参数信息、以及所述初始位姿数据,确定使得目标函数的值最小的位姿数据,将使得目标函数的值最小的位姿数据作为该采集区间对应的优化后的位姿数据,其中,所述目标函数为进行位姿数据优化前后的地面参数信息之差的绝对值,与进行位姿数据优化前后的位姿数据之差的绝对值之和。
在一种可能的实施方式中,所述调整模块704,在基于每个采集区间对应的所述初始位姿数据、以及优化后的位姿数据,对所述外参数据进行调整时,用于:基于每个采集区间对应的初始位姿数据,及优化后的位姿数据,确定该采集区间内的第一平均位姿变化量;基于每个采集区间内的所述第一平均位姿变化量,确定所述多个采集区间对应的第二平均位姿变化量;基于所述第二平均位姿变化量,对所述外参数据进行调整。
在一种可能的实施方式中,所述调整模块704,在基于所述第二平均位姿变化量,对所述外参数据进行调整时,用于:将所述第二平均位姿变化量与调整前的外参数据的乘积确定为调整后的外参数据。
在一种可能的实施方式中,所述定位数据中包含在多个第一采集时间点分别采集的位置数据,所述点云数据中包含在多个第二采集时间点分别采集的点云数据;所述基于所述定位数据和所述点云数据,确定所述雷达装置在采集所述点云数据时的位置数据。相应地,所述第一确定模块702,在基于所述定位数据和所述点云数据,确定所述雷达装置在所述定位装置采集所述定位数据时的位置数据时,用于:针对每一个所述第二采集时间点,在所述多个第一采集时间点中存在与该第二采集时间点重叠的目标第一采集时间点的情况下,基于所述目标第一采集时间点对应的位置数据确定所述雷达装置在该第二采集时间点的位置数据;在所述多个第一采集时间点中不存在所述目标第一采集时间点的情况下,基于所述多个第一采集时间点中与该第二采集时间点相隔时间最短的两个相邻第一采集时间点分别对应的位置数据,确定所述雷达装置在该第二采集时间点的位置数据,该第二采集时间点位于所述两个相邻第一采集时间点之间。
关于装置中的各模块的处理流程、以及各模块之间的交互流程的描述可以参照上述方法实施例中的相关说明,这里不再详述。
在一些实施例中,本公开实施例提供的装置具有的功能或包含的模板可以用于执行上文方法实施例描述的方法,其具体实现可以参照上文方法实施例的描述,为了简 洁,这里不再赘述。
基于同一技术构思,本公开实施例还提供了一种计算机设备。参照图8所示,该计算机设备包括处理器801、存储器802、和总线803。其中,存储器802用于存储执行指令,包括内存8021和外部存储器8022;这里的内存8021也称内存储器,用于暂时存放处理器801中的运算数据,以及与硬盘等外部存储器8022交换的数据,处理器801通过内存8021与外部存储器8022进行数据交换。
当计算机设备800运行时,处理器801与存储器802之间通过总线803通信,使得处理器801在执行以下指令:获取定位装置采集的定位数据和雷达装置采集的点云数据;其中,所述定位装置和所述雷达装置部署在同一车辆上;基于所述定位数据和所述点云数据,确定所述雷达装置在采集点云数据时的位置数据;基于所述点云数据确定表征地面的地面参数信息,以及基于所述位置数据确定所述雷达装置的初始位姿数据;基于所述地面参数信息、以及所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整。
其中,处理器801执行的具体处理过程可参照上述方法实施例中的描述,这里不再展开说明。
本公开实施例还提供一种计算机可读存储介质,该计算机可读存储介质上存储有计算机程序,该计算机程序被处理器运行时执行上述方法实施例中所述的数据处理方法的步骤。其中,该存储介质可以是易失性或非易失的计算机可读取存储介质。
本公开实施例所提供的数据处理方法的计算机程序产品,包括存储了程序代码的计算机可读存储介质,所述程序代码包括的指令可用于执行上述方法实施例中所述的数据处理方法的步骤。该计算机程序产品可以具体通过硬件、软件或其结合的方式实现。在一个可选实施例中,所述计算机程序产品具体体现为计算机存储介质,在另一个可选实施例中,计算机程序产品具体体现为软件产品,例如软件开发包(Software Development Kit,SDK)等等。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的系统和装置的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。在本公开所提供的几个实施例中,应该理解到,所揭露的系统、装置和方法,可以通过其它的方式实现。以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,又例如,多个单元或组件 可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些通信接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本公开各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个处理器可执行的非易失的计算机可读取存储介质中。基于这样的理解,本公开的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本公开各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是:以上所述实施例,仅为本公开的具体实施方式,用以说明本公开的技术方案,而非对其限制,本公开的保护范围并不局限于此,尽管参照前述实施例对本公开进行了详细的说明,本领域的普通技术人员应当理解:任何熟悉本技术领域的技术人员在本公开揭露的技术范围内,其依然可以对前述实施例所记载的技术方案进行修改或可轻易想到变化,或者对其中部分技术特征进行等同替换;而这些修改、变化或者替换,并不使相应技术方案的本质脱离本公开实施例技术方案的精神和范围,都应涵盖在本公开的保护范围之内。因此,本公开的保护范围应所述以权利要求的保护范围为准。

Claims (14)

  1. 一种数据处理方法,包括:
    获取定位装置采集的定位数据和雷达装置采集的点云数据;其中,所述定位装置和所述雷达装置部署在同一车辆上;
    基于所述定位数据和所述点云数据,确定所述雷达装置在采集所述点云数据时的位置数据;
    基于所述点云数据确定表征地面的地面参数信息,以及基于所述位置数据确定所述雷达装置的初始位姿数据;
    基于所述地面参数信息、以及所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整。
  2. 根据权利要求1所述的方法,其特征在于,所述点云数据中包括多个雷达扫描点的坐标信息;所述基于所述点云数据确定表征地面的地面参数信息,包括:
    基于所述点云数据进行平面拟合,得到拟合平面信息;
    基于所述点云数据中雷达扫描点的坐标信息,确定每一个雷达扫描点与所述拟合平面信息所指示的平面之间的距离;
    基于所述确定的每一个雷达扫描点与所述拟合平面信息所指示的平面之间的距离,进行点云数据筛选,并
    基于筛选后的点云数据,返回所述基于所述点云数据进行平面拟合的步骤,直到达到预设迭代条件,
    基于最终筛选的点云数据确定所述地面参数信息。
  3. 根据权利要求2所述的方法,其特征在于,在基于所述点云数据进行平面拟合之前,执行以下至少一种数据筛选过程:
    对所述点云数据进行降采样处理,使得所述点云数据中雷达扫描点的分布密度符合预设条件;
    筛选出对应的坐标信息位于目标坐标范围内的雷达扫描点,其中所述目标坐标范围是根据所述雷达装置的预设安装高度确定的。
  4. 根据权利要求1所述的方法,其特征在于,所述位置数据包括高度数据;所述方法还包括:
    基于所述位置数据中的高度数据,以及所述位置数据和所述点云数据之间的对应关系,将所述点云数据划分至多个采集区间。
  5. 根据权利要求4所述的方法,其特征在于,所述基于所述位置数据中的高度数据,以及所述位置数据和所述点云数据之间的对应关系,将所述点云数据划分至多个采集区间,包括:
    对所述高度数据进行过滤,确定过滤后保留的高度数据中的至少一个极值点;
    以各个极值点对应的采集时间点作为采集区间的切分点,将所述点云数据划分至多个采集区间。
  6. 根据权利要求4所述的方法,其特征在于,
    所述基于所述点云数据确定表征地面的地面参数信息,以及基于所述位置数据确定所述雷达装置的初始位姿数据,包括:
    针对每个所述采集区间,
    基于该采集区间内的点云数据,确定该采集区间对应的地面参数信息,以及
    基于所述雷达装置在采集该采集区间内的点云数据时的位置数据,确定该采集区间对应的所述雷达装置的初始位姿数据;
    所述基于所述地面参数信息、以及所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整,包括:
    基于每个所述采集区间对应的所述地面参数信息和所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整。
  7. 根据权利要求6所述的方法,其特征在于,所述基于每个所述采集区间对应的所述地面参数信息和所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整,包括:
    基于每个所述采集区间对应的所述地面参数信息和所述初始位姿数据,确定每个所述采集区间对应的优化后的位姿数据;
    基于每个所述采集区间对应的所述初始位姿数据、以及所述优化后的位姿数据,对所述外参数据进行调整。
  8. 根据权利要求7所述的方法,其特征在于,基于每个所述采集区间对应的所述地面参数信息和所述初始位姿数据,确定每个所述采集区间对应的优化后的位姿数据,包括:
    针对任一个所述采集区间,
    基于该采集区间对应的所述地面参数信息、以及所述初始位姿数据,确定使得目标函数的值最小的位姿数据,
    将使得所述目标函数的值最小的位姿数据作为该采集区间对应的优化后的位姿数据,
    其中,所述目标函数为进行位姿数据优化前后的地面参数信息之差的绝对值,与进行位姿数据优化前后的位姿数据之差的绝对值之和。
  9. 根据权利要求7所述的方法,其特征在于,所述基于每个所述采集区间对应的所述初始位姿数据、以及所述优化后的位姿数据,对所述外参数据进行调整,包括:
    基于每个所述采集区间对应的初始位姿数据及所述优化后的位姿数据,确定该采集区间内的第一平均位姿变化量;
    基于每个所述采集区间内的所述第一平均位姿变化量,确定所述多个采集区间对应的第二平均位姿变化量;
    基于所述第二平均位姿变化量,对所述外参数据进行调整。
  10. 根据权利要求9所述的方法,其特征在于,所述基于所述第二平均位姿变化量,对所述外参数据进行调整,包括:
    将所述第二平均位姿变化量与调整前的外参数据的乘积确定为调整后的外参数据。
  11. 根据权利要求1所述的方法,其特征在于,所述定位数据中包含在多个第一采集时间点分别采集的位置数据,所述点云数据中包含在多个第二采集时间点分别采集的点云数据;所述基于所述定位数据和所述点云数据,确定所述雷达装置在采集所述点云数据时的位置数据,包括:
    针对每一个所述第二采集时间点,
    在所述多个第一采集时间点中存在与该第二采集时间点重叠的目标第一采集时间点的情况下,基于所述目标第一采集时间点对应的位置数据确定所述雷达装置在该第二采集时间点的位置数据;
    在所述多个第一采集时间点中不存在所述目标第一采集时间点的情况下,基于所述多个第一采集时间点中与该第二采集时间点相隔时间最短的两个相邻第一采集时间点分别对应的位置数据,确定所述雷达装置在该第二采集时间点的位置数据,该第二采集时间点位于所述两个相邻第一采集时间点之间。
  12. 一种数据处理装置,包括:
    获取模块,用于获取定位装置采集的定位数据和雷达装置采集的点云数据;其中,所述定位装置和所述雷达装置部署在同一车辆上;
    第一确定模块,用于基于所述定位数据和所述点云数据,确定所述雷达装置在采集所述点云数据时的位置数据;
    第二确定模块,用于基于所述点云数据确定表征地面的地面参数信息,以及基于所述位置数据确定所述雷达装置的初始位姿数据;
    调整模块,用于基于所述地面参数信息、以及所述初始位姿数据,对表征所述定位装置与所述雷达装置之间的相对位置关系的外参数据进行调整。
  13. 一种计算机设备,包括处理器、存储器和总线,所述存储器存储有所述处理器可执行的机器可读指令,当计算机设备运行时,所述处理器与所述存储器之间通过总线通信,所述机器可读指令被所述处理器执行时执行如权利要求1至11任一所述的数据处理方法。
  14. 一种计算机可读存储介质,其上存储有计算机程序,该计算机程序被处理器运行时执行如权利要求1至11任意一项所述的数据处理方法。
PCT/CN2021/102869 2020-10-30 2021-06-28 一种数据处理方法及装置 WO2022088723A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
KR1020227007210A KR20220058901A (ko) 2020-10-30 2021-06-28 데이터 처리 방법 및 장치
JP2022514853A JP2023503767A (ja) 2020-10-30 2021-06-28 データの処理方法及び装置

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011190058.9 2020-10-30
CN202011190058.9A CN112164063A (zh) 2020-10-30 2020-10-30 一种数据处理方法及装置

Publications (1)

Publication Number Publication Date
WO2022088723A1 true WO2022088723A1 (zh) 2022-05-05

Family

ID=73865259

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/102869 WO2022088723A1 (zh) 2020-10-30 2021-06-28 一种数据处理方法及装置

Country Status (4)

Country Link
JP (1) JP2023503767A (zh)
KR (1) KR20220058901A (zh)
CN (1) CN112164063A (zh)
WO (1) WO2022088723A1 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112164063A (zh) * 2020-10-30 2021-01-01 上海商汤临港智能科技有限公司 一种数据处理方法及装置
CN112835007B (zh) * 2021-01-07 2023-04-18 北京百度网讯科技有限公司 点云数据转换方法、装置、电子设备和存储介质
CN112946591A (zh) * 2021-02-26 2021-06-11 商汤集团有限公司 外参标定方法、装置、电子设备及存储介质
CN112946612B (zh) * 2021-03-29 2024-05-17 上海商汤临港智能科技有限公司 外参标定方法、装置、电子设备及存储介质
CN113484843A (zh) * 2021-06-02 2021-10-08 福瑞泰克智能系统有限公司 一种激光雷达与组合导航间外参数的确定方法及装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180357503A1 (en) * 2017-06-13 2018-12-13 TuSimple Sensor calibration and time system for ground truth static scene sparse flow generation
CN110443786A (zh) * 2019-07-25 2019-11-12 深圳一清创新科技有限公司 激光雷达点云滤波方法、装置、计算机设备和存储介质
CN111208492A (zh) * 2018-11-21 2020-05-29 长沙智能驾驶研究院有限公司 车载激光雷达外参标定方法及装置、计算机设备及存储介质
CN111435163A (zh) * 2020-03-18 2020-07-21 深圳市镭神智能系统有限公司 地面点云数据过滤方法、装置、探测系统及存储介质
CN112164063A (zh) * 2020-10-30 2021-01-01 上海商汤临港智能科技有限公司 一种数据处理方法及装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180357503A1 (en) * 2017-06-13 2018-12-13 TuSimple Sensor calibration and time system for ground truth static scene sparse flow generation
CN111208492A (zh) * 2018-11-21 2020-05-29 长沙智能驾驶研究院有限公司 车载激光雷达外参标定方法及装置、计算机设备及存储介质
CN110443786A (zh) * 2019-07-25 2019-11-12 深圳一清创新科技有限公司 激光雷达点云滤波方法、装置、计算机设备和存储介质
CN111435163A (zh) * 2020-03-18 2020-07-21 深圳市镭神智能系统有限公司 地面点云数据过滤方法、装置、探测系统及存储介质
CN112164063A (zh) * 2020-10-30 2021-01-01 上海商汤临港智能科技有限公司 一种数据处理方法及装置

Also Published As

Publication number Publication date
JP2023503767A (ja) 2023-02-01
KR20220058901A (ko) 2022-05-10
CN112164063A (zh) 2021-01-01

Similar Documents

Publication Publication Date Title
WO2022088723A1 (zh) 一种数据处理方法及装置
CN108152831B (zh) 一种激光雷达障碍物识别方法及系统
JP2020530569A (ja) 車両センサの較正及び位置特定
WO2022142185A1 (zh) 位姿数据的确定方法、装置、电子设备及车辆
CN110223380B (zh) 融合航拍与地面视角图像的场景建模方法、系统、装置
CN110763239B (zh) 滤波组合激光slam建图方法及装置
CN114216454B (zh) 一种gps拒止环境下基于异源图像匹配的无人机自主导航定位方法
CN113674412B (zh) 基于位姿融合优化的室内地图构建方法、系统及存储介质
CN112051575A (zh) 一种毫米波雷达与激光雷达的调整方法及相关装置
CN110736456A (zh) 稀疏环境下基于特征提取的二维激光实时定位方法
CN114077249B (zh) 一种作业方法、作业设备、装置、存储介质
CN115683141A (zh) 一种未知环境下用于自动驾驶的局部参考路径生成方法
WO2021081958A1 (zh) 地形检测方法、可移动平台、控制设备、系统及存储介质
JP3874363B1 (ja) 位置評定装置、位置評定方法および位置評定プログラム
CN117367412B (zh) 一种融合捆集调整的紧耦合激光惯导里程计与建图方法
CN117029870A (zh) 一种基于路面点云的激光里程计
CN117451032A (zh) 一种低算力与松耦合的激光雷达和imu的slam方法及系统
CN116879917A (zh) 一种激光雷达地形匹配辅助导航方法及系统
CN113155126A (zh) 一种基于视觉导航的多机协同目标高精度定位系统及方法
KR102130687B1 (ko) 다중 센서 플랫폼 간 정보 융합을 위한 시스템
WO2023226155A1 (zh) 多源数据融合定位方法、装置、设备及计算机存储介质
WO2023226154A1 (zh) 自主定位方法、装置、设备及计算机可读存储介质
CN116465393A (zh) 基于面阵激光传感器的同步定位和建图方法及装置
CN114519671B (zh) 无人机遥感影像动态快速拼接方法
CN115512054A (zh) 三维时空连续点云地图的构建方法

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2022514853

Country of ref document: JP

Kind code of ref document: A

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21884464

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 522432313

Country of ref document: SA

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 02.10.2023)