WO2023202157A1 - 铲斗坐标标定方法和装置、更新方法和设备、挖掘机 - Google Patents

铲斗坐标标定方法和装置、更新方法和设备、挖掘机 Download PDF

Info

Publication number
WO2023202157A1
WO2023202157A1 PCT/CN2022/143871 CN2022143871W WO2023202157A1 WO 2023202157 A1 WO2023202157 A1 WO 2023202157A1 CN 2022143871 W CN2022143871 W CN 2022143871W WO 2023202157 A1 WO2023202157 A1 WO 2023202157A1
Authority
WO
WIPO (PCT)
Prior art keywords
bucket
coordinate
coordinates
calibration
radar
Prior art date
Application number
PCT/CN2022/143871
Other languages
English (en)
French (fr)
Inventor
马厚雪
张坚
赵宇
濮洪钧
Original Assignee
江苏徐工工程机械研究院有限公司
江苏徐工国重实验室科技有限公司
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 江苏徐工工程机械研究院有限公司, 江苏徐工国重实验室科技有限公司 filed Critical 江苏徐工工程机械研究院有限公司
Publication of WO2023202157A1 publication Critical patent/WO2023202157A1/zh

Links

Images

Classifications

    • EFIXED CONSTRUCTIONS
    • E02HYDRAULIC ENGINEERING; FOUNDATIONS; SOIL SHIFTING
    • E02FDREDGING; SOIL-SHIFTING
    • E02F3/00Dredgers; Soil-shifting machines
    • E02F3/04Dredgers; Soil-shifting machines mechanically-driven
    • E02F3/28Dredgers; Soil-shifting machines mechanically-driven with digging tools mounted on a dipper- or bucket-arm, i.e. there is either one arm or a pair of arms, e.g. dippers, buckets
    • E02F3/30Dredgers; Soil-shifting machines mechanically-driven with digging tools mounted on a dipper- or bucket-arm, i.e. there is either one arm or a pair of arms, e.g. dippers, buckets with a dipper-arm pivoted on a cantilever beam, i.e. boom
    • E02F3/32Dredgers; Soil-shifting machines mechanically-driven with digging tools mounted on a dipper- or bucket-arm, i.e. there is either one arm or a pair of arms, e.g. dippers, buckets with a dipper-arm pivoted on a cantilever beam, i.e. boom working downwardly and towards the machine, e.g. with backhoes
    • EFIXED CONSTRUCTIONS
    • E02HYDRAULIC ENGINEERING; FOUNDATIONS; SOIL SHIFTING
    • E02FDREDGING; SOIL-SHIFTING
    • E02F9/00Component parts of dredgers or soil-shifting machines, not restricted to one of the kinds covered by groups E02F3/00 - E02F7/00
    • E02F9/20Drives; Control devices
    • EFIXED CONSTRUCTIONS
    • E02HYDRAULIC ENGINEERING; FOUNDATIONS; SOIL SHIFTING
    • E02FDREDGING; SOIL-SHIFTING
    • E02F9/00Component parts of dredgers or soil-shifting machines, not restricted to one of the kinds covered by groups E02F3/00 - E02F7/00
    • E02F9/26Indicating devices
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01BMEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00Measuring arrangements characterised by the use of optical techniques
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01BMEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B21/00Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
    • G01B21/22Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring angles or tapers; for testing the alignment of axes
    • 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
    • G01S17/02Systems using the reflection of electromagnetic waves other than radio waves
    • G01S17/06Systems determining position data of a target
    • 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

Definitions

  • the present disclosure relates to the field of intelligent engineering machinery, and in particular to a bucket coordinate calibration method and device, an update method and equipment, and an excavator.
  • sensing equipment can be used to detect the size of the material pile and the location of the target excavation point, and inform the excavator of the target location to realize unmanned automatic excavation operations of the excavator.
  • a bucket coordinate calibration method including:
  • the coordinate calibration matrix is determined, where the coordinate calibration matrix is the coordinate calibration of the middle bucket teeth in the radar system to the excavator coordinates.
  • the coordinate calibration matrix of the system is the coordinate calibration matrix of the system.
  • obtaining the radar point cloud data and angle sensor data of the bucket includes: collecting the radar point cloud data and angle sensor data of the bucket when the bucket is in multiple different positions.
  • determining the coordinates of the middle bucket teeth of the bucket in the radar coordinate system based on the radar point cloud data of the bucket includes: based on the radar point cloud data collected by the bucket at each spatial position, Determine the coordinates of the middle bucket tooth in the radar coordinate system.
  • determining the coordinates of the middle bucket tooth in the excavator coordinate system based on the angle sensor data of the bucket includes: determining based on the angle sensor data collected by the bucket at each spatial position. The coordinates of the middle bucket tooth in the excavator coordinate system.
  • determining the coordinates of the middle bucket teeth of the bucket in the radar coordinate system based on the radar point cloud data collected by the bucket at each spatial location includes: based on the radar point cloud data collected by the bucket at each spatial location.
  • the radar point cloud data is based on the implicit shape model algorithm to determine the coordinates of the middle bucket tooth in the radar coordinate system.
  • determining the coordinates of the middle bucket teeth of the bucket in the excavator coordinate system based on the angle sensor data collected by the bucket at each spatial position includes: based on the angle sensor data collected by the bucket at each spatial position.
  • the angle sensor data is used to obtain the positive solution of kinematics of the excavator device and determine the coordinates of the middle bucket tooth in the excavator coordinate system.
  • determining the coordinate calibration matrix based on the coordinates of the middle bucket teeth in the radar coordinate system and the excavator coordinate system includes:
  • the coordinate calibration matrix is a coordinate rotation and translation transformation matrix.
  • determining the coordinate calibration matrix according to the training set data includes:
  • direct linear transformation is used to determine the coordinate calibration matrix.
  • determining the coordinate calibration matrix based on the training set data further includes:
  • the interior points that meet the conditions are recorded and the coordinate calibration matrix is updated;
  • a coordinate calibration update method including:
  • the bucket coordinate calibration method as described in any of the above embodiments is used to determine a new coordinate calibration matrix
  • the coordinate calibration update method further includes:
  • the bucket radar point cloud data is collected to determine one coordinate of the middle bucket tooth in the radar coordinate system
  • the number of position point data pairs is accumulated, and then the step of determining whether the number of collected position point data pairs reaches a predetermined number of position points is performed again.
  • determining one coordinate of the middle bucket tooth in the radar coordinate system includes: based on an implicit shape model algorithm, obtaining one coordinate of the middle bucket tooth in the radar coordinate system. ; Determine whether the model similarity is greater than the predetermined similarity; if the model similarity is greater than the predetermined similarity, use one coordinate of the middle bucket tooth in the radar coordinate system.
  • determining a coordinate of the middle bucket tooth in the excavator coordinate system includes: obtaining the correct kinematics solution of the excavator device based on the angle sensor data, and determining the position of the middle bucket tooth in the excavator. 1 coordinate in the machine coordinate system.
  • a bucket coordinate calibration device including:
  • a data acquisition module configured to acquire radar point cloud data and angle sensor data of the bucket
  • the positioning module is configured to determine the coordinates of the middle bucket tooth in the radar coordinate system based on the bucket's radar point cloud data; and determine the coordinates of the bucket middle bucket tooth in the excavator coordinate system based on the bucket's angle sensor data. coordinate;
  • the calibration module is configured to determine a coordinate calibration matrix based on the coordinates of the middle bucket teeth of the bucket in the radar coordinate system and the excavator coordinate system, wherein the coordinate calibration matrix is the coordinates of the middle bucket teeth of the bucket in the radar system.
  • the coordinates are calibrated to the coordinate calibration matrix of the excavator coordinate system.
  • the bucket coordinate calibration device is used to perform operations to implement the bucket coordinate calibration method described in any of the above embodiments.
  • a coordinate calibration update device including:
  • a judgment device configured to judge whether the online error of the coordinate calibration matrix is greater than the predetermined allowable error; when the online error of the coordinate calibration matrix is greater than the predetermined allowable error, it is determined whether the number of collected position point data pairs reaches the predetermined number of position points ;
  • the bucket coordinate calibration device is configured to use the bucket coordinate calibration method to determine a new coordinate calibration matrix when the number of collected position point data pairs is equal to the number of predetermined position points;
  • the updating device is configured to update the coordinate calibration matrix.
  • the bucket coordinate calibration device is the bucket coordinate calibration device described in any of the above embodiments.
  • the coordinate calibration update device is configured to perform operations that implement the coordinate calibration update method described in any of the above embodiments.
  • a computer device including:
  • Memory used to store instructions
  • a processor configured to execute the instructions, so that the computer device executes and implements the bucket coordinate calibration method as described in any of the above embodiments, and/or executes and implements the coordinate calibration update as described in any of the above embodiments. method operation.
  • a calibration system including a laser radar and an angle sensor, and also includes at least one of a computer device, a coordinate calibration update device, and a bucket coordinate calibration device, wherein the computer device is as follows
  • the coordinate calibration and updating device is the coordinate calibration and updating device according to any of the above embodiments
  • the bucket coordinate calibration device is the bucket coordinate calibration device according to any of the above embodiments.
  • Bucket coordinate calibration device is provided.
  • an excavator including a lidar, and at least one of a computer device, a coordinate calibration update device, and a bucket coordinate calibration device, wherein the computer device is any one of the above
  • the coordinate calibration update device is the coordinate calibration update device as described in any of the above embodiments
  • the bucket coordinate calibration device is the bucket coordinate calibration as described in any of the above embodiments. device.
  • a computer-readable storage medium stores computer instructions, and when the instructions are executed by a processor, the shovel as described in any of the above embodiments is implemented. bucket coordinate calibration method, and/or implement operations of the coordinate calibration update method described in any of the above embodiments.
  • Figure 1 is a schematic diagram of the excavation working conditions of bulk materials of an excavator according to some embodiments of the present disclosure.
  • Figure 2 is a schematic diagram of some embodiments of the bucket coordinate calibration method of the present disclosure.
  • Figure 3 is a schematic diagram of other embodiments of the bucket coordinate calibration method of the present disclosure.
  • Figure 4 is a schematic diagram of some embodiments of the coordinate calibration update method of the present disclosure.
  • Figure 5 is a schematic diagram of other embodiments of the coordinate calibration update method of the present disclosure.
  • Figure 6 is a schematic diagram of some embodiments of the bucket coordinate calibration device of the present disclosure.
  • Figure 7 is a schematic diagram of some embodiments of the coordinate calibration update device of the present disclosure.
  • Figure 8 is a schematic structural diagram of some embodiments of a computer device of the present disclosure.
  • any specific values are to be construed as illustrative only and not as limiting. Accordingly, other examples of the exemplary embodiments may have different values.
  • the position of the bucket is detected through an angle sensor, which is installed on the excavator.
  • the automatic excavation operation function to the target excavation point is realized through bucket trajectory planning and control.
  • the measurement of the target excavation point must be accurate, and the prerequisite for accurate measurement is coordinate calibration, that is, the coordinates of the bucket and the target excavation point in the radar coordinate system are accurately calibrated to the excavator coordinate system Under this method, the bucket coordinates and the coordinates of the target excavation point are unified in the excavator coordinate system.
  • the scene feature method requires designing a specific scene for sensor calibration and cannot achieve online calibration of the sensor.
  • the related technology involves a sensor positioning system that can calculate the position of at least one or more autonomous vehicle sensors based on surface data.
  • a sensor positioning system that can calculate the position of at least one or more autonomous vehicle sensors based on surface data.
  • it requires multiple measurement conversions; second, it does not have error verification and online update functions.
  • the related art also discloses a method for calibrating sensor equipment installed on a machine, which can realize the calibration of the sensor equipment installed on the machine.
  • it is necessary to collect terrain point clouds with multiple characteristics and perform the process twice. Only by registration can the sensor equipment installed on the machine be calibrated; secondly, when the relative position of the sensor changes, online calibration cannot be performed.
  • Another related technology can achieve offline calibration, but firstly, it requires specific chessboard calibration boards and "L"-shaped joint calibration targets; secondly, it cannot be calibrated online.
  • Another related technology can realize offline calibration, but firstly, it requires a specific calibration version; secondly, the acquisition of the first coordinate value of each feature point on the calibration board in the coordinate system of the lower body of the working machine requires Use measuring tools such as tape measures for manual measurement, or control operating machinery for automatic measurement. However, manual measurement is troublesome and difficult to measure accurately. How to automatically measure is not explained; thirdly, it cannot be calibrated online.
  • the present disclosure provides a bucket coordinate calibration method and device, an update method and device, and an excavator, which can use laser radar and excavator angle sensors without adding external calibration equipment. Calibrate the bucket coordinates below.
  • the present disclosure will be described below through specific embodiments.
  • Figure 1 is a schematic diagram of the excavation working conditions of bulk materials of an excavator according to some embodiments of the present disclosure.
  • the operation scene shown in Figure 1 includes an excavator, materials to be excavated, lidar sensing equipment, and a transport vehicle (as an unloading point, not shown in the figure).
  • the excavator is parked near the material to be excavated, and its excavation radius can cover the material area.
  • the excavator can move as the location of the material changes.
  • lidar may be a lidar sensing device.
  • the lidar sensing device is installed outside the material to be excavated, and is used to collect point clouds of the material to be excavated in order to obtain a suitable target excavation point; and is also used to collect the bucket
  • the point cloud containing the middle bucket teeth can be collected, and the coordinates PL of the middle bucket teeth in the radar coordinate system can be obtained through the implicit shape model algorithm.
  • an angle sensor is installed on the excavator to collect the angle information of the excavator.
  • the position of the middle bucket tooth in the excavator coordinate system is calculated.
  • the coordinates of the middle bucket teeth of the bucket in the radar system are calibrated to the excavator coordinate system, so that the coordinates of the middle bucket teeth of the bucket and the coordinates of the target excavation point are within Unification under the excavator coordinate system.
  • FIG. 2 is a schematic diagram of some embodiments of the bucket coordinate calibration method of the present disclosure.
  • this embodiment can be executed by the disclosed bucket coordinate calibration device or the disclosed calibration system or the disclosed computer device or the disclosed coordinate calibration update device.
  • the method in the embodiment of Figure 2 may include at least one of steps 21 to 24, wherein:
  • Step 21 Obtain the radar point cloud data and angle sensor data of the bucket.
  • step 21 may include: collecting radar point cloud data and angle sensor data of the bucket when the bucket is in multiple different positions.
  • Step 22 Determine the coordinates of the middle tooth of the bucket in the radar coordinate system based on the radar point cloud data of the bucket.
  • step 22 may include: determining the coordinates of the middle bucket teeth of the bucket in the radar coordinate system based on the radar point cloud data collected by the bucket at each spatial position.
  • step 22 may include: determining the coordinates of the middle bucket tooth in the radar coordinate system based on the implicit shape model algorithm based on the radar point cloud data collected by the bucket at each spatial position.
  • Step 23 Determine the coordinates of the middle tooth of the bucket in the excavator coordinate system based on the angle sensor data of the bucket.
  • step 23 may include: determining the coordinates of the middle bucket teeth of the bucket in the excavator coordinate system based on the angle sensor data collected by the bucket at each spatial position.
  • step 23 may include: obtaining the correct kinematics solution of the excavator device based on the angle sensor data collected by the bucket at each spatial position, and determining the position of the middle bucket teeth of the bucket in the excavator coordinate system. coordinate.
  • Step 24 Determine the coordinate calibration matrix based on the coordinates of the middle bucket tooth in the radar coordinate system and the excavator coordinate system, where the coordinate calibration matrix is the coordinate calibration of the middle bucket tooth in the radar system to Coordinate calibration matrix of the excavator coordinate system.
  • the coordinate calibration matrix is a coordinate rotation and translation transformation matrix.
  • step 24 may include at least one of steps 241 to 243, wherein:
  • Step 241 Construct data pairs of radar coordinate system coordinates and excavator coordinate system coordinates based on the coordinates of the middle bucket tooth in the radar coordinate system and the excavator coordinate system, and divide the multiple data pairs into training sets and testing sets set.
  • Step 242 Determine the coordinate calibration matrix based on the training set data.
  • Step 243 Use the test set number to verify the coordinate calibration matrix.
  • This disclosure can use lidar and excavator angle sensors to calibrate bucket coordinates without adding external calibration equipment.
  • FIG. 3 is a schematic diagram of other embodiments of the bucket coordinate calibration method of the present disclosure.
  • this embodiment can be executed by the disclosed bucket coordinate calibration device or the disclosed calibration system or the disclosed computer device or the disclosed coordinate calibration update device.
  • the method in the embodiment of Figure 3 may include at least one of steps 31 to 36, wherein:
  • Step 31 Collect the radar point cloud data and angle sensor data of the bucket at N different positions.
  • N different positions refer to the movement of the bucket to N different spatial positions relative to the excavator coordinate system.
  • N is set in the program, and its value is at least greater than 4.
  • Bucket radar point cloud data is data collected based on the radar coordinate system.
  • Angle sensor data is data collected based on the excavator coordinate system.
  • system detection and control when collecting data, can be used to make N different spatial positions have greater differences, such as different rotation intervals, different bucket postures, and different arm positions.
  • Frame angle is used for collection to avoid the collection location being too concentrated and causing data correlation, which will lead to the problem of inability to converge when the coordinate calibration matrix calculation is performed in subsequent steps.
  • the system when collecting data, can be detected and controlled so that the bucket is within the radar's viewing angle at N different spatial positions, so that the radar can collect as much information about the bucket as possible.
  • Point cloud ensures the accuracy of subsequent configuration using implicit mode algorithms and improves similarity.
  • the slewing system, excavator arm, and bucket can move freely, but the positions of the excavator and the radar should be kept relatively fixed, otherwise Due to changes in the relative position of the excavator and the radar, the calibration may be inaccurate; of course, the system can also be configured to recalibrate when a change in relative position is detected.
  • Step 32 Based on the implicit shape model algorithm, calculate the coordinates PL of the middle bucket teeth of the N buckets in the radar coordinate system.
  • step 32 may include: based on the bucket radar point cloud collected at each spatial position of the bucket, based on the implicit shape model algorithm, calculate the shape of the middle bucket tooth in the radar coordinate system. Coordinates PL.
  • Step 33 Based on the forward kinematic solution of the angle sensor, calculate the coordinates PW of the corresponding N bucket middle bucket teeth in the excavator coordinate system;
  • step 33 may include: obtaining the coordinates PW of the middle bucket teeth of the bucket in the excavator coordinate system based on the angle sensor data collected by the bucket at each spatial position based on the forward kinematics solution.
  • Step 34 Construct (PL, PW) data pairs and randomly divide them into training sets and test sets.
  • step 34 may include: matching the middle bucket tooth coordinates PL and PW obtained in the above steps one-to-one according to the corresponding positions, constructing (PL, PW) data pairs, and randomly dividing the training set and test set, the division ratio can be set according to the cumulative number of collected points.
  • Step 35 Use the RANSAC (RANdom SAmple Consensus, Random Sampling Consensus) estimation algorithm on the training set data to obtain the coordinate calibration matrix R
  • RANSAC Random SAmple Consensus, Random Sampling Consensus
  • This disclosure uses the RANSAC estimation algorithm to avoid the interference of noisy data, and the estimated R
  • the purpose of coordinate calibration is to find the appropriate R
  • DLT Direct Linear Transformation
  • the implicit shape model algorithm is used to calculate the bucket coordinates based on the bucket point cloud.
  • the bottom layer is to use the registration method of the actual bucket point cloud and the model point cloud.
  • the bucket point cloud collected on site and the model Typically, there is registration similarity.
  • the higher the similarity the higher the accuracy of the calculated bucket coordinates. That is, the accuracy of the bucket coordinates is affected by the model similarity. Therefore, this solution uses the RANSAC random sampling estimation algorithm to solve the coordinate calibration matrix R.
  • step 242 of the FIG. 2 embodiment or step 35 of the FIG. 3 embodiment may include at least one of steps 351 to 359, wherein:
  • Step 351 Initialize relevant parameters, where the relevant parameters include parameters such as the number of iterations, threshold, maximum number of interior points, and probability of interior points.
  • Steps 352 to 359 are iterative calculations.
  • Step 352 Randomly select a predetermined number of first data pairs, where the data pairs are point pairs.
  • the predetermined number may be four.
  • step 352 may include randomly selecting 4 point pairs.
  • Step 353 Determine whether the first data pair is collinear. If collinearity is exceeded, return to step 351.
  • Step 354 If the first data pair is not collinear, use direct linear transformation to determine the coordinate calibration matrix, that is, use DLT to calculate R
  • Step 355 use the coordinate calibration matrix R
  • Step 356 Calculate the distance deviation between the transformed excavator coordinate system coordinates and the actual excavator coordinate system coordinates.
  • Step 357 Determine whether the distance deviation is less than a predetermined distance threshold.
  • Step 358 According to the number of iterations and the predetermined distance threshold, interior points that meet the conditions are recorded, and the coordinate calibration matrix R
  • Step 359 Calculate the interior point probability and update the number of iterations based on the interior point probability.
  • the present disclosure can robustly estimate model parameters using the RANSAC estimation algorithm, and can estimate high-precision parameters from a data set containing a large number of outlier points.
  • Step 36 Use the test set number to verify the coordinate calibration matrix R
  • This disclosure uses the test set number to verify the coordinate calibration matrix R
  • the coordinates of the middle bucket tooth in the radar coordinate system obtained through point cloud calculation are as follows:
  • test set is as follows:
  • the maximum error is 4.35cm and the mean square error is 1.89cm. Meet the needs of bulk material excavation operations.
  • Figure 4 is a schematic diagram of some embodiments of the coordinate calibration update method of the present disclosure. Preferably, this embodiment can be executed by either the disclosed calibration system or the disclosed computer device or the disclosed coordinate calibration update device.
  • the method in the embodiment of Figure 4 may include at least one of steps 41 to 44, wherein:
  • Step 41 Determine whether the online error of the coordinate calibration matrix is greater than the predetermined allowable error.
  • Step 42 If the online error of the coordinate calibration matrix is greater than the predetermined allowable error, determine whether the number of collected position point data pairs reaches the predetermined number of position points.
  • Step 43 When the number of collected position point data pairs is equal to the predetermined number of position points, use the bucket coordinate calibration method as described in any of the above embodiments (for example, the embodiment in Figure 2 or Figure 3) to determine a new coordinate calibration. matrix.
  • Step 44 Update the coordinate calibration matrix.
  • Figure 5 is a schematic diagram of other embodiments of the coordinate calibration update method of the present disclosure.
  • this embodiment can be executed by either the disclosed calibration system or the disclosed computer device or the disclosed coordinate calibration update device.
  • the method in the embodiment of Figure 5 may include at least one of steps S1 to S9, wherein:
  • Step S1 Set automatic calibration parameters: number of position points N, predetermined similarity A, division ratio B, and predetermined allowable error C.
  • the number of position points N refers to the movement of the bucket to N different spatial positions relative to the excavator coordinate system. N is set in the program, and its value is at least greater than 4.
  • the predetermined similarity A refers to the degree of registration similarity between the implicit shape model algorithm bucket point cloud and the model bucket point cloud.
  • the division ratio B refers to the division ratio of the (PL, PW) training set and the test set.
  • the predetermined allowable error C refers to an error that meets usage requirements.
  • the automatic calibration parameters can be set through the configuration file.
  • Step S2 Determine whether the online error is greater than the set predetermined allowable error C.
  • the online error is calculated by the PW-R*PL+T formula.
  • the online error will be greater than the set allowable error.
  • the automatic calibration program will be entered and step S3 will be entered. After the calibration is successful, the calibration program will be exited directly.
  • Step S3 Determine whether the number of collected point pairs (number of position point data pairs) is less than the set number N of position points. When the number of collected position point data pairs is less than the predetermined number of position points, step S4 is executed; when the number of collected position point data pairs is equal to the predetermined number of position points, step S8 is executed.
  • Step S4 Collect bucket radar point cloud data and determine one coordinate of the middle bucket tooth in the radar coordinate system.
  • step S4 may include: collecting bucket radar point cloud data, and obtaining the middle tooth coordinate PL1 of a bucket based on an implicit shape model algorithm.
  • the bucket when collecting data, the bucket is collected within the appropriate visual angle range of the radar according to the different positions of the bucket relative to the excavator, so as to collect a reasonable bucket point cloud and increase model configuration. Accurate success rate and improve model similarity.
  • Step S5 Determine whether the model similarity is greater than the predetermined setting value A.
  • step S6 is entered to obtain the bucket middle position.
  • Step S6 Collect the angle sensor data of the bucket and determine the coordinate of the middle tooth of the bucket in the excavator coordinate system.
  • step S6 may include: collecting the angle sensor data of the bucket, and calculating the middle bucket tooth coordinate PW1 of one bucket based on the forward kinematic solution of the angle sensor.
  • Step S7 Accumulation of location points.
  • step S7 may include counting the number of (PL1, PW1) point pairs that are successfully paired.
  • step S8 is entered.
  • Step S8 Use the calibration algorithm to calibrate the coordinate calibration matrix R
  • step S8 may include: in the case that the number of collected position point data pairs is equal to the predetermined number of position points, use the method as described in any of the above embodiments (for example, the embodiment of Figure 2 or Figure 3).
  • the bucket coordinate calibration method determines the new coordinate calibration matrix.
  • Step S9 Automatically update the coordinate calibration matrix R
  • T can be automatically updated or the operator can be reminded to decide whether to use it.
  • the present disclosure provides a method based on Coordinate calibration method and automatic update method of lidar and angle sensor:
  • the present disclosure does not require adding any calibration equipment to the existing system to complete the calibration.
  • this disclosure uses model recognition filtering and uses the RANSAC estimation algorithm to improve the calibration accuracy.
  • this disclosure randomly divides the data set and provides an online verification function.
  • the present disclosure can perform automatic calibration updates when the relative positions of the excavator and the radar change, and when the coordinate calibration matrix generates drift errors.
  • FIG. 6 is a schematic diagram of some embodiments of the bucket coordinate calibration device of the present disclosure.
  • the bucket coordinate calibration device of the present disclosure may include a data acquisition module 61, a positioning module 62 and a calibration module 63, wherein:
  • the data acquisition module 61 is configured to acquire radar point cloud data and angle sensor data of the bucket.
  • the data acquisition module 61 is configured to collect radar point cloud data and angle sensor data of the bucket when the bucket is in multiple different positions.
  • the positioning module 62 is configured to determine the coordinates of the middle tooth of the bucket in the radar coordinate system based on the radar point cloud data of the bucket; and determine the coordinates of the middle tooth of the bucket in the excavator coordinate system based on the angle sensor data of the bucket. coordinate of.
  • the positioning module 62 is configured to determine the coordinates of the middle bucket teeth of the bucket in the radar coordinate system based on the radar point cloud data of the bucket.
  • the collected radar point cloud data determines the coordinates of the middle bucket tooth in the radar coordinate system.
  • the positioning module 62 determines the coordinates of the middle bucket teeth of the bucket in the radar coordinate system according to the radar point cloud data of the bucket.
  • the radar point cloud data collected by the bucket at each spatial position is based on the implicit shape model algorithm to determine the coordinates of the middle bucket teeth in the radar coordinate system.
  • the positioning module 62 is configured to determine the coordinates of the middle bucket teeth of the bucket in the excavator coordinate system based on angle sensor data of the bucket.
  • the collected angle sensor data determines the coordinates of the middle bucket tooth in the excavator coordinate system.
  • the positioning module 62 is configured to determine the coordinates of the middle bucket teeth of the bucket in the excavator coordinate system based on angle sensor data of the bucket. Using the angle sensor data collected, the correct kinematics solution of the excavator device is obtained, and the coordinates of the middle bucket tooth in the excavator coordinate system are determined.
  • the calibration module 63 is configured to determine a coordinate calibration matrix based on the coordinates of the middle bucket teeth of the bucket in the radar coordinate system and the excavator coordinate system, where the coordinate calibration matrix is the coordinates of the middle bucket teeth of the bucket in the radar system.
  • the coordinates are calibrated to the coordinate calibration matrix of the excavator coordinate system.
  • the coordinate calibration matrix is a coordinate rotation and translation transformation matrix.
  • the calibration module 63 is configured to determine the coordinate calibration matrix according to the coordinates of the middle bucket teeth in the radar coordinate system and the excavator coordinate system.
  • the coordinates of the bucket teeth in the radar coordinate system and the excavator coordinate system construct data pairs of radar coordinate system coordinates and excavator coordinate system coordinates, and divide multiple data pairs into training sets and test sets; based on the training set data, Determine the coordinate calibration matrix; use the test set number to verify the coordinate calibration matrix.
  • the calibration module 63 when determining the coordinate calibration matrix according to the training set data, is configured to initialize relevant parameters, wherein the relevant parameters include the number of iterations; randomly select a predetermined number of A data pair; determine whether the first data pair is collinear; when the first data pair is not collinear, use direct linear transformation to determine the coordinate calibration matrix; use the coordinate calibration matrix to convert the radar coordinate system coordinates of the second data pair Transform to obtain the excavator coordinate system coordinates, where the second data pair is other data pairs in the training set except the first data pair; calculate the distance deviation between the transformed excavator coordinate system coordinates and the actual excavator coordinate system coordinates; determine the distance deviation Whether it is less than the predetermined distance threshold; judge based on the number of iterations and the predetermined distance threshold, record the interior points that meet the conditions, and update the coordinate calibration matrix; calculate the interior point probability and update the number of iterations based on the interior point probability.
  • the relevant parameters include the number of iterations; randomly select a predetermined number of
  • the bucket coordinate calibration device is used to perform operations to implement the bucket coordinate calibration method described in any of the above embodiments (for example, the embodiment of FIG. 2 or FIG. 3 ).
  • Figure 7 is a schematic diagram of some embodiments of the coordinate calibration update device of the present disclosure.
  • the coordinate calibration update device of the present disclosure may include a judgment device 71, a bucket coordinate calibration device 72 and an update device 73, wherein:
  • the judgment device 71 is configured to judge whether the online error of the coordinate calibration matrix is greater than the predetermined allowable error; when the online error of the coordinate calibration matrix is greater than the predetermined allowable error, it is determined whether the number of collected position point data pairs reaches the predetermined position point. quantity.
  • the bucket coordinate calibration device 72 is configured to determine a new coordinate calibration matrix using the bucket coordinate calibration method when the number of collected position point data pairs is equal to the number of predetermined position points.
  • the updating device 73 is configured to update the coordinate calibration matrix.
  • the bucket coordinate calibration device is the bucket coordinate calibration device described in any of the above embodiments (for example, the embodiment in FIG. 6 ).
  • the judgment device 71 is also configured to collect bucket radar point cloud data when the number of collected position point data pairs is less than the predetermined number of position points, and determine whether the middle bucket tooth of the bucket is in the radar position. 1 coordinate in the coordinate system; collect angle sensor data of the bucket, determine 1 coordinate of the middle tooth of the bucket in the excavator coordinate system; accumulate the number of position point data, and then perform judgment on the collected position point data again Operation on whether the quantity reaches the predetermined number of position points.
  • the judgment device 71 when determining one coordinate of the middle bucket tooth in the radar coordinate system, is configured to obtain the position of the middle bucket tooth in the radar coordinate system based on the implicit shape model algorithm. One coordinate in the coordinate system; determine whether the model similarity is greater than a predetermined similarity; if the model similarity is greater than the predetermined similarity, use one coordinate of the middle bucket tooth in the radar coordinate system.
  • the judgment device 71 when determining one coordinate of the middle bucket tooth in the excavator coordinate system, is configured to obtain the correct kinematics solution of the excavator device based on the angle sensor data, and determine The coordinate of the middle tooth of the bucket in the excavator coordinate system.
  • the coordinate calibration update device is configured to perform operations that implement the coordinate calibration update method described in any of the above embodiments (for example, the embodiment of FIG. 4 or FIG. 5 ).
  • the above embodiments of the present disclosure provide a coordinate calibration device and automatic update equipment based on lidar and angle sensors.
  • Lidar and excavator angle sensors are used to calibrate and automatically update bucket coordinates without adding external calibration equipment. , to achieve the unification of the bucket coordinates and the coordinates of the target excavation point in the excavator coordinate system.
  • Figure 8 is a schematic structural diagram of some embodiments of a computer device of the present disclosure. As shown in FIG. 8 , the computer device includes a memory 81 and a processor 82 .
  • the memory 81 is used to store instructions, and the processor 82 is coupled to the memory 81.
  • the processor 82 is configured to execute the method involved in implementing the above embodiments (such as any embodiment of FIG. 2 to FIG. 5) based on the instructions stored in the memory.
  • the computer device also includes a communication interface 83 for information interaction with other devices.
  • the computer device also includes a bus 84, through which the processor 82, the communication interface 83, and the memory 81 complete communication with each other.
  • the memory 81 may include high-speed RAM memory, or may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
  • the memory 81 may also be a memory array.
  • the storage 81 may also be divided into blocks, and the blocks may be combined into virtual volumes according to certain rules.
  • processor 82 may be a central processing unit (CPU), or may be an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present disclosure.
  • CPU central processing unit
  • ASIC application specific integrated circuit
  • a calibration system including a laser radar and an angle sensor, and also includes at least one of a computer device, a coordinate calibration update device, and a bucket coordinate calibration device, wherein,
  • the computer device is a computer device as described in any of the above embodiments (such as the embodiment in Figure 8)
  • the coordinate calibration update device is a coordinate calibration update device as described in any of the above embodiments (such as the embodiment in Figure 7)
  • the bucket coordinate calibration device is the bucket coordinate calibration device described in any of the above embodiments (for example, the embodiment in FIG. 6 ).
  • an excavator including a lidar, and at least one of a computer device, a coordinate calibration update device, and a bucket coordinate calibration device, wherein the computer
  • the device is a computer device as described in any of the above embodiments (such as the embodiment of Figure 8)
  • the coordinate calibration update device is a coordinate calibration update device as described in any of the above embodiments (such as the embodiment of Figure 7)
  • the bucket coordinate calibration device is the bucket coordinate calibration device described in any of the above embodiments (for example, the embodiment in FIG. 6 ).
  • the present disclosure provides a coordinate calibration method and automatic update system based on lidar and angle sensors.
  • lidar and excavator angle sensors By using lidar and excavator angle sensors, bucket coordinates can be calibrated and automatically updated without adding external calibration equipment, which can be achieved
  • the bucket coordinates and the coordinates of the target excavation point are unified in the excavator coordinate system.
  • a computer-readable storage medium stores computer instructions, which when executed by a processor implement any of the above embodiments (for example, FIG. 2 or the bucket coordinate calibration method described in the embodiment of Figure 3), and/or implement the operations of the coordinate calibration update method described in any of the above embodiments (for example, the embodiment of Figure 4 or Figure 5).
  • the computer-readable storage medium may be a non-transitory computer-readable storage medium.
  • embodiments of the present disclosure may be provided as methods, apparatuses, or computer program products. Accordingly, the present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment that combines software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable non-transitory storage media (including, but not limited to, disk memory, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein. .
  • These computer program instructions may also be stored in a computer-readable memory that causes a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction means, the instructions
  • the device implements the functions specified in a process or processes of the flowchart and/or a block or blocks of the block diagram.
  • These computer program instructions may also be loaded onto a computer or other programmable data processing device, causing a series of operating steps to be performed on the computer or other programmable device to produce computer-implemented processing, thereby executing on the computer or other programmable device.
  • Instructions provide steps for implementing the functions specified in a process or processes of a flowchart diagram and/or a block or blocks of a block diagram.
  • the computer device, bucket coordinate calibration device, data acquisition module, positioning module, calibration module, coordinate calibration update device, judgment device, bucket coordinate calibration device and update device described above can be implemented to perform what is described in this application.
  • PLC programmable logic controller
  • DSP digital signal processor
  • ASIC application specific integrated circuit
  • FPGA field programmable gate array
  • the program can be stored in a non-transitory computer-readable storage medium.
  • the storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Civil Engineering (AREA)
  • General Engineering & Computer Science (AREA)
  • Structural Engineering (AREA)
  • Mining & Mineral Resources (AREA)
  • Electromagnetism (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Mechanical Engineering (AREA)
  • Operation Control Of Excavators (AREA)

Abstract

一种铲斗坐标标定方法和装置、更新方法和设备、挖掘机。该铲斗坐标标定方法包括:获取铲斗的雷达点云数据和角度传感器数据(21);根据铲斗的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标(22);根据铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标(23);根据铲斗中间斗齿在雷达坐标系下和在挖掘机坐标系下的坐标,确定坐标标定矩阵(24),其中,坐标标定矩阵为铲斗中间斗齿在雷达系下的坐标标定到挖掘机坐标系的坐标标定矩阵。该方法可以利用激光雷达和挖掘机角度传感器,在不增加外部标定设备的情况下对铲斗坐标进行标定。

Description

铲斗坐标标定方法和装置、更新方法和设备、挖掘机
相关申请的交叉引用
本申请是以CN申请号为202211674601.1,申请日为2022年12月26日的申请为基础,并主张其优先权,该CN申请的公开内容在此作为整体引入本申请中。
技术领域
本公开涉及工程机械智能化领域,特别涉及一种铲斗坐标标定方法和装置、更新方法和设备、挖掘机。
背景技术
相关技术挖掘机散状物料挖掘作业工况而言,希望通过感知设备来检测料堆大小及目标挖掘点位置,并将目标位置告知挖掘机,实现挖掘机无人化的自动挖掘作业。
发明内容
根据本公开的一个方面,提供一种铲斗坐标标定方法,包括:
获取铲斗的雷达点云数据和角度传感器数据;
根据铲斗的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标;
根据铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标;
根据铲斗中间斗齿在雷达坐标系下和在挖掘机坐标系下的坐标,确定坐标标定矩阵,其中,所述坐标标定矩阵为铲斗中间斗齿在雷达系下的坐标标定到挖掘机坐标系的坐标标定矩阵。
在本公开的一些实施例中,所述获取铲斗的雷达点云数据和角度传感器数据包括:在铲斗处于多个不同位置下,采集铲斗的雷达点云数据和角度传感器数据。
在本公开的一些实施例中,所述根据铲斗的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标包括:根据铲斗在每个空间位置采集的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标。
在本公开的一些实施例中,所述根据铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标包括:根据铲斗在每个空间位置采集的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标。
在本公开的一些实施例中,所述根据铲斗在每个空间位置采集的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标包括:根据铲斗在每个空间位置采集的雷达点云数据,基于隐式形状模型算法,确定铲斗中间斗齿在雷达坐标系下的坐标。
在本公开的一些实施例中,所述根据铲斗在每个空间位置采集的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标包括:根据铲斗在每个空间位置采集的角度传感器数据,求得挖机装置运动学正解,确定铲斗中间斗齿在挖掘机坐标系下的坐标。
在本公开的一些实施例中,所述根据铲斗中间斗齿在雷达坐标系下和在挖掘机坐标系下的坐标,确定坐标标定矩阵包括:
根据铲斗中间斗齿在雷达坐标系下和在挖掘机坐标系下的坐标构建雷达坐标系坐标和挖掘机坐标系坐标的数据对,并将多个数据对划分为训练集和测试集;
根据训练集数据,确定坐标标定矩阵;
使用测试集数验证坐标标定矩阵。
在本公开的一些实施例中,所述坐标标定矩阵为坐标旋转平移变换矩阵。
在本公开的一些实施例中,所述根据训练集数据,确定坐标标定矩阵包括:
初始化相关参数,其中,所述相关参数包括迭代次数;
随机选择预定数量的第一数据对;
判断第一数据对是否共线;
在第一数据对不共线的情况下,采用直接线性变换确定坐标标定矩阵。
在本公开的一些实施例中,所述根据训练集数据,确定坐标标定矩阵还包括:
采用坐标标定矩阵,将第二数据对中的雷达坐标系坐标变换得到挖掘机坐标系坐标,其中,第二数据对为训练集中除第一数据对外的其它数据对;
计算变换得到的挖掘机坐标系坐标与实际挖掘机坐标系坐标的距离偏差;
判断距离偏差是否小于预定距离阈值;
根据迭代次数、预定距离阈值判断,记录符合条件的内点,更新坐标标定矩阵;
计算内点概率并根据内点概率更新迭代次数。
根据本公开的另一方面,提供一种坐标标定更新方法,包括:
判断坐标标定矩阵的在线误差是否大于预定许用误差;
若坐标标定矩阵的在线误差大于预定许用误差,则判断采集的位置点数据对数量是否达到预定位置点数量;
在采集的位置点数据对数量等于预定位置点数量的情况下,采用如上述任一实施例所 述的铲斗坐标标定方法确定新的坐标标定矩阵;
对坐标标定矩阵进行更新。
在本公开的一些实施例中,所述坐标标定更新方法还包括:
在采集的位置点数据对数量小于预定位置点数量的情况下,采集铲斗雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的1个坐标;
采集铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的1个坐标;
进行位置点数据对数量累计,之后再次执行判断采集的位置点数据对数量是否达到预定位置点数量的步骤。
在本公开的一些实施例中,所述确定铲斗中间斗齿在雷达坐标系下的1个坐标包括:基于隐式形状模型算法,得到铲斗中间斗齿在雷达坐标系下的1个坐标;判断模型相似度是否大于预定相似度;在模型相似度大于预定相似度的情况下,使用所述铲斗中间斗齿在雷达坐标系下的1个坐标。
在本公开的一些实施例中,所述确定铲斗中间斗齿在挖掘机坐标系下的1个坐标包括:基于角度传感器数据求得挖机装置运动学正解,确定铲斗中间斗齿在挖掘机坐标系下的1个坐标。
根据本公开的另一方面,提供一种铲斗坐标标定装置,包括:
数据获取模块,被配置为获取铲斗的雷达点云数据和角度传感器数据;
定位模块,被配置为根据铲斗的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标;根据铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标;
标定模块,被配置为根据铲斗中间斗齿在雷达坐标系下和在挖掘机坐标系下的坐标,确定坐标标定矩阵,其中,所述坐标标定矩阵为铲斗中间斗齿在雷达系下的坐标标定到挖掘机坐标系的坐标标定矩阵。
在本公开的一些实施例中,所述铲斗坐标标定装置用于执行实现如上述任一实施例所述的铲斗坐标标定方法的操作。
根据本公开的另一方面,提供一种坐标标定更新设备,包括:
判断装置,被配置为判断坐标标定矩阵的在线误差是否大于预定许用误差;在坐标标定矩阵的在线误差大于预定许用误差的情况下,判断采集的位置点数据对数量是否达到预定位置点数量;
铲斗坐标标定装置,被配置为在采集的位置点数据对数量等于预定位置点数量的情况下,采用铲斗坐标标定方法确定新的坐标标定矩阵;
更新装置,被配置为对坐标标定矩阵进行更新。
在本公开的一些实施例中,所述铲斗坐标标定装置为如上述任一实施例所述的铲斗坐标标定装置。
在本公开的一些实施例中,所述坐标标定更新设备用于执行实现如上述任一实施例所述的坐标标定更新方法的操作。
根据本公开的另一方面,提供一种计算机装置,包括:
存储器,用于存储指令;
处理器,用于执行所述指令,使得所述计算机装置执行实现如权上述任一实施例所述的铲斗坐标标定方法,和/或执行实现如上述任一实施例所述的坐标标定更新方法的操作。
根据本公开的另一方面,提供一种标定***,包括激光雷达和角度传感器,还包括计算机装置、坐标标定更新设备和铲斗坐标标定装置中的至少一项,其中,所述计算机装置为如上述任一实施例所述的计算机装置,所述坐标标定更新设备为如上述任一实施例所述的坐标标定更新设备,所述铲斗坐标标定装置为如上述任一实施例所述的铲斗坐标标定装置。
根据本公开的另一方面,提供一种挖掘机,包括激光雷达,还包括计算机装置、坐标标定更新设备和铲斗坐标标定装置中的至少一项,其中,所述计算机装置为如上述任一实施例所述的计算机装置,所述坐标标定更新设备为如上述任一实施例所述的坐标标定更新设备,所述铲斗坐标标定装置为如上述任一实施例所述的铲斗坐标标定装置。
根据本公开的另一方面,提供一种计算机可读存储介质,其中,所述计算机可读存储介质存储有计算机指令,所述指令被处理器执行时实现如上述任一实施例所述的铲斗坐标标定方法,和/或实现如上述任一实施例所述的坐标标定更新方法的操作。
附图说明
为了更清楚地说明本公开实施例或相关技术中的技术方案,下面将对实施例或相关技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本公开的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1为本公开一些实施例挖掘机散状物料的挖掘作业工况的示意图。
图2为本公开铲斗坐标标定方法一些实施例的示意图。
图3为本公开铲斗坐标标定方法另一些实施例的示意图。
图4为本公开坐标标定更新方法一些实施例的示意图。
图5为本公开坐标标定更新方法另一些实施例的示意图。
图6为本公开铲斗坐标标定装置一些实施例的示意图。
图7为本公开坐标标定更新设备一些实施例的示意图。
图8为本公开计算机装置一些实施例的结构示意图。
具体实施方式
下面将结合本公开实施例中的附图,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。以下对至少一个示例性实施例的描述实际上仅仅是说明性的,决不作为对本公开及其应用或使用的任何限制。基于本公开中的实施例,本领域普通技术人员在没有做出创造性劳动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。
除非另外具体说明,否则在这些实施例中阐述的部件和步骤的相对布置、数字表达式和数值不限制本公开的范围。
同时,应当明白,为了便于描述,附图中所示出的各个部分的尺寸并不是按照实际的比例关系绘制的。
对于相关领域普通技术人员已知的技术、方法和设备可能不作详细讨论,但在适当情况下,所述技术、方法和设备应当被视为授权说明书的一部分。
在这里示出和讨论的所有示例中,任何具体值应被解释为仅仅是示例性的,而不是作为限制。因此,示例性实施例的其它示例可以具有不同的值。
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讨论。
发明人通过研究发现:在挖掘机散状物料挖掘作业现场,通过激光雷达来检测料堆大小及目标挖掘点位置,其中激光雷达安装在挖掘机外侧。通过角度传感器来检测铲斗的位置,其中角度传感器安装在挖掘机上。通过铲斗轨迹规划和控制实现向目标挖掘点的自动挖掘作业功能。在这种场景下,要想实现精准挖掘,目标挖掘点的测量必须准确,而精确测量的前提是坐标标定,即将铲斗、目标挖掘点在雷达坐标系下的坐标准确标定到挖掘机坐标系下,实现铲斗坐标与目标挖掘点的坐标在挖掘机坐标系下的统一。
相关技术坐标标定常用的方法有:直接测量法、人工取点法、场景特征法等。相关技术这些方法还存在以下不足:直接测量法,标定精度低,直接测量和人工取点往往存在各 种误差,包括人为操作带来的误差。
发明人通过研究还发现:相关技术直接测量法、人工取点法、场景特征法还存在以下不足:
1)人工取点法,费时耗力,直接测试和人工取点都需要人为干预,如果对大量的设备进行标定,则会带来大量的工作量,难以实现智能***的量产化。
2)场景特征法,需要为传感器的标定设计特定的场景,无法实现对传感器的在线标定。
3)上述方法均无在作业过程中在线验证标定误差功能,尤其是当挖掘机与雷达的相对位置变化后,无法在线更新标定。
相关技术涉及一种传感器定位***,能够基于表面数据计算至少一个或多个自动驾驶汽车传感器的位置,但其一,需要多次测量转换;其二,不具有误差验证,在线更新功能。
相关技术还公开了一种用于校准安装在机器上的传感器设备的方法,能够实现安装在机器上的传感器装载的校准,但其一,需要采集具有多个特征的地形点云,进行两次配准才能实现对安装在机器上的传感器设备进行校准;其二,当传感器相对位置变化时,不能在线校准。
另一种相关技术能够实现离线标定,但其一,需要特定的包括棋盘标定板和“L”字形联合标定靶;其二,不能在线标定。
另一种相关技术能够实现离线标定,但其一,需要特定的标定版;其二,且标定板上的各特征点在作业机械的下车体坐标系中的第一坐标值的获取,需要通过测量工具如卷尺等进行手动测量,或控制作业机械进行自动测量。但手动测量麻烦而且很难测准,如何自动测量未做说明;其三,不能在线标定。
鉴于以上技术问题中的至少一项,本公开提供了一种铲斗坐标标定方法和装置、更新方法和设备、挖掘机,可以利用激光雷达和挖掘机角度传感器,在不增加外部标定设备的情况下对铲斗坐标进行标定。下面通过具体实施例对本公开进行说明。
图1为本公开一些实施例挖掘机散状物料的挖掘作业工况的示意图。如图1所示的作业场景包括挖掘机、待挖掘物料、激光雷达感知设备、运输车(作为卸料点,图中未给出)等组成。挖掘机停靠在待挖掘物料附近,其挖掘作业半径能够覆盖物料区域,在另一个场景中,当物料总量或位置随着挖掘作业施工变化时,挖掘机可以随着物料的位置变化而移动。
在本公开的一些实施例中,激光雷达可以为激光雷达感知设备。
在本公开的一些实施例中,如图1所示,激光雷达感知设备架设在待挖掘物料外侧,用于采集待挖掘物料的点云、以便得到适合的目标挖掘点;同时用于采集铲斗的点云,当铲斗中间斗齿露出在物料外面时,能够采集到含有铲斗中间斗齿的点云,通过隐式形状模型算法得到铲斗中间斗齿在雷达坐标系下的坐标PL。
在本公开的一些实施例中,如图1所示,挖掘机上安装有角度传感器,用于采集挖掘机的角度信息,通过运动学正解算法,计算得到铲斗中间斗齿在挖掘机坐标系下的坐标PW。
要想实现精准挖掘,必须进行坐标标定,本公开上述实施例将铲斗中间斗齿在雷达系下的坐标标定到挖掘机坐标系下,实现铲斗中间斗齿坐标与目标挖掘点的坐标在挖掘机坐标系下的统一。
图2为本公开铲斗坐标标定方法一些实施例的示意图。优选的,本实施例可由本公开铲斗坐标标定装置或本公开标定***或本公开计算机装置或本公开坐标标定更新设备执行。图2实施例的方法可以包括步骤21至步骤24中的至少一个步骤,其中:
步骤21,获取铲斗的雷达点云数据和角度传感器数据。
在本公开的一些实施例中,步骤21可以包括:在铲斗处于多个不同位置下,采集铲斗的雷达点云数据和角度传感器数据。
步骤22,根据铲斗的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标。
在本公开的一些实施例中,步骤22可以包括:根据铲斗在每个空间位置采集的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标。
在本公开的一些实施例中,步骤22可以包括:根据铲斗在每个空间位置采集的雷达点云数据,基于隐式形状模型算法,确定铲斗中间斗齿在雷达坐标系下的坐标。
步骤23,根据铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标。
在本公开的一些实施例中,步骤23可以包括:根据铲斗在每个空间位置采集的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标。
在本公开的一些实施例中,步骤23可以包括:根据铲斗在每个空间位置采集的角度传感器数据,求得挖机装置运动学正解,确定铲斗中间斗齿在挖掘机坐标系下的坐标。
步骤24,根据铲斗中间斗齿在雷达坐标系下和在挖掘机坐标系下的坐标,确定坐标标定矩阵,其中,所述坐标标定矩阵为铲斗中间斗齿在雷达系下的坐标标定到挖掘机坐标系的坐标标定矩阵。
在本公开的一些实施例中,所述坐标标定矩阵为坐标旋转平移变换矩阵。
在本公开的一些实施例中,步骤24可以包括步骤241-步骤243中的至少一个步骤,其中:
步骤241,根据铲斗中间斗齿在雷达坐标系下和在挖掘机坐标系下的坐标构建雷达坐标系坐标和挖掘机坐标系坐标的数据对,并将多个数据对划分为训练集和测试集。
步骤242,根据训练集数据,确定坐标标定矩阵。
步骤243,使用测试集数验证坐标标定矩阵。
本公开可以利用激光雷达和挖掘机角度传感器,在不增加外部标定设备的情况下对铲斗坐标进行标定。
图3为本公开铲斗坐标标定方法另一些实施例的示意图。优选的,本实施例可由本公开铲斗坐标标定装置或本公开标定***或本公开计算机装置或本公开坐标标定更新设备执行。图3实施例的方法可以包括步骤31至步骤36中的至少一个步骤,其中:
步骤31:在N个不同位置下,采集铲斗的雷达点云数据和角度传感器数据。
在本公开的一些实施例中,N个不同位置,指的是铲斗相对于挖掘机坐标系,运动到N个不同的空间位置,N在程序中设定,其值至少大于4。铲斗雷达点云数据是基于雷达坐标系采集的数据。角度传感器数据是基于挖掘机坐标系采集的数据。
在本公开的一些实施例中,在采集数据时,可通过***检测、控制使得N个不同的空间位置具有较大的差异性,比如按不同的回转间隔、不同的铲斗姿态、不同的臂架角度,来进行采集,避免采集位置过于集中,产生数据相关性现象,从而导致后续步骤中进行坐标标定矩阵计算时无法收敛的问题。
在本公开的一些实施例中,在采集数据时,可通过***检测、控制使得在N个不同的空间位置时,铲斗在雷达的视角范围内,使得雷达能够采集到铲斗尽可能多的点云,保障后续采用隐式模式算法的配置精度,提高相似度。
在本公开的一些实施例中,在采集数据,进行一次校准的过程中,上车回转***、挖掘机臂、铲斗可以***,但应保持挖掘机下车与雷达的位置相对固定,否则因挖掘机和雷达相对位置变化,可能会导致标定不准确;当然,***也可配置为,当检测相对位置变化时,重新标定。
步骤32,基于隐式形状模型算法,计算得到的N个铲斗中间斗齿在雷达坐标系下的坐标PL。
在本公开的一些实施例中,步骤32可以包括:针对铲斗在每个空间位置采集的铲斗雷达点云,基于隐式形状模型算法,计算得到铲斗中间斗齿在雷达坐标系下的坐标PL。
步骤33:基于角度传感器运动学正解,计算得到对应的N个铲斗中间斗齿在挖掘机坐标系下的坐标PW;
在本公开的一些实施例中,步骤33可以包括:针对铲斗在每个空间位置采集的角度传感器数据,基于运动学正解,得到铲斗中间斗齿在挖掘机坐标系下的坐标PW。
步骤34:构建(PL,PW)数据对,并随机划分为训练集和测试集。
在本公开的一些实施例中,步骤34可以包括:将上述步骤得到的铲斗中间斗齿坐标PL、PW按对应的位置一一对应,构建(PL,PW)数据对,并随机划分训练集和测试集,划分比例可根据累计采集的点的数量设定。
步骤35:对训练集数据使用RANSAC(RAndom SAmple Consensus,随机采样一致)估计算法得到坐标标定矩阵R|T。
在本公开的一些实施例中,步骤35可以包括:根据坐标标定模型PW=R*PL+T,对训练集数据使用RANSAC估计算法,计算得到坐标标定矩阵R|T。本公开使用RANSAC估计算法可以避免噪声数据的干扰,估计得到R|T准确可靠。
在本公开的一些实施例中,坐标标定的目的,就是找到合适的R|T,将铲斗中间斗齿在雷达坐标系下的坐标变换到挖掘机坐标系下,并计算与对应点的误差,期望所有数据对的均方误差最小,即求公式(1)的最小值。
Figure PCTCN2022143871-appb-000001
发明人通过研究发现:相关技术使用最小二乘法,DLT(Direct Linear Transformation,直接线性变换)对矩阵直接进行计算,但是该方法有时会因矩阵的数值计算、点对应不准确、尤其是存在离群点等原因,导致计算结果不稳定。
本公开方案的上述步骤中使用隐式形状模型算法根据铲斗点云来计算铲斗坐标,底层是利用实际铲斗点云与模型点云配准的方式,现场采集的铲斗点云与模型典型存在配准相似度,相似度越高,计算的铲斗坐标准确度越高,即铲斗坐标的准确度受模型相似度的影响,因此本方案使用RANSAC随机采样估计算法求解坐标标定矩阵R|T。
在本公开的一些实施例中,图2实施例的步骤242或图3实施例的步骤35可以包括步骤351至步骤359中的至少一个步骤,其中:
步骤351,初始化相关参数,其中,所述相关参数包括迭代次数、阈值、最大内点数、内点概率等参数。
步骤352至步骤359为迭代计算。
步骤352,随机选择预定数量的第一数据对,其中,所述数据对为点对。
在本公开的一些实施例中,预定数量可以为4。
在本公开的一些实施例中,步骤352可以包括:随机选择4个点对。
步骤353,判断第一数据对是否共线。如过共线,则返回步骤351。
步骤354,在第一数据对不共线的情况下,采用直接线性变换确定坐标标定矩阵,即,使用DLT计算R|Ti。
步骤355,采用坐标标定矩阵R|Ti,将第二数据对中的雷达坐标系坐标PLi变换得到挖掘机坐标系坐标PWi,其中,第二数据对为训练集中除第一数据对外的其它数据对。
步骤356,计算变换得到的挖掘机坐标系坐标与实际挖掘机坐标系坐标的距离偏差。
步骤357,判断距离偏差是否小于预定距离阈值。
步骤358,根据迭代次数、预定距离阈值判断,记录符合条件的内点,更新坐标标定矩阵R|T。
步骤359,计算内点概率并根据内点概率更新迭代次数。
经验证,本公开使用RANSAC估计算法,能够鲁棒的估计模型参数,能从包含大量局外点的数据集中估计出高精度的参数。
步骤36:使用测试集数验证坐标标定矩阵R|T。
本公开使用测试集数验证坐标标定矩阵R|T,如果误差满足使用要求,则完成标定。后续挖掘机作业过程,使用R|T将PW变换到雷达坐标系下,进行目标点挖掘和卸料点卸料。
下面通过具体实施例对本公开铲斗坐标标定方法进行说明。
具体实施例1:
经过现场试验,采集N=8个不同位置的PL,PW;其中,
通过点云计算得到的铲斗中间斗齿在雷达坐标系下的坐标,数据如下:
PL={3.233 2.986 3.233 3.634 3.972 2.986 2.789 2.651
-0.415 -0.738 -0.888 -0.367 -0.662 -0.23 -0.52 -0.696
0.057 -0.204 -0.38 -0.058 -0.266 -0.322 -0.074 0.253}
通过角度传感器运动学正解,得到铲斗中间斗齿在挖机坐标系下的坐标:
PW={4.27 4.20 3.94 4.04 3.60 4.6 4.53 4.43//x
0.74 1.13 1.07 0.50 0.45 0.80 1.12 1.33//y
0.36 0.11 -0.07 0.24 0.01 0.01 0.24 0.58}//z
进行数据集划分,划分比例6:4,得到训练集,如下:
PL_train={3.233 2.986 3.233 3.634 3.972
-0.415 -0.738 -0.888 -0.367 -0.662
0.057 -0.204 -0.38 -0.058 -0.266}
PW_train={4.27 4.20 3.94 4.04 3.603
0.74 1.13 1.07 0.50 0.45 0.80
0.36 0.11 -0.07 0.24 0.01}
测试集如下:
PL_test={2.986 2.789 2.651
-0.23 -0.52 -0.696
-0.322 -0.074 0.253}
PW_test={4.6 4.53 4.43
0.80 1.12 1.33
0.01 0.24 0.58}
使用RANSAC估计算法坐标标定矩阵,得到R|T,其中:
R={-0.7098 0.7033 -0.0408
-0.7028 -0.7109 -0.0273
-0.0482 0.0093 0.9988}
T={6.8549
2.7214
0.4704}
使用测试进行验证,res=PW_test–(PL_test*R+T),得到偏差矩阵,可以出每个点在x,y,z方向上的误差大小res,均方误差RE2,如下:
res={-0.0131 -0.0173 0.0435
-0.0048 0.0131 0.0162
-0.0073 0.0172 0.0088}
RE2=0.0189
最大误差为4.35cm,均方误差1.89cm。满足散装物料挖掘作业需求。
图4为本公开坐标标定更新方法一些实施例的示意图。优选的,本实施例可由或本公开标定***或本公开计算机装置或本公开坐标标定更新设备执行。图4实施例的方法可以包括步骤41至步骤44中的至少一个步骤,其中:
步骤41,判断坐标标定矩阵的在线误差是否大于预定许用误差。
步骤42,若坐标标定矩阵的在线误差大于预定许用误差,则判断采集的位置点数据对数量是否达到预定位置点数量。
步骤43,在采集的位置点数据对数量等于预定位置点数量的情况下,采用如上述任一实施例(例如图2或图3实施例)所述的铲斗坐标标定方法确定新的坐标标定矩阵。
步骤44,对坐标标定矩阵进行更新。
图5为本公开坐标标定更新方法另一些实施例的示意图。优选的,本实施例可由或本公开标定***或本公开计算机装置或本公开坐标标定更新设备执行。图5实施例的方法可以包括步骤S1至步骤S9中的至少一个步骤,其中:
步骤S1:设置自动标定参数:位置点数N,预定相似度A,划分比例B,预定许用误差C。
在本公开的一些实施例中,位置点数N(位置点数量N),指的是铲斗相对于挖掘机坐标系,运动到N个不同的空间位置,N在程序中设定,其值至少大于4。
在本公开的一些实施例中,预定相似度A,指的是隐式形状模型算法铲斗点云与模型铲斗点云的配准相似程度。
在本公开的一些实施例中,划分比例B,指的是(PL,PW)训练集与测试集的划分比例。
在本公开的一些实施例中,预定许用误差C,指的是满足使用要求的误差。所述自动标定参数可通过配置文件设定。
步骤S2:判断在线误差是否大于设定的预定许用误差C。
在本公开的一些实施例中,在线误差通过PW-R*PL+T公式计算。***初次使用时、或者挖掘机与雷达的相对位置变化时,在线误差会大于设定的许用误差,这时会进入自动标定程序,进入步骤S3,标定成功后,会直接退出标定程序。
步骤S3:判断采集的点对数(位置点数据对数量)是否小于设定的位置点数量N。在采集的位置点数据对数量小于预定位置点数量的情况下,执行步骤S4;在采集的位置点数据对数量等于预定位置点数量的情况下,执行步骤S8。
步骤S4:采集铲斗雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的1个坐标。
在本公开的一些实施例中,步骤S4可以包括:采集铲斗雷达点云数据,基于隐式形状模型算法,得到1个铲斗中间斗齿坐标PL1。
在本公开的一些实施例中,采集数据时,根据铲斗相对于挖掘机的不同位置,铲斗在 雷达的合适视觉角范围内进行采集,以便采集到合理的铲斗点云,增加模型配准成功率,提高模型相似度。
步骤S5:判断模型相似度是否大于预定设定值A。
在本公开的一些实施例中,若模型相似对大于预定设定值A,则接受基于隐式形状模型算法得到1个铲斗中间斗齿坐标PL1,并进入步骤S6,获取该位置铲斗中间斗齿在挖掘机坐标系下的坐标PW1。
步骤S6:采集铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的1个坐标。
在本公开的一些实施例中,步骤S6可以包括:采集铲斗的角度传感器数据,基于角度传感器运动学正解,计算1个铲斗中间斗齿坐标PW1。
步骤S7:位置点数累计。
在本公开的一些实施例中,步骤S7可以包括:统计配对成功的(PL1,PW1)点对数量。当数据等于N时进入步骤S8。
步骤S8:使用标定算法标定坐标标定矩阵R|T。
在本公开的一些实施例中,步骤S8可以包括:在采集的位置点数据对数量等于预定位置点数量的情况下,采用如上述任一实施例(例如图2或图3实施例)所述的铲斗坐标标定方法确定新的坐标标定矩阵。
步骤S9:自动更新坐标标定矩阵R|T。
本公开标定成功后,可自动更新R|T或提醒操作手来决定是否使用。
针对直接测量法、人工取点法、场景特征法等标定方法存在的标定精度低、费时耗力、需要特殊场景或专用标定设备、无在线验证,不能在线标定的问题,本公开提供一种基于激光雷达和角度传感器的坐标标定方法及自动更新方法:
其一,本公开不需要增加任何标定设备在现有***上即可完成标定。
其二,本公开使用模型相识度过滤,采用RANSAC估计算法提高了标定精度。
其三,本公开对数据集随机划分,提供了在线验证功能。
其四,本公开能够在挖掘机与雷达的相对位置变化时,以及坐标标定矩阵产生漂移误差时,进行自动标定更新。
图6为本公开铲斗坐标标定装置一些实施例的示意图。如图6所示,本公开铲斗坐标标定装置可以包括数据获取模块61、定位模块62和标定模块63,其中:
数据获取模块61,被配置为获取铲斗的雷达点云数据和角度传感器数据。
在本公开的一些实施例中,数据获取模块61,被配置为在铲斗处于多个不同位置下,采集铲斗的雷达点云数据和角度传感器数据。
定位模块62,被配置为根据铲斗的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标;根据铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标。
在本公开的一些实施例中,定位模块62在根据铲斗的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标的情况下,被配置为根据铲斗在每个空间位置采集的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标。
在本公开的一些实施例中,在本公开的一些实施例中,定位模块62在根据铲斗的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标的情况下,根据铲斗在每个空间位置采集的雷达点云数据,基于隐式形状模型算法,确定铲斗中间斗齿在雷达坐标系下的坐标。
在本公开的一些实施例中,定位模块62在根据铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标的情况下,被配置为根据铲斗在每个空间位置采集的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标。
在本公开的一些实施例中,定位模块62在根据铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标的情况下,被配置为根据铲斗在每个空间位置采集的角度传感器数据,求得挖机装置运动学正解,确定铲斗中间斗齿在挖掘机坐标系下的坐标。
标定模块63,被配置为根据铲斗中间斗齿在雷达坐标系下和在挖掘机坐标系下的坐标,确定坐标标定矩阵,其中,所述坐标标定矩阵为铲斗中间斗齿在雷达系下的坐标标定到挖掘机坐标系的坐标标定矩阵。
在本公开的一些实施例中,所述坐标标定矩阵为坐标旋转平移变换矩阵。
在本公开的一些实施例中,标定模块63,在根据铲斗中间斗齿在雷达坐标系下和在挖掘机坐标系下的坐标,确定坐标标定矩阵的情况下,被配置为根据铲斗中间斗齿在雷达坐标系下和在挖掘机坐标系下的坐标构建雷达坐标系坐标和挖掘机坐标系坐标的数据对,并将多个数据对划分为训练集和测试集;根据训练集数据,确定坐标标定矩阵;使用测试集数验证坐标标定矩阵。
在本公开的一些实施例中,标定模块63,在根据训练集数据,确定坐标标定矩阵的情况下,被配置为初始化相关参数,其中,所述相关参数包括迭代次数;随机选择预定数量的第一数据对;判断第一数据对是否共线;在第一数据对不共线的情况下,采用直接线性 变换确定坐标标定矩阵;采用坐标标定矩阵,将第二数据对中的雷达坐标系坐标变换得到挖掘机坐标系坐标,其中,第二数据对为训练集中除第一数据对外的其它数据对;计算变换得到的挖掘机坐标系坐标与实际挖掘机坐标系坐标的距离偏差;判断距离偏差是否小于预定距离阈值;根据迭代次数、预定距离阈值判断,记录符合条件的内点,更新坐标标定矩阵;计算内点概率并根据内点概率更新迭代次数。
在本公开的一些实施例中,所述铲斗坐标标定装置用于执行实现如上述任一实施例(例如图2或图3实施例)所述的铲斗坐标标定方法的操作。
图7为本公开坐标标定更新设备一些实施例的示意图。如图7所示,本公开坐标标定更新设备可以包括判断装置71、铲斗坐标标定装置72和更新装置73,其中:
判断装置71,被配置为判断坐标标定矩阵的在线误差是否大于预定许用误差;在坐标标定矩阵的在线误差大于预定许用误差的情况下,判断采集的位置点数据对数量是否达到预定位置点数量。
铲斗坐标标定装置72,被配置为在采集的位置点数据对数量等于预定位置点数量的情况下,采用铲斗坐标标定方法确定新的坐标标定矩阵。
更新装置73,被配置为对坐标标定矩阵进行更新。
在本公开的一些实施例中,所述铲斗坐标标定装置为如上述任一实施例(例如图6实施例)所述的铲斗坐标标定装置。
在本公开的一些实施例中,判断装置71,还被配置为在采集的位置点数据对数量小于预定位置点数量的情况下,采集铲斗雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的1个坐标;采集铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的1个坐标;进行位置点数据对数量累计,之后再次执行判断采集的位置点数据对数量是否达到预定位置点数量的操作。
在本公开的一些实施例中,判断装置71在确定铲斗中间斗齿在雷达坐标系下的1个坐标的情况下,被配置为基于隐式形状模型算法,得到铲斗中间斗齿在雷达坐标系下的1个坐标;判断模型相似度是否大于预定相似度;在模型相似度大于预定相似度的情况下,使用所述铲斗中间斗齿在雷达坐标系下的1个坐标。
在本公开的一些实施例中,判断装置71在确定铲斗中间斗齿在挖掘机坐标系下的1个坐标的情况下,被配置为基于角度传感器数据求得挖机装置运动学正解,确定铲斗中间斗齿在挖掘机坐标系下的1个坐标。
在本公开的一些实施例中,所述坐标标定更新设备用于执行实现如上述任一实施例(例如图4或图5实施例)所述的坐标标定更新方法的操作。
本公开上述实施例提供一种基于激光雷达和角度传感器的坐标标定装置及自动更新设备,利用激光雷达和挖掘机角度传感器,在不增加外部标定设备的情况下对铲斗坐标进行标定及自动更新,实现铲斗坐标与目标挖掘点的坐标在挖掘机坐标系下的统一。
图8为本公开计算机装置一些实施例的结构示意图。如图8所示,计算机装置包括存储器81和处理器82。
存储器81用于存储指令,处理器82耦合到存储器81,处理器82被配置为基于存储器存储的指令执行实现上述实施例(例如图2-图5任一实施例)涉及的方法。
如图8所示,该计算机装置还包括通信接口83,用于与其它设备进行信息交互。同时,该计算机装置还包括总线84,处理器82、通信接口83、以及存储器81通过总线84完成相互间的通信。
存储器81可以包含高速RAM存储器,也可还包括非易失性存储器(non-volatile memory),例如至少一个磁盘存储器。存储器81也可以是存储器阵列。存储器81还可能被分块,并且块可按一定的规则组合成虚拟卷。
此外,处理器82可以是一个中央处理器CPU,或者可以是专用集成电路ASIC,或是被配置成实施本公开实施例的一个或多个集成电路。
根据本公开的另一方面,如图1所示,提供一种标定***,包括激光雷达和角度传感器,还包括计算机装置、坐标标定更新设备和铲斗坐标标定装置中的至少一项,其中,所述计算机装置为如上述任一实施例(例如图8实施例)所述的计算机装置,所述坐标标定更新设备为如上述任一实施例(例如图7实施例)所述的坐标标定更新设备,所述铲斗坐标标定装置为如上述任一实施例(例如图6实施例)所述的铲斗坐标标定装置。
根据本公开的另一方面,如图1所示,提供一种挖掘机,包括激光雷达,还包括计算机装置、坐标标定更新设备和铲斗坐标标定装置中的至少一项,其中,所述计算机装置为如上述任一实施例(例如图8实施例)所述的计算机装置,所述坐标标定更新设备为如上述任一实施例(例如图7实施例)所述的坐标标定更新设备,所述铲斗坐标标定装置为如上述任一实施例(例如图6实施例)所述的铲斗坐标标定装置。
本公开提供一种基于激光雷达和角度传感器的坐标标定方法及自动更新***,利用激光雷达和挖掘机角度传感器,在不增加外部标定设备的情况下对铲斗坐标进行标定及自动 更新,可以实现铲斗坐标与目标挖掘点的坐标在挖掘机坐标系下的统一。
根据本公开的另一方面,提供一种计算机可读存储介质,其中,所述计算机可读存储介质存储有计算机指令,所述指令被处理器执行时实现如上述任一实施例(例如图2或图3实施例)所述的铲斗坐标标定方法,和/或实现如上述任一实施例(例如图4或图5实施例)所述的坐标标定更新方法的操作。
在本公开的一些实施例中,所述计算机可读存储介质可以为非瞬时性计算机可读存储介质。
本领域内的技术人员应明白,本公开的实施例可提供为方法、装置、或计算机程序产品。因此,本公开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,本公开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非瞬时性存储介质(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。
本公开是参照根据本公开实施例的方法、设备(***)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设备上,使得在计算机或其他可编程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。
在上面所描述的计算机装置、铲斗坐标标定装置、数据获取模块、定位模块、标定模块、坐标标定更新设备、判断装置、铲斗坐标标定装置和更新装置可以实现为用于执行本申请所描述功能的通用处理器、可编程逻辑控制器(PLC)、数字信号处理器(DSP)、专 用集成电路(ASIC)、现场可编程门阵列(FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件或者其任意适当组合。
至此,已经详细描述了本公开。为了避免遮蔽本公开的构思,没有描述本领域所公知的一些细节。本领域技术人员根据上面的描述,完全可以明白如何实施这里公开的技术方案。
本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成,也可以通过程序来指示相关的硬件完成,所述的程序可以存储于一种非瞬时性计算机可读存储介质中,上述提到的存储介质可以是只读存储器,磁盘或光盘等。
本公开的描述是为了示例和描述起见而给出的,而并不是无遗漏的或者将本公开限于所公开的形式。很多修改和变化对于本领域的普通技术人员而言是显然的。选择和描述实施例是为了更好说明本公开的原理和实际应用,并且使本领域的普通技术人员能够理解本公开从而设计适于特定用途的带有各种修改的各种实施例。

Claims (19)

  1. 一种铲斗坐标标定方法,包括:
    获取铲斗的雷达点云数据和角度传感器数据;
    根据铲斗的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标;
    根据铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标;
    根据铲斗中间斗齿在雷达坐标系下和在挖掘机坐标系下的坐标,确定坐标标定矩阵,其中,所述坐标标定矩阵为铲斗中间斗齿在雷达系下的坐标标定到挖掘机坐标系的坐标标定矩阵。
  2. 根据权利要求1所述的铲斗坐标标定方法,其中:
    所述获取铲斗的雷达点云数据和角度传感器数据包括:在铲斗处于多个不同位置下,采集铲斗的雷达点云数据和角度传感器数据;
    所述根据铲斗的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标包括:根据铲斗在每个空间位置采集的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标;
    所述根据铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标包括:根据铲斗在每个空间位置采集的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标。
  3. 根据权利要求2所述的铲斗坐标标定方法,其中:
    所述根据铲斗在每个空间位置采集的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标包括:根据铲斗在每个空间位置采集的雷达点云数据,基于隐式形状模型算法,确定铲斗中间斗齿在雷达坐标系下的坐标;
    所述根据铲斗在每个空间位置采集的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标包括:根据铲斗在每个空间位置采集的角度传感器数据,求得挖机装置运动学正解,确定铲斗中间斗齿在挖掘机坐标系下的坐标。
  4. 根据权利要求1-3中任一项所述的铲斗坐标标定方法,其中,所述根据铲斗中间斗齿在雷达坐标系下和在挖掘机坐标系下的坐标,确定坐标标定矩阵包括:
    根据铲斗中间斗齿在雷达坐标系下和在挖掘机坐标系下的坐标构建雷达坐标系坐标 和挖掘机坐标系坐标的数据对,并将多个数据对划分为训练集和测试集;
    根据训练集数据,确定坐标标定矩阵;
    使用测试集数验证坐标标定矩阵。
  5. 根据权利要求1-3中任一项所述的铲斗坐标标定方法,其中,所述坐标标定矩阵为坐标旋转平移变换矩阵。
  6. 根据权利要求4所述的铲斗坐标标定方法,其中,所述根据训练集数据,确定坐标标定矩阵包括:
    初始化相关参数,其中,所述相关参数包括迭代次数;
    随机选择预定数量的第一数据对;
    判断第一数据对是否共线;
    在第一数据对不共线的情况下,采用直接线性变换确定坐标标定矩阵。
  7. 根据权利要求6所述的铲斗坐标标定方法,其中,所述根据训练集数据,确定坐标标定矩阵还包括:
    采用坐标标定矩阵,将第二数据对中的雷达坐标系坐标变换得到挖掘机坐标系坐标,其中,第二数据对为训练集中除第一数据对外的其它数据对;
    计算变换得到的挖掘机坐标系坐标与实际挖掘机坐标系坐标的距离偏差;
    判断距离偏差是否小于预定距离阈值;
    根据迭代次数、预定距离阈值判断,记录符合条件的内点,更新坐标标定矩阵;
    计算内点概率并根据内点概率更新迭代次数。
  8. 一种坐标标定更新方法,包括:
    判断坐标标定矩阵的在线误差是否大于预定许用误差;
    若坐标标定矩阵的在线误差大于预定许用误差,则判断采集的位置点数据对数量是否达到预定位置点数量;
    在采集的位置点数据对数量等于预定位置点数量的情况下,采用如权利要求1-7中任一项所述的铲斗坐标标定方法确定新的坐标标定矩阵;
    对坐标标定矩阵进行更新。
  9. 根据权利要求8所述的坐标标定更新方法,还包括:
    在采集的位置点数据对数量小于预定位置点数量的情况下,采集铲斗雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的1个坐标;
    采集铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的1个坐标;
    进行位置点数据对数量累计,之后再次执行判断采集的位置点数据对数量是否达到预定位置点数量的步骤。
  10. 根据权利要求9所述的坐标标定更新方法,其中:
    所述确定铲斗中间斗齿在雷达坐标系下的1个坐标包括:基于隐式形状模型算法,得到铲斗中间斗齿在雷达坐标系下的1个坐标;判断模型相似度是否大于预定相似度;在模型相似度大于预定相似度的情况下,使用所述铲斗中间斗齿在雷达坐标系下的1个坐标;
    所述确定铲斗中间斗齿在挖掘机坐标系下的1个坐标包括:基于角度传感器数据求得挖机装置运动学正解,确定铲斗中间斗齿在挖掘机坐标系下的1个坐标。
  11. 一种铲斗坐标标定装置,包括:
    数据获取模块,被配置为获取铲斗的雷达点云数据和角度传感器数据;
    定位模块,被配置为根据铲斗的雷达点云数据,确定铲斗中间斗齿在雷达坐标系下的坐标;根据铲斗的角度传感器数据,确定铲斗中间斗齿在挖掘机坐标系下的坐标;
    标定模块,被配置为根据铲斗中间斗齿在雷达坐标系下和在挖掘机坐标系下的坐标,确定坐标标定矩阵,其中,所述坐标标定矩阵为铲斗中间斗齿在雷达系下的坐标标定到挖掘机坐标系的坐标标定矩阵。
  12. 根据权利要求11所述的铲斗坐标标定装置,其中,所述铲斗坐标标定装置用于执行实现如权利要求2-7中任一项所述的铲斗坐标标定方法的操作。
  13. 一种坐标标定更新设备,包括:
    判断装置,被配置为判断坐标标定矩阵的在线误差是否大于预定许用误差;在坐标标定矩阵的在线误差大于预定许用误差的情况下,判断采集的位置点数据对数量是否达到预定位置点数量;
    铲斗坐标标定装置,被配置为在采集的位置点数据对数量等于预定位置点数量的情 况下,采用铲斗坐标标定方法确定新的坐标标定矩阵;
    更新装置,被配置为对坐标标定矩阵进行更新。
  14. 根据权利要求13所述的坐标标定更新设备,其中,所述铲斗坐标标定装置为如权利要求11或12所述的铲斗坐标标定装置。
  15. 根据权利要求13所述的坐标标定更新设备,其中,所述坐标标定更新设备用于执行实现如权利要求8-10中任一项所述的坐标标定更新方法的操作。
  16. 一种计算机装置,包括:
    存储器,用于存储指令;
    处理器,用于执行所述指令,使得所述计算机装置执行实现如权利要求1-7中任一项所述的铲斗坐标标定方法,和/或执行实现如权利要求8-10中任一项所述的坐标标定更新方法的操作。
  17. 一种标定***,包括激光雷达和角度传感器,还包括计算机装置、坐标标定更新设备和铲斗坐标标定装置中的至少一项,其中,所述计算机装置为如权利要求16所述的计算机装置,所述坐标标定更新设备为如权利要求13-15中任一项所述的坐标标定更新设备,所述铲斗坐标标定装置为如权利要求11或12所述的铲斗坐标标定装置。
  18. 一种挖掘机,包括激光雷达,还包括计算机装置、坐标标定更新设备和铲斗坐标标定装置中的至少一项,其中,所述计算机装置为如权利要求16所述的计算机装置,所述坐标标定更新设备为如权利要求13-15中任一项所述的坐标标定更新设备,所述铲斗坐标标定装置为如权利要求11或12所述的铲斗坐标标定装置。
  19. 一种计算机可读存储介质,其中,所述计算机可读存储介质存储有计算机指令,所述指令被处理器执行时实现如权利要求1-7中任一项所述的铲斗坐标标定方法,和/或实现如权利要求8-10中任一项所述的坐标标定更新方法的操作。
PCT/CN2022/143871 2022-12-26 2022-12-30 铲斗坐标标定方法和装置、更新方法和设备、挖掘机 WO2023202157A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211674601.1 2022-12-26
CN202211674601.1A CN115950356A (zh) 2022-12-26 2022-12-26 铲斗坐标标定方法和装置、更新方法和设备、挖掘机

Publications (1)

Publication Number Publication Date
WO2023202157A1 true WO2023202157A1 (zh) 2023-10-26

Family

ID=87288498

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/143871 WO2023202157A1 (zh) 2022-12-26 2022-12-30 铲斗坐标标定方法和装置、更新方法和设备、挖掘机

Country Status (2)

Country Link
CN (1) CN115950356A (zh)
WO (1) WO2023202157A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117435853A (zh) * 2023-12-21 2024-01-23 山东科技大学 一种破土点位坐标计算方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109614743A (zh) * 2018-12-26 2019-04-12 广州海达安控智能科技有限公司 挖掘机及其铲斗定位方法、电子设备、存储介质
JP2020002708A (ja) * 2018-06-29 2020-01-09 日立建機株式会社 作業機械
CN111679306A (zh) * 2020-06-18 2020-09-18 万宝矿产有限公司 一种基于卫星导航的挖掘机智能化高精度定位方法
CN111708033A (zh) * 2020-06-17 2020-09-25 北京百度网讯科技有限公司 坐标系标定方法、装置、电子设备及存储介质
CN112095710A (zh) * 2020-09-16 2020-12-18 上海三一重机股份有限公司 挖掘机位姿显示方法、装置及其所应用的挖掘机
CN113034603A (zh) * 2019-12-09 2021-06-25 百度在线网络技术(北京)有限公司 用于确定标定参数的方法和装置
CN114186189A (zh) * 2021-11-19 2022-03-15 合肥联宝信息技术有限公司 坐标转换矩阵的计算方法、装置、设备及可读存储介质

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2020002708A (ja) * 2018-06-29 2020-01-09 日立建機株式会社 作業機械
CN109614743A (zh) * 2018-12-26 2019-04-12 广州海达安控智能科技有限公司 挖掘机及其铲斗定位方法、电子设备、存储介质
CN113034603A (zh) * 2019-12-09 2021-06-25 百度在线网络技术(北京)有限公司 用于确定标定参数的方法和装置
CN111708033A (zh) * 2020-06-17 2020-09-25 北京百度网讯科技有限公司 坐标系标定方法、装置、电子设备及存储介质
CN111679306A (zh) * 2020-06-18 2020-09-18 万宝矿产有限公司 一种基于卫星导航的挖掘机智能化高精度定位方法
CN112095710A (zh) * 2020-09-16 2020-12-18 上海三一重机股份有限公司 挖掘机位姿显示方法、装置及其所应用的挖掘机
CN114186189A (zh) * 2021-11-19 2022-03-15 合肥联宝信息技术有限公司 坐标转换矩阵的计算方法、装置、设备及可读存储介质

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117435853A (zh) * 2023-12-21 2024-01-23 山东科技大学 一种破土点位坐标计算方法
CN117435853B (zh) * 2023-12-21 2024-03-01 山东科技大学 一种破土点位坐标计算方法

Also Published As

Publication number Publication date
CN115950356A (zh) 2023-04-11

Similar Documents

Publication Publication Date Title
US11417008B2 (en) Estimating a volume of contents in a container of a work vehicle
US10852139B2 (en) Positioning method, positioning device, and robot
US10689830B2 (en) Container angle sensing using vision sensor for feedback loop control
EP3160690B1 (en) Automatic in-situ registration and calibration of robotic arm/sensor/workspace system
Martinelli et al. Estimating the odometry error of a mobile robot during navigation
CN113781582A (zh) 基于激光雷达和惯导联合标定的同步定位与地图创建方法
JP5776324B2 (ja) 地図処理方法及びプログラム、並びにロボットシステム
CN104715469A (zh) 一种数据处理方法及电子设备
WO2023202157A1 (zh) 铲斗坐标标定方法和装置、更新方法和设备、挖掘机
CN105136145A (zh) 一种基于卡尔曼滤波的四旋翼无人机姿态数据融合的方法
WO2022110451A1 (zh) 机器人定位方法、装置、计算机可读存储介质及机器人
WO2020132924A1 (zh) 机器人传感器的外参标定方法、装置、机器人及存储介质
CN113835422B (zh) 一种视觉地图构建方法和移动机器人
WO2022134144A1 (zh) 一种机器人质心规划方法、装置、可读存储介质及机器人
WO2022179094A1 (zh) 车载激光雷达外参数联合标定方法、***、介质及设备
CN110209180A (zh) 一种基于HuberM-Cubature卡尔曼滤波的无人水下航行器目标跟踪方法
Kallasi et al. A novel calibration method for industrial AGVs
Zheng et al. Traj-lo: In defense of lidar-only odometry using an effective continuous-time trajectory
CN112837314B (zh) 基于2D-LiDAR和Kinect的果树冠层参数检测***和方法
Nasir et al. Autonomous mobile robot localization using Kalman filter
CN114111791A (zh) 一种智能机器人室内自主导航方法、***及存储介质
CN109848988A (zh) 一种基于历史多帧点云信息融合的扫描匹配方法及***
CN117367412A (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
WO2022237375A1 (zh) 用于定位装置和里程计的标定方法、程序产品和标定装置
CN112697153A (zh) 自主移动设备的定位方法、电子设备及存储介质

Legal Events

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

Ref document number: 22938375

Country of ref document: EP

Kind code of ref document: A1