CN112415494B - AGV double-laser-radar position calibration method, device, equipment and storage medium - Google Patents
AGV double-laser-radar position calibration method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN112415494B CN112415494B CN202011461196.6A CN202011461196A CN112415494B CN 112415494 B CN112415494 B CN 112415494B CN 202011461196 A CN202011461196 A CN 202011461196A CN 112415494 B CN112415494 B CN 112415494B
- Authority
- CN
- China
- Prior art keywords
- point cloud
- double
- laser radars
- agv
- cloud data
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO 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
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
- G01S7/497—Means for monitoring or calibrating
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The invention discloses an AGV double-laser-radar position calibration method, an apparatus, a device and a storage medium, wherein the AGV double-laser-radar position calibration method comprises the following steps: acquiring double-laser-radar scanning data of the AVG provided with the double laser radars in a preset calibration environment, wherein the double laser radars are respectively arranged on two opposite sides of the AVG, and the scanning areas of the double laser radars have overlapping areas; converting scanning radar data of a scanning area overlapped by the double laser radars into point cloud data respectively corresponding to the double laser radars; and determining the calibration result of the double laser radars by using an ICP algorithm by taking one of the point cloud data respectively corresponding to the double laser radars as source point cloud data and the other point cloud data as target point cloud data. The AGV double-laser-radar position calibration method, the device, the system and the storage medium disclosed by the embodiment of the invention improve the positioning accuracy of the AVG with the double laser radars.
Description
Technical Field
The embodiment of the invention provides intelligent robot technology, and particularly relates to an AGV double-laser-radar position calibration method, device, equipment and storage medium.
Background
An Automatic Guided Vehicle (AGV) has the characteristics of high automation degree, safety, flexibility and the like, is gradually a key technology of a flexible production line and a modern storage system, and has been widely applied to the fields of automation production such as intelligent manufacturing and logistics.
In the manufacturing field, the AGV is required to be accurately positioned under the conditions of large load and narrow operation space in the production process, so that high requirements are provided for the positioning accuracy of the AGV. Because laser radar (Light Detection And Ranging, LiDAR) can realize area scanning, consequently current AGV carries out all-round scanning to automobile body surrounding environment through placing the two laser radar at locomotive And rear of a vehicle to fix a position And navigate according to two laser radar's data.
However, in the current AGV positioning and navigation, synchronous positioning and Mapping (SLAM) is adopted, and SLAM means that an AGV creates a map in a completely unknown environment under the condition that the position of the AGV is uncertain, and meanwhile, the map is used for autonomous positioning and navigation. However, due to the limitation of machining precision or installation precision, it is difficult to ensure that the relative positions of the two laser radars on the AGV strictly meet the design requirements, and these engineering errors directly affect the positioning precision of the AGV, so the calibration of the relative positions of the two laser radars is very important.
Disclosure of Invention
The invention provides an AGV double-laser-radar position calibration method, device and system and a storage medium, and improves the positioning accuracy of an AGV with double laser radars.
In a first aspect, an embodiment of the present invention provides an AGV dual laser radar position calibration method, including:
acquiring double-laser-radar scanning data of an AGV provided with double laser radars in a preset calibration environment, wherein the double laser radars are respectively arranged on two opposite sides of the AGV and the scanning areas of the double laser radars have overlapping areas;
converting scanning radar data of a scanning area overlapped by the double laser radars into point cloud data respectively corresponding to the double laser radars;
and determining the calibration result of the double laser radars by using an ICP algorithm by taking one of the point cloud data respectively corresponding to the double laser radars as source point cloud data and the other point cloud data as target point cloud data.
In a possible implementation manner of the first aspect, determining a calibration result of a dual laser radar by using an ICP algorithm with one of point cloud data respectively corresponding to the dual laser radar as source point cloud data and the other as target point cloud data includes:
determining a theoretical conversion matrix of the double laser radars according to the installation positions of the double laser radars on the AGV;
generating a source point cloud set according to one of the point cloud data respectively corresponding to the double laser radars, and determining a theoretical target point cloud set of the other point cloud data under a source point cloud data coordinate system according to a theoretical conversion matrix;
performing iterative computation on the theoretical conversion matrix according to the distance between the source point cloud set and the average near point pair in the target point cloud set until the distance between the source point cloud set and the average near point pair in the target point cloud set is smaller than a preset distance threshold or the iteration times reach preset iteration times;
and taking the product of the position of the laser radar corresponding to the target point cloud set and the conversion matrix after iteration as a calibration result.
In a possible implementation manner of the first aspect, performing iterative computation on the theoretical transformation matrix according to the distance between the source point cloud set and the average near point pair in the target point cloud set until the distance between the source point cloud set and the average near point pair in the target point cloud set is smaller than a preset distance threshold or the iteration number reaches a preset iteration number includes:
in each iterative calculation, a first point set with the closest point in the target point cloud set and the distance smaller than a preset distance threshold is searched in the source point cloud set, and a corresponding adjacent point set in the source point cloud set is a second point set;
decomposing SVD (singular value decomposition) on the singular values of the first point set and the second point set by using a least square method to obtain a rotation matrix and a translation vector of single iteration;
and updating the conversion matrix and the target point cloud set according to the rotation matrix and the translation vector of the single iteration.
In a possible implementation manner of the first aspect, updating the transformation matrix and the target point cloud set according to the rotation matrix and the translation vector of a single iteration includes:
according to
Updating a conversion matrix Tc, wherein r is a rotation matrix and t is a translation phasor;
and updating the target point cloud set according to p2 '═ Tc × p2_ L1, wherein p 2' is the updated target point cloud set, and p2_ L1 is the target point cloud set before updating.
In a possible implementation manner of the first aspect, the dual laser radars are respectively installed at two opposite corners of the AGV, and a scanning angle of the dual laser radars is greater than or equal to 270 degrees.
In a second aspect, an embodiment of the present invention provides an AGV dual laser radar position calibration apparatus, including:
the system comprises a radar data acquisition module, a calibration module and a calibration module, wherein the radar data acquisition module is used for acquiring double laser radar scanning data of an AGV provided with double laser radars in a preset calibration environment, the double laser radars are respectively arranged on two opposite sides of the AGV, and scanning areas of the double laser radars have overlapping areas;
the data conversion module is used for converting scanning radar data of a scanning area overlapped by the double laser radars into point cloud data respectively corresponding to the double laser radars;
and the position calibration module is used for determining a calibration result of the double laser radars by using an ICP (inductively coupled plasma) algorithm by taking one of the point cloud data respectively corresponding to the double laser radars as source point cloud data and the other point cloud data as target point cloud data.
In a possible implementation manner of the second aspect, the position calibration device is specifically configured to determine a theoretical conversion matrix of the dual laser radar according to an installation position of the dual laser radar on the AGV; generating a source point cloud set according to one of the point cloud data respectively corresponding to the double laser radars, and determining a theoretical target point cloud set of the other point cloud data under a source point cloud data coordinate system according to a theoretical conversion matrix; performing iterative computation on the theoretical conversion matrix according to the distance between the source point cloud set and the average near point pair in the target point cloud set until the distance between the source point cloud set and the average near point pair in the target point cloud set is smaller than a preset distance threshold or the iteration times reach preset iteration times; and taking the product of the target point cloud set and the conversion matrix after iteration as a calibration result.
In a third aspect, an embodiment of the present invention provides an AGV with dual laser radars, including:
the automatic guided vehicle comprises an AGV and double laser radars arranged on two opposite sides of the AGV, wherein scanning areas of the double laser radars are provided with overlapping areas;
the AGV double-laser-radar position calibration device is used for acquiring double-laser-radar scanning data of the AGV in a preset calibration environment; converting scanning radar data of a scanning area overlapped by the double laser radars into point cloud data respectively corresponding to the double laser radars; and determining the calibration result of the double laser radars by using an ICP algorithm by taking one of the point cloud data respectively corresponding to the double laser radars as source point cloud data and the other point cloud data as target point cloud data.
In one possible implementation manner of the third aspect, the AGV with dual laser radars further includes: and the double-laser radar data correction device is used for correcting the scanning data of the double-laser radar according to the calibration result of the double-laser radar.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the AGV dual-laser radar position calibration method according to any one of the implementations of the first aspect.
According to the AGV double-laser-radar position calibration method, the AGV double-laser-radar position calibration device, the AGV double-laser-radar scanning data with the double laser radars in the preset calibration environment are obtained firstly, then the scanning radar data of the scanning area overlapped by the double laser radars are converted into point cloud data respectively corresponding to the double laser radars, finally one of the point cloud data respectively corresponding to the double laser radars is used as source point cloud data, the other point cloud data is used as target point cloud data, the ICP algorithm is used for determining the calibration result of the double laser radars, the scanning result of the double laser radars is calibrated by the ICP algorithm, errors caused by manual calibration are avoided, and the positioning accuracy of the AGV with the double laser radars is improved.
Drawings
FIG. 1 is a flowchart illustrating an AGV double laser radar position calibration method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of an AGV with a dual laser radar;
FIG. 3 is a flowchart illustrating another AGV dual laser radar position calibration method according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of an AGV double-laser-radar position calibration device according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of an AGV with dual lidar according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Fig. 1 is a flowchart of an AGV dual laser radar position calibration method according to an embodiment of the present invention, and as shown in fig. 1, the AGV dual laser radar position calibration method according to the embodiment includes:
and S101, acquiring double-laser radar scanning data of the AGV provided with the double-laser radar in a preset calibration environment, wherein the double-laser radar is respectively arranged on two opposite sides of the AGV, and the scanning areas of the double-laser radar have overlapping areas.
The AGV double-laser-radar position calibration method provided by the embodiment is used for calibrating two laser radars of the AGV provided with the double laser radars so as to eliminate measurement errors caused by installation or manufacturing and other reasons of the two laser radars. The two laser radars of the AGV provided with the double laser radars are calibrated, and the difference between the actual measurement data of the two laser radars and the theoretical measurement data determined according to the installation position is actually determined, so that the actual measurement data of the two laser radars can be corrected according to the calibration result, and accurate data support is provided for the navigation, positioning and other work of the AGV.
The laser radar comprises a single beam narrow band laser transmitter and a receiving system. The laser transmitter sends out laser pulse waves, when the laser waves are emitted to the surface of an object, part of energy returns, and when the laser receiver receives the returned laser waves and the returned energy is enough to trigger a threshold value, the laser scanner calculates the distance value between the laser scanner and the object; the laser scanner continuously emits laser pulse waves, which impinge on a mirror rotating at high speed, and the laser pulse waves are emitted in various directions to form a two-dimensional area scan. This scanning of the two-dimensional area can perform the following two functions: 1) setting protection areas with different shapes in the scanning range of a scanner, and sending an alarm signal when an object enters the areas; 2) within the scanning range of the scanner, the scanner outputs the distance of each measuring point, and according to the distance information, the outline, coordinate positioning and the like of the object can be calculated.
Firstly, the AGV double-laser-radar position calibration method needs to acquire double-laser-radar scanning data of the AGV provided with the double laser radars in a preset calibration environment. Because AGV itself has certain volume, consequently install single laser radar on the AGV and can't carry out the measurement at no dead angle around the AGV, consequently can install two laser radar on the AGV at present, constitute the AGV that has two laser radar. Two laser radars of installation on the AGV can install respectively in the relative both sides of AGV, for example the locomotive department and the rear of a vehicle department of AGV. Install two laser radar on AGV all have certain scanning range, in order to realize demarcating two laser radar's position, two laser radar's scanning area need have the coincidence region. The scanning areas of the two laser radars on the AGV may have one overlapping area or multiple overlapping areas, and it should be noted that the scanning areas of the two laser radars need to exceed a certain area size, so that an accurate calibration result can be obtained.
In addition, the preset calibration environment is an environment with an obvious geometric shape, when the laser radar scans objects with an obvious set shape, the scanned data has obvious characteristics, so that the scanned data of the two laser radars can have obvious difference, and the calibration of the double-radar scanned data can be realized according to the obvious difference between the two scanned data.
In order to improve the scanning range of the laser radar, two laser radars on the AGV can be respectively installed at two opposite corners of the AGV. For example, as shown in FIG. 2, FIG. 2 is a schematic diagram of an AGV with a dual laser radar installed. Wherein, install first laser radar 22 and second laser radar 23 on AGV 21, the scanning area of first laser radar 22 is first region 24, and the scanning area of second laser radar 23 is second region 25. The first region 24 and the second region 25 have an overlap region 26. Two coinciding zones 26 are shown in fig. 2. As shown in fig. 2, the scanning angles of the first laser radar 22 and the second laser radar 23 on the AGV are 270 degrees, so that the scanning areas of the first laser radar 22 and the second laser radar 23 have the maximum overlapping area 26, so as to improve the accuracy of the calibration of the dual laser radars.
And S102, converting the scanning radar data of the scanning area overlapped by the double laser radars into point cloud data respectively corresponding to the double laser radars.
The traditional method for calibrating two laser radars on the AGV is to calibrate by a manual method. For example, the AGV can be statically placed at a wall corner where two wall surfaces intersect to ensure that two laser radars (a laser radar A and a laser radar B) can simultaneously scan wall surface 1 and wall surface 2 wall surface characteristics, then the AGV body is manually rotated for N times to obtain N groups of linear characteristic equations of the wall surface 1 and the wall surface 2 under the coordinate systems of the laser radar A and the laser radar B, and the average angle deviation value of the radar A relative to the radar B is calculated; and then, calculating the coordinates of the intersection points under the N groups of two radars to obtain N groups of average relative offset distances of the radar A relative to the radar B, and obtaining the relative position of the double laser radars. It is clear, however, that this solution is very susceptible to both wall flatness and lidar measurement errors fluctuating with measured distance.
Therefore, in this embodiment, the relative positions of the two laser radars are calibrated in a manner of Iterative Closest Point (ICP) algorithm registration. The ICP is a data registration algorithm, corresponding point pairs between source point clouds and target point clouds are obtained, a rotational translation matrix is constructed based on the corresponding point pairs, the source point clouds are transformed to a coordinate system of the target point clouds by using the obtained matrix, an error function of the transformed source point clouds and the transformed target point clouds is estimated, and if the error function value is larger than a threshold value, the operation is iterated until a given error requirement is met.
When the ICP algorithm is applied, firstly, scanning radar data of a scanning area where the double laser radars are overlapped needs to be converted into point cloud data respectively corresponding to the double laser radars, and the point cloud data refers to a set of vectors in a three-dimensional coordinate system.
And S103, determining a calibration result of the double laser radars by using an ICP algorithm by taking one of the point cloud data respectively corresponding to the double laser radars as source point cloud data and the other point cloud data as target point cloud data.
And after point cloud data respectively corresponding to the two laser radars are obtained, the point cloud data of one laser radar is used as source point cloud data, and the point cloud data of the other laser radar is used as target point cloud data. And then calculating according to the source point cloud data and the target point cloud data by using an ICP (inductively coupled plasma) algorithm to finally obtain a calibration result of the double laser radar.
The specific method for determining the calibration result of the dual laser radar by using the ICP algorithm can comprise the following steps: determining a theoretical conversion matrix of the double laser radars according to the installation positions of the double laser radars on the AGV; generating a source point cloud set according to one of the point cloud data respectively corresponding to the double laser radars, and determining a theoretical target point cloud set of the other point cloud data under a source point cloud data coordinate system according to a theoretical conversion matrix; performing iterative computation on the theoretical conversion matrix according to the distance between the source point cloud set and the average near point pair in the target point cloud set until the distance between the source point cloud set and the average near point pair in the target point cloud set is smaller than a preset distance threshold or the iteration times reach preset iteration times; and taking the product of the position of the laser radar corresponding to the target point cloud set and the conversion matrix after iteration as a calibration result.
The ICP algorithm is an iterative algorithm, and requires iterative computation on data for a plurality of times until a result reaches a preset condition or the number of iterations reaches a preset number. In this embodiment, first, theoretical conversion matrices of two laser radars need to be determined according to the installation positions of the two laser radars on the AGV, that is, the theoretical relative position relationship and the corresponding coordinate conversion relationship of the two laser radars are determined according to the theoretical installation positions of the two laser radars. The theoretical installation positions and the actual relative positions of the two laser radars have certain errors, and the calibration of the double laser radars needs to determine the errors.
And then generating a source point cloud set according to one of the point cloud data respectively corresponding to the two laser radars, and determining a theoretical target point cloud set of the other point cloud data in a source point cloud data coordinate system according to a theoretical conversion matrix, namely taking one laser radar as source data and converting scanning data of the other laser radar into a coordinate system of the source data so as to calculate. And finally, performing iterative computation on the theoretical conversion matrix according to the distance between the source point cloud set and the average near point pair in the target point cloud set until the distance between the source point cloud set and the average near point pair in the target point cloud set is smaller than a preset distance threshold or the iteration frequency reaches a preset iteration frequency. And judging whether the result meets the condition of ending iteration in each iteration calculation, if so, stopping the iteration calculation, and otherwise, performing the next iteration. And each iteration is carried out on the theoretical conversion matrix according to the distance between the source point cloud set and the average adjacent point pair in the target point cloud set. And after each iterative computation, updating the conversion matrix, and obtaining the conversion matrix after the iterative computation is completed.
And finally, after the iterative computation is completed, taking the product of the position of the laser radar corresponding to the target point cloud set and the conversion matrix after the iterative computation as a calibration result. The iterative conversion matrix represents the measurement error between the two laser radars, and the actual error between the two laser radars can be eliminated by multiplying the positions of the laser radars corresponding to the target point cloud set by the iterative conversion matrix.
The AGV double-laser-radar position calibration method comprises the steps of firstly obtaining double-laser-radar scanning data of an AGV provided with double laser radars in a preset calibration environment, then converting the scanning radar data of a scanning area overlapped by the double laser radars into point cloud data corresponding to the double laser radars respectively, finally taking one of the point cloud data corresponding to the double laser radars respectively as source point cloud data, taking the other one as target point cloud data, determining a calibration result of the double laser radars by using an ICP algorithm, and since the scanning result of the double laser radars is calibrated by using the ICP algorithm, errors caused by manual calibration are avoided, and the positioning accuracy of the AGV with the double laser radars is improved.
Fig. 3 is a flowchart of another AGV dual-lidar position calibration method according to an embodiment of the present invention, and as shown in fig. 3, the AGV dual-lidar position calibration method according to the embodiment includes:
step S301, acquiring double-laser radar scanning data of the AGV provided with the double-laser radar in a preset calibration environment, wherein the double-laser radar is respectively arranged on two opposite sides of the AGV, and the scanning areas of the double-laser radar have overlapping areas.
Step S302, converting the scanning radar data of the scanning area overlapped by the double laser radars into point cloud data respectively corresponding to the double laser radars.
In the present embodiment, the scanning data of the two laser radars are set to s1 and s2, respectively.
And step S303, determining a theoretical conversion matrix of the double laser radars according to the installation positions of the double laser radars on the AGV.
Taking the point cloud data corresponding to s1 as a theoretical value and the point cloud data corresponding to s2 as a measurement value, a theoretical transformation matrix T _ L1_ L2 can be obtained.
Step S304, a source point cloud set is generated according to one of the point cloud data respectively corresponding to the double laser radars, and a theoretical target point cloud set of the other point cloud data under a source point cloud data coordinate system is determined according to a theoretical conversion matrix.
Let the source cloud set generated from the point cloud data corresponding to s1 be p1, and let the target cloud set generated from the point cloud data corresponding to s2 be p 2. The theoretical target point cloud set of the target point cloud data in the source point cloud data coordinate system is p2_ L1.
Step S305, iterative calculation is carried out on the theoretical conversion matrix according to the distance between the source point cloud set and the average near point pair in the target point cloud set until the distance between the source point cloud set and the average near point pair in the target point cloud set is smaller than a preset distance threshold or the iteration frequency reaches a preset iteration frequency.
Specifically, for each iterative calculation of the ICP algorithm, it may be: in each iterative calculation, a first point set with the closest point in the target point cloud set and the distance smaller than a preset distance threshold is searched in the source point cloud set, and a corresponding adjacent point set in the source point cloud set is a second point set; performing Singular Value Decomposition (SVD) on the first point set and the second point set by using a least square method to obtain a rotation matrix and a translation vector of a single iteration; and updating the conversion matrix and the target point cloud set according to the rotation matrix and the translation vector of the single iteration.
First, let the iteration initial value target point cloud set p2 ═ p2_ L1, the initial transformation matrix Tc ═ I, and I is the identity matrix.
Then, a first point set p2n with the closest point in p 2' and the distance smaller than a preset distance threshold is searched for in the source point cloud set p1, and a corresponding adjacent point set in the source point cloud set p1 is a second point set p1 n.
Then, SVD decomposition is carried out according to p1n and p2n by using a least square method, and a rotation matrix r and a translation vector t of a single iteration are obtained. The transformation matrix Tc and the target point cloud set p 2' are then updated according to the rotation matrix r and the translation vector t of a single iteration.
Wherein, according to
And updating the conversion matrix Tc, updating the target point cloud set according to p2 '═ Tc × p2_ L1, wherein p 2' is the updated target point cloud set, and p2_ L1 is the target point cloud set before updating.
If the distance between the average proximity point pair (p1n to p2n) is still greater than the preset distance budget and the iteration number is less than the preset iteration number, the step is repeated until the distance between the average proximity point pair in the source point cloud set and the target point cloud set is less than the preset distance threshold or the iteration number reaches the preset iteration number.
And S306, taking the product of the position of the laser radar corresponding to the target point cloud set and the conversion matrix after iteration as a calibration result.
After the iterative computation is completed, an actual value of L2 relative to L1 is obtained, where L2 is the installation position of the lidar corresponding to the target point cloud set, and L2 is the installation position of the lidar corresponding to the source point cloud set. T _ L1_ L2 ═ Tc × T _ L1_ L2, which is the required calibration result.
Fig. 4 is a schematic structural diagram of an AGV dual-laser-radar position calibration device according to an embodiment of the present invention, and as shown in fig. 4, the AGV dual-laser-radar position calibration device according to the embodiment includes:
radar data acquisition module 41 for acquire the two laser radar scan data of installing two laser radar's AGV in predetermineeing the calibration environment, two laser radar install respectively in AGV's relative both sides, and two laser radar's scan area has the coincidence region.
And the data conversion module 42 is configured to convert the scanning radar data of the scanning area where the two laser radars overlap into point cloud data corresponding to the two laser radars respectively.
And the position calibration module 43 is configured to determine a calibration result of the dual laser radar by using an ICP algorithm, where one of the point cloud data respectively corresponding to the dual laser radar is used as source point cloud data, and the other is used as target point cloud data.
The double-laser-radar-position calibrating device for the AGV provided by the embodiment is used for realizing the technical scheme of the double-laser-radar-position calibrating method for the AGV shown in the figure 1, the realization principle and the technical effect are similar, and the details are not repeated here.
In the embodiment shown in fig. 4, the position calibration device 43 is specifically configured to determine a theoretical conversion matrix of the dual laser radar according to the installation position of the dual laser radar on the AGV; generating a source point cloud set according to one of the point cloud data respectively corresponding to the double laser radars, and determining a theoretical target point cloud set of the other point cloud data under a source point cloud data coordinate system according to a theoretical conversion matrix; performing iterative computation on the theoretical conversion matrix according to the distance between the source point cloud set and the average near point pair in the target point cloud set until the distance between the source point cloud set and the average near point pair in the target point cloud set is smaller than a preset distance threshold or the iteration times reach preset iteration times; and taking the product of the target point cloud set and the conversion matrix after iteration as a calibration result.
Fig. 5 is a schematic structural diagram of an AGV with dual laser radars according to an embodiment of the present invention, and as shown in fig. 5, the AGV with dual laser radars according to this embodiment includes:
AGV 51 and set up in the two laser radar of the relative both sides of AGV 51, be first laser radar 52 and second laser radar 53 respectively, first laser radar 52 and second laser radar 53's scanning area has coincidence region. An AGV dual laser radar position calibration device 54, configured to acquire dual laser radar scanning data of the AGV 51 in a preset calibration environment; converting scanning radar data of a scanning area overlapped by the double laser radars into point cloud data respectively corresponding to the double laser radars; and determining the calibration result of the double laser radars by using an iterative closest point ICP algorithm by taking one of the point cloud data respectively corresponding to the double laser radars as source point cloud data and the other point cloud data as target point cloud data.
The AGV dual-lidar position calibration device 54 may be the AGV dual-lidar position calibration device shown in fig. 4.
Further, the AGV with dual laser radars shown in fig. 5 further includes a dual laser radar data correction device, configured to correct the scanning data of the dual laser radars according to the calibration result of the dual laser radars.
The present invention also provides a storage medium containing computer executable instructions which, when executed by a computer processor, perform a method for AGV dual laser radar position calibration, the method comprising: acquiring double-laser-radar scanning data of an AGV provided with double laser radars in a preset calibration environment, wherein the double laser radars are respectively arranged on two opposite sides of the AGV and the scanning areas of the double laser radars have overlapping areas; converting scanning radar data of a scanning area overlapped by the double laser radars into point cloud data respectively corresponding to the double laser radars; determining the calibration result of the double laser radars by using an ICP (inductively coupled plasma) algorithm by taking one of the point cloud data respectively corresponding to the double laser radars as source point cloud data and the other point cloud data as target point cloud data
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments illustrated herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (8)
1. A method for calibrating the position of an AGV (automatic guided vehicle) with double laser radars is characterized by comprising the following steps:
acquiring double-laser-radar scanning data of an Automatic Guided Vehicle (AGV) provided with double laser radars in a preset calibration environment, wherein the double laser radars are respectively arranged on two opposite sides of the AGV and the scanning areas of the double laser radars have overlapping areas;
converting scanning radar data of a scanning area overlapped by the double laser radars into point cloud data respectively corresponding to the double laser radars;
determining a calibration result of the double laser radars by using an iterative closest point ICP (inductively coupled plasma) algorithm, wherein one of the point cloud data respectively corresponding to the double laser radars is used as source point cloud data, and the other point cloud data is used as target point cloud data;
the determining the calibration result of the dual laser radar by using an ICP algorithm with one of the point cloud data respectively corresponding to the dual laser radar as source point cloud data and the other as target point cloud data comprises the following steps:
determining a theoretical conversion matrix of the double laser radars according to the installation positions of the double laser radars on the AGV;
generating a source point cloud set according to one of the point cloud data respectively corresponding to the double laser radars, and determining a theoretical target point cloud set of the other point cloud data under a source point cloud data coordinate system according to the theoretical conversion matrix;
performing iterative computation on the theoretical conversion matrix according to the distance between the average proximity point pair in the source point cloud set and the average proximity point pair in the target point cloud set until the distance between the average proximity point pair in the source point cloud set and the average proximity point pair in the target point cloud set is smaller than a preset distance threshold or the iteration times reach preset iteration times;
and taking the product of the position of the laser radar corresponding to the target point cloud set and the conversion matrix after iteration as the calibration result.
2. The method of claim 1, wherein iteratively calculating the theoretical transformation matrix according to the distance between the source point cloud set and the average neighbor pair in the target point cloud set until the distance between the source point cloud set and the average neighbor pair in the target point cloud set is less than a preset distance threshold or the iteration number reaches a preset iteration number comprises:
in each iterative calculation, a first point set with the closest point and the distance less than a preset distance threshold value in the target point cloud set is searched in a source point cloud set, and a corresponding adjacent point set in the source point cloud set is a second point set;
decomposing SVD (singular value decomposition) on the singular values of the first point set and the second point set by using a least square method to obtain a rotation matrix and a translation vector of single iteration;
and updating the conversion matrix and the target point cloud set according to the rotation matrix and the translation vector of the single iteration.
3. The method of claim 2, wherein updating the transformation matrix and the set of target point clouds according to the rotation matrix and the translation vector of the single iteration comprises:
according to
Updating a conversion matrix Tc, wherein r is a rotation matrix and t is a translation phasor;
and updating the target point cloud set according to p2 '= Tc p2_ L1, wherein p 2' is the updated target point cloud set, and p2_ L1 is the target point cloud set before updating.
4. The method according to any one of claims 1 to 3, wherein the dual laser radars are respectively installed at two opposite corners of the AG V, and a scanning angle of the dual laser radars is greater than or equal to 270 degrees.
5. The utility model provides a two laser radar position calibration device of AGV which characterized in that includes:
the system comprises a radar data acquisition module, a calibration module and a calibration module, wherein the radar data acquisition module is used for acquiring double-laser-radar scanning data of an Automatic Guided Vehicle (AGV) provided with double laser radars in a preset calibration environment, the double laser radars are respectively arranged on two opposite sides of the AGV, and scanning areas of the double laser radars have overlapping areas;
the data conversion module is used for converting scanning radar data of a scanning area overlapped by the double laser radars into point cloud data respectively corresponding to the double laser radars;
the position calibration module is used for determining a calibration result of the double laser radars by using an Iterative Closest Point (ICP) algorithm with one of the point cloud data respectively corresponding to the double laser radars as source point cloud data and the other as target point cloud data;
the position calibration device is specifically used for determining a theoretical conversion matrix of the double laser radars according to the installation positions of the double laser radars on the AGV; generating a source point cloud set according to one of the point cloud data respectively corresponding to the double laser radars, and determining a theoretical target point cloud set of the other point cloud data under a source point cloud data coordinate system according to the theoretical conversion matrix; performing iterative computation on the theoretical conversion matrix according to the distance between the average proximity point pair in the source point cloud set and the average proximity point pair in the target point cloud set until the distance between the average proximity point pair in the source point cloud set and the average proximity point pair in the target point cloud set is smaller than a preset distance threshold or the iteration times reach preset iteration times; and taking the product of the target point cloud set and the conversion matrix after iteration as the calibration result.
6. An AGV with dual laser radars, comprising:
the automatic guided vehicle AGV comprises an automatic guided vehicle AGV and double laser radars arranged on two opposite sides of the automatic guided vehicle AGV, wherein scanning areas of the double laser radars are provided with overlapping areas;
the AGV double-laser-radar position calibration device is used for acquiring double-laser-radar scanning data of the AGV in a preset calibration environment; converting scanning radar data of a scanning area overlapped by the double laser radars into point cloud data respectively corresponding to the double laser radars; determining a calibration result of the double laser radars by using an iterative closest point ICP (inductively coupled plasma) algorithm, wherein one of the point cloud data respectively corresponding to the double laser radars is used as source point cloud data, and the other point cloud data is used as target point cloud data;
the determining the calibration result of the dual laser radar by using an ICP algorithm with one of the point cloud data respectively corresponding to the dual laser radar as source point cloud data and the other as target point cloud data comprises the following steps: determining a theoretical conversion matrix of the double laser radars according to the installation positions of the double laser radars on the AGV; generating a source point cloud set according to one of the point cloud data respectively corresponding to the double laser radars, and determining a theoretical target point cloud set of the other point cloud data under a source point cloud data coordinate system according to the theoretical conversion matrix; performing iterative computation on the theoretical conversion matrix according to the distance between the average proximity point pair in the source point cloud set and the average proximity point pair in the target point cloud set until the distance between the average proximity point pair in the source point cloud set and the average proximity point pair in the target point cloud set is smaller than a preset distance threshold or the iteration times reach preset iteration times; and taking the product of the position of the laser radar corresponding to the target point cloud set and the conversion matrix after iteration as the calibration result.
7. The AGV with dual lidar of claim 6, further comprising: and the double-laser radar data correction device is used for correcting the scanning data of the double-laser radar according to the calibration result of the double-laser radar.
8. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a AGV dual-lidar position calibration method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011461196.6A CN112415494B (en) | 2020-12-11 | 2020-12-11 | AGV double-laser-radar position calibration method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011461196.6A CN112415494B (en) | 2020-12-11 | 2020-12-11 | AGV double-laser-radar position calibration method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112415494A CN112415494A (en) | 2021-02-26 |
| CN112415494B true CN112415494B (en) | 2021-08-13 |
Family
ID=74776674
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011461196.6A Active CN112415494B (en) | 2020-12-11 | 2020-12-11 | AGV double-laser-radar position calibration method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112415494B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113109792A (en) * | 2021-03-30 | 2021-07-13 | 深圳市优必选科技股份有限公司 | Laser radar calibration method, laser radar calibration device and intelligent equipment |
| CN116069051B (en) * | 2021-10-29 | 2024-03-19 | 北京三快在线科技有限公司 | Unmanned aerial vehicle control method, device, equipment and readable storage medium |
| CN114194685A (en) * | 2021-12-23 | 2022-03-18 | 山东新华医疗器械股份有限公司 | Stacking AGV control system, method and device |
| CN115390051B (en) * | 2022-10-27 | 2023-03-24 | 深圳煜炜光学科技有限公司 | Laser radar calibration method, device, equipment and storage medium |
| CN115542340B (en) * | 2022-12-01 | 2023-05-05 | 广东工业大学 | Road side double-solid-state laser radar point cloud fusion method based on distance known reference object |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3349042B1 (en) * | 2017-01-11 | 2019-05-08 | Sick Ag | Surveillance sensor and floor-bound vehicle |
| CN108596979A (en) * | 2018-03-27 | 2018-09-28 | 深圳市智能机器人研究院 | A kind of caliberating device and method for laser radar and depth camera |
| CN110221276B (en) * | 2019-05-31 | 2023-09-29 | 文远知行有限公司 | Laser radar calibration method, device, computer equipment and storage medium |
| CN110658530B (en) * | 2019-08-01 | 2024-02-23 | 北京联合大学 | Map construction method and system based on double-laser-radar data fusion and map |
| CN110988849B (en) * | 2019-12-25 | 2022-07-08 | 北京万集科技股份有限公司 | Calibration method and device of radar system, electronic equipment and storage medium |
-
2020
- 2020-12-11 CN CN202011461196.6A patent/CN112415494B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN112415494A (en) | 2021-02-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112415494B (en) | AGV double-laser-radar position calibration method, device, equipment and storage medium | |
| Peršić et al. | Extrinsic 6dof calibration of a radar–lidar–camera system enhanced by radar cross section estimates evaluation | |
| CN109548141B (en) | Indoor environment base station coordinate position calibration method based on Kalman filtering algorithm | |
| US20190278288A1 (en) | Simultaneous localization and mapping methods of mobile robot in motion area | |
| JP6781265B2 (en) | Radar installation judgment using unstructured data | |
| WO2021189468A1 (en) | Attitude correction method, apparatus and system for laser radar | |
| CN112305521B (en) | Double-laser-radar relative position calibration method, device, equipment and storage medium | |
| CN111522022B (en) | Dynamic target detection method of robot based on laser radar | |
| CN111380573B (en) | Method for calibrating the orientation of a moving object sensor | |
| Park et al. | Radar localization and mapping for indoor disaster environments via multi-modal registration to prior LiDAR map | |
| CN111913169B (en) | Laser radar internal reference and point cloud data correction method, device and storage medium | |
| JP2018194562A (en) | Method and device for detecting road boundary | |
| CN112904358B (en) | Laser positioning method based on geometric information | |
| US20220026562A1 (en) | System and method for radar based mapping for autonomous robotic devices | |
| US20240004389A1 (en) | Robot traveling method, and electronic device | |
| CN112285739B (en) | Data processing method, device, equipment and storage medium | |
| Zhou | A closed-form algorithm for the least-squares trilateration problem | |
| Wei et al. | Active self-calibration of hand-mounted laser range finders | |
| CN115856898A (en) | Area array three-dimensional imaging sonar point position reduction method suitable for full sea depth | |
| CN113447908A (en) | GPS Beidou coordinate acquisition method and device for fixed radar detection target | |
| CN115421153B (en) | Laser radar and UWB combined positioning method and system based on extended Kalman filtering | |
| Xu et al. | LiDAR-Link: Observability-Aware Probabilistic Plane-Based Extrinsic Calibration for Non-Overlapping Solid-State LiDARs | |
| WO2022160879A1 (en) | Method and apparatus for determining conversion parameters | |
| KR100929924B1 (en) | Common coordinate generation method using indoor location tracking system | |
| JP7220246B2 (en) | Position detection method, device, equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |