CN114581606A - Bulk door machine cabin based on 3D laser radar and modeling method for materials in cabin - Google Patents
Bulk door machine cabin based on 3D laser radar and modeling method for materials in cabin Download PDFInfo
- Publication number
- CN114581606A CN114581606A CN202210166061.XA CN202210166061A CN114581606A CN 114581606 A CN114581606 A CN 114581606A CN 202210166061 A CN202210166061 A CN 202210166061A CN 114581606 A CN114581606 A CN 114581606A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- radar
- coordinate system
- gantry crane
- initial point
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 58
- 239000000463 material Substances 0.000 title claims abstract description 35
- 230000009466 transformation Effects 0.000 claims abstract description 58
- 239000011159 matrix material Substances 0.000 claims abstract description 46
- 238000001914 filtration Methods 0.000 claims description 55
- 230000008569 process Effects 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 16
- 238000009616 inductively coupled plasma Methods 0.000 claims description 11
- 238000013519 translation Methods 0.000 claims description 9
- 238000012216 screening Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 5
- 230000009471 action Effects 0.000 claims description 4
- 230000005484 gravity Effects 0.000 claims description 4
- 230000001131 transforming effect Effects 0.000 claims description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 3
- 238000005303 weighing Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- 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
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/87—Combinations of systems using electromagnetic waves other than radio waves
-
- 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
- G01S7/481—Constructional features, e.g. arrangements of optical elements
- G01S7/4817—Constructional features, e.g. arrangements of optical elements relating to scanning
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
- G06T7/337—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
- G06T7/593—Depth or shape recovery from multiple images from stereo images
- G06T7/596—Depth or shape recovery from multiple images from stereo images from three or more stereo images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
- G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Networks & Wireless Communication (AREA)
- Remote Sensing (AREA)
- Electromagnetism (AREA)
- Software Systems (AREA)
- Geometry (AREA)
- Computer Graphics (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The invention provides a bulk door machine cabin and a cabin material modeling method based on a 3D laser radar, wherein a first radar is arranged on a revolving body of a door machine, and a second radar with a self-stabilizing device is arranged at the head of a trunk bridge of a boom of the door machine;establishing a plurality of coordinate systems including a gantry crane base coordinate system, a gantry crane revolving body coordinate system, a first radar coordinate system and a second radar coordinate system; determining a coordinate transformation matrix of the two radars under a gantry crane revolving body coordinate system so as to obtain the positions of the two radars on a revolving body; the two radars scan according to the operation instruction to obtain a first initial point cloud and a second initial point cloud and register to obtain a transformation matrix T12And correcting the second initial point cloud, and combining the first initial point cloud and the corrected second initial point cloud to obtain the final point cloud. The invention uses the 3D laser radar with better scanning effect, improves the point cloud registration method, and realizes that the scanning modeling is difficult to be efficiently and accurately carried out by using a simple scanning device.
Description
Technical Field
The invention relates to the field of port bulk cargo door machine automation operation, in particular to a bulk cargo door machine cabin based on a 3D laser radar and a material modeling method in the cabin.
Background
The portal crane is a shore equipment important for port operation, and the working process of the portal crane is to unload bulk cargo (coal, grain and the like) from a cargo ship to a designated area on a dock or carry out reverse loading operation through a grab bucket. Due to the view angle of a driver of the portal crane, the state of materials in a cabin is difficult to master during loading operation, and a camera is generally required to be arranged above a grab bucket to increase the view of the driver; or arranging a finger on the cargo ship, and guiding the driver to operate by the command finger; both of these approaches obviously require manual operation.
With the increasing demand of automatic loading and unloading of the gantry crane, a cargo ship and materials need to be accurately modeled through scanning equipment, so that information such as the position of a cabin hatch and the state of the materials is acquired and transmitted to a control system, and the gantry crane is controlled to perform automatic operation.
The gantry crane mainly structurally comprises a base at the bottom and a revolving body at the upper part. The base part can linearly move along the track, the revolving body part can rotate 360 degrees around the revolving center, and the arm support part is a four-bar mechanism and can radially extend and retract.
In the aspect of scanning and modeling of a bulk cargo ship, the mainstream technical scheme is that a single-line 2D laser radar with a holder is installed at a proper position of a nose bridge head and at a cab of a door machine, the radar rotates for a certain angle along with a holder rotating shaft while scanning a cargo ship operation area, and the rotary scanning of the radar and the movement around the holder form 3D point cloud of an interest area. And combining the two point clouds to form a complete 3D point cloud under different visual angles.
This scanning method has several main disadvantages:
1. the scanning device has a complex structure, consists of a laser radar and a servo driving system, and is inconvenient to maintain in a later period due to multiple mechanical and electrical parts, so that the stability of the system is reduced;
2. the scanning efficiency is low, the resolution of the point cloud on the rotating shaft of the holder is determined by the rotating speed of the driving system, the higher the rotating speed is, the lower the resolution is, in order to obtain higher precision, the rotating speed of the driving system needs to be lower, so that the scanning efficiency is low, and the operation efficiency is influenced;
3. on one hand, the influence is caused by the fact that the radar of the single-line 2D laser radar may vibrate or shake in the scanning process, and the longer the scanning process is, the more obvious the influence is, point clouds obtained by the radar like the nose bridge head are distorted, and the modeling result is influenced; on the other hand, when point cloud merging is performed, only the point cloud registration algorithm with the widest application range at present, namely the iterative closest point algorithm, is usually used, but in the cargo ship operation area scanning modeling process, the positions of all radars are not fixed and unchanged, the point cloud obtained by scanning is unstable in effect and low in overlapping rate, so that the iterative closest point algorithm is not suitable for the cargo ship operation area scanning modeling process, the accuracy of the point cloud obtained by radar scanning is not high when merging and registering is performed, and the modeling result is not accurate enough.
Therefore, currently, there is no universally applicable method that can solve the problem that it is difficult to efficiently and accurately perform scan modeling using a simple scanning device.
Disclosure of Invention
In view of the above, the invention provides a bulk door machine cabin based on a 3D laser radar and a modeling method for materials in the cabin, which are used for solving the problem that a simple scanning device is difficult to efficiently and accurately scan and model.
The technical scheme of the invention is realized as follows:
the invention discloses a 3D laser radar-based bulk door machine cabin and a modeling method for materials in the cabin, wherein the method comprises the following steps:
s1, mounting the first radar on a revolving body of the gantry crane, mounting the second radar with a self-stabilizing device on the head of a trunk bridge of a boom of the gantry crane, wherein the second radar is under the action of gravity, and the scanning direction of the second radar is always kept vertically downward in the operation process of the gantry crane; the first radar and the second radar are both 3D laser radars;
s2, establishing a plurality of coordinate systems including a gantry crane base coordinate system, a gantry crane revolving body coordinate system, a first radar coordinate system and a second radar coordinate system; determining coordinate transformation matrixes of the first radar and the second radar under a coordinate system of a revolving body of the portal crane to obtain the accurate position of the first radar on the revolving body and the estimated position of the second radar on the revolving body, and finishing the preliminary calibration of the first radar and the second radar;
s3, according to the operation instruction, the first radar and the second radar scan the operation area at the same time to obtain a first initial point cloud and a second initial point cloud;
s4, registering the first initial point cloud and the second initial point cloud to obtain a transformation matrix T12And correcting the second initial point cloud, and combining the first initial point cloud and the corrected second initial point cloud to obtain the final point cloud.
According to the method, two 3D laser radars arranged at the revolving body and the trunk bridge of the portal crane are used for rapidly scanning and modeling the cargo ship cabin and bulk cargo materials in the cabin in the portal crane operation area to form 3D point cloud with small dead zone and high precision; the 3D laser radar does not need to be additionally provided with an external auxiliary motion device, has a simple structure, is fast in scanning, can generate point clouds meeting the requirements of resolution and density in a short time, is high in scanning efficiency, and is low in torsion resistance of the obtained point clouds; and point clouds obtained by scanning the two radars are accurately registered, so that a more accurate modeling result can be obtained.
In addition to the above technical means, preferably, in step S1, the attaching the first radar to the revolving body of the door machine specifically includes:
the first radar is arranged at a place where a scanning operation area near the pitching arm frame is not shielded.
On the basis of the above technical solution, preferably, step S2 specifically includes:
s2-1, establishing a plurality of coordinate systems including a gantry crane base coordinate system, a gantry crane revolving body coordinate system, a first radar coordinate system and a second radar coordinate system; the gantry crane base coordinate system takes the rotation center of the gantry crane as an origin, is vertical to the horizontal plane and upwards as a Z axis, is vertical to the wharf shoreline and points to the water side as a Y axis, and determines the X axis direction according to the right-hand rule; the origin of the coordinate system of the gantry crane rotary body, the Z axis and the coordinate system of the base are superposed, the Y axis always points to the direction of the arm support, and the direction of the X axis is determined according to the right-hand rule;
s2-2, only one rotation angle theta is different between the gantry crane revolving body coordinate system and the gantry crane base, the value of the rotation angle theta is obtained through a sensor which is installed on a revolving body and comprises an absolute value encoder, and a transformation matrix of the value of the rotation angle theta is recorded as TWR；
S2-3, the first radar coordinate system and the second radar coordinate system are clearly defined in the product, and the coordinate transformation matrix of the first radar and the second radar under the gantry crane revolving body coordinate system is obtained through preliminary calibration of the first radar and the second radar, so that the accurate position of the first radar on the revolving body and the estimated position of the second radar on the revolving body are obtained.
On the basis of the above technical solution, preferably, step S2-3 specifically includes:
s2-3-1, calibrating external parameters of the first radar by using the inherent object on the wharf, and determining a coordinate transformation matrix T of the first radar relative to a gantry crane rotator coordinate systemRL1＝[R t]；
S2-3-2，Roughly determining a coordinate transformation matrix T of a coordinate system of a second radar and a gantry crane rotary body according to a sensor on the gantry crane comprising an absolute value encoderRL2。
On the basis of the above technical solution, preferably, step S2-3-1 specifically includes:
the method comprises the following steps that a first radar carries out 3D scanning on a wharf scene to obtain calibration point cloud, and features are extracted from the calibration point cloud for calibration;
extracting mutually perpendicular ground and electrical room plane at the lower part of the gantry crane from the calibrated point cloud, determining three rotational degrees of freedom, and obtaining a coordinate transformation matrix TRL1The rotating moiety of (a), noted as R; according to the mechanical size of the gantry crane, selecting a plurality of points with known coordinates on the gantry crane structure from the calibration point cloud, according to the relative relation between the points in the calibration point cloud and the actual scene, calculating an average value P1 of the selected points, obtaining a corresponding point P2 of P1 in the actual scene through rotation and translation transformation, and calculating a translation degree of freedom t by using the average value:
P2＝R*P1+t
obtaining a translation part t of the coordinate transformation matrix; determining a coordinate transformation matrix T of the first radar relative to a gantry crane rotor coordinate systemRL1＝[R t]。
According to the method, the accurate position of the first radar on the revolving body and the estimated position of the second radar on the revolving body are determined, so that the second radar can be conveniently registered and corrected subsequently, the accurate position of the second radar on the revolving body is obtained, and point clouds obtained by scanning the first radar and the second radar can be accurately matched.
On the basis of the above technical solution, preferably, step S3 specifically includes:
the first radar and the second radar scan the working area at the same time to respectively obtain a point cloud, and the two point clouds are converted into a coordinate system of a base of the gantry crane according to a coordinate transformation matrix chain of the first radar and the second radar; let any coordinate in two point clouds be PL1、PL2The coordinates of the two are respectively P under the coordinate system of the gantry crane baseW1、PW2Then, there are:
PW1＝TWR*TRL1*PL1
PW2＝TWR*TRL2*PL2
two point clouds P under base coordinate system of door weighing machineW1And PW2The first initial point cloud and the second initial point cloud are respectively.
According to the method, the first radar and the second radar respectively scan the operation area to obtain the point cloud, and preparation is made for subsequent modeling.
On the basis of the above technical solution, preferably, step S4 specifically includes:
s4-1, preprocessing the first initial point cloud and the second initial point cloud, wherein the preprocessing comprises removing irrelevant point clouds from the first initial point cloud and the second initial point cloud in a direct filtering mode, carrying out radius filtering, and removing noise points to obtain two corresponding filtered point clouds; continuing to execute step S4-2;
s4-2, performing down-sampling processing on the two filtered point clouds through voxel filtering to obtain a first sparse point cloud and a second sparse point cloud which correspond to the point clouds and have proper scale; continuing to execute step S4-3;
s4-3, carrying out coarse registration on the first sparse point cloud and the second sparse point cloud; the rough registration comprises the steps of calculating the FPFH characteristics of two point clouds by taking the first sparse point cloud as a target point cloud and the second sparse point cloud as a source point cloud according to a registration algorithm module provided by a PCL (personal computer) library, setting parameters, and performing registration calculation on the two point clouds to obtain a coordinate transformation matrix T from the second sparse point cloud to the first sparse point cloud1(ii) a And according to T1Performing coordinate transformation on the second sparse point cloud to obtain an updated second initial point cloud; continuing to execute step S4-4;
s4-4, performing straight-through filtering and radius filtering processing on the first sparse point cloud and the updated second initial point cloud again, modifying parameters during straight-through filtering, reducing the range of the X direction, and canceling the relaxation of a threshold value; continuing to execute step S4-5;
s4-5, performing downsampling processing on the two-point cloud, and setting the side length of a filter to be lfin(ii) a Step of continuing executionStep S4-6;
s4-6, carrying out ICP fine registration on the two point clouds; the fine registration comprises the steps of setting parameters by taking the first sparse point cloud as a target point cloud and the updated second initial point cloud as a source point cloud, and carrying out ICP (inductively coupled plasma) registration calculation to obtain a coordinate transformation matrix T2(ii) a And according to T2Performing coordinate transformation on the updated second initial point cloud to obtain a second initial point cloud under a final gantry crane base coordinate system; continuing to execute step S4-7;
s4-7, transforming the matrix T according to the coordinates1And T2Obtaining a transformation matrix T12＝T1*T2And then correcting the second initial point cloud under the portal crane base coordinate system: p isW1＝T12*PW2Obtaining a corrected second initial point cloud; and merging the first initial point cloud and the corrected second initial point cloud to obtain a final point cloud.
According to the method, the operation area point clouds, namely the first initial point cloud and the second initial point cloud, which are respectively obtained by the first radar and the second radar are processed and accurately registered, so that a more accurate modeling result can be obtained; the position of the first radar is fixed, and the position of the second radar is dynamically changed, so that the second initial point cloud is registered and corrected, the accurate position of the second radar on the revolving body is determined, the second initial point cloud under the base coordinate system of the gantry crane is obtained, and the merging with the first initial point cloud is facilitated.
On the basis of the above technical solution, preferably, step S4-1 specifically includes:
setting the length of the initial filter edge to be lmaxThe expected point cloud size is countminPerforming through filtering processing;
setting the Y value of the wharf coastline as Y in a gantry crane coordinate system for the first initial point cloudlScreening out-50.0<x<50.0 and yl<y<ylA first initial point cloud in the range of + 50.0;
setting the Y value of the wharf coastline as Y in the gantry crane coordinate system for the second initial point cloudlRelaxing the filtering threshold by delta1Screening out-50.0-delta.3.01<x<50.0+δ1And y isl-δ1<y<yl+50.0+δ1A second initial point cloud within the range;
and performing radius filtering on the first initial point cloud and the second initial point cloud which are subjected to the straight-through filtering, and removing noise points to obtain two corresponding filtering point clouds.
According to the method, the first initial point cloud and the second initial point cloud are preprocessed to remove irrelevant point clouds, and when the second initial point cloud is subjected to the first direct filtering processing, the coordinate transformation matrix T of the second radar is usedRL2It is not an exact value and therefore some relaxation is made to the filtering threshold.
On the basis of the above technical solution, preferably, step S4-2 specifically includes:
for the first initial point cloud, length of side lcur＝lmaxVoxel filtering is carried out on the point cloud, and the scale of the obtained point cloud is countcurThe following determination is performed:
i. if countcur≥countminIf the point cloud is dense, the voxel filtering process is ended, and the final point cloud scale is countcur；
And ii, vice versa, the current point cloud is over sparse, and the side length is halved, namely lcur＝lcur2.0, carrying out voxel filtering again;
if result countcur<countminContinuing to execute the step ii, and repeating the step ii until the voxel filtering process is finished; otherwise, the existence of a proper filter side length l epsilon (l) is showncur～2lcur) Voxel filtering is carried out according to the side length to obtain the size of countminThe first sparse point cloud is recorded with the filtering side length of lfin；
Setting the length of the filter edge to be l for the second sparse point cloudfinAnd carrying out voxel filtering to obtain a second sparse point cloud.
According to the method, two point clouds with basically consistent uniformity and moderate density are obtained, and subsequent coarse registration is facilitated.
In a second aspect of the present invention, a computer-readable storage medium is disclosed, wherein a 3D laser radar-based modeling method program for a material in a ship cabin of a bulk door machine and a cabin is stored on the storage medium, and when the 3D laser radar-based modeling method program for a material in a ship cabin of a bulk door machine and a cabin is executed, the 3D laser radar-based modeling method for a material in a ship cabin of a bulk door machine and a cabin is implemented according to the first aspect of the present invention.
Compared with the prior art, the bulk door machine cabin and the cabin material modeling method based on the 3D laser radar have the following beneficial effects:
(1) two 3D laser radars arranged at the revolving body and the trunk bridge of the gantry crane are used for quickly scanning and modeling the cargo ship cabin and bulk cargo materials in the cabin in the operation area of the gantry crane to form 3D point cloud with small dead zone and high precision; the 3D laser radar does not need to be additionally provided with an external auxiliary motion device, has a simple structure, can generate point clouds meeting the requirements of resolution and density in a short time, and has high scanning efficiency;
(2) respectively determining coordinate transformation matrixes of the fixed first radar and the second radar with dynamically changed positions relative to a gantry crane revolving body coordinate system, and determining the coordinate transformation matrixes of the first radar and the second radar under the gantry crane revolving body coordinate system, so as to determine the relative positions of the two radars under a gantry crane base coordinate system in point clouds obtained by scanning an operation area;
(3) and accurately registering the point clouds, namely a first initial point cloud and a second initial point cloud, obtained by scanning the operation area by two radars, wherein the second initial point cloud with dynamically changed position is registered and corrected according to the first initial point cloud, the accurate position of the second radar on the revolving body is determined, the second initial point cloud under the base coordinate system of the gantry crane is obtained, and finally the second initial point cloud and the first initial point cloud under the base coordinate system of the gantry crane are combined, so that a more accurate modeling result can be obtained.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of the operation of a bulk door machine cabin and a cabin material modeling method based on a 3D laser radar;
FIG. 2 is a schematic diagram of a door machine structure and a radar installation position in the bulk cargo door machine cabin and cabin interior material modeling method based on a 3D laser radar of the invention;
FIG. 3 is a schematic diagram of a coordinate system of a door machine in a bulk door machine cabin and in-cabin material modeling method based on a 3D laser radar of the invention;
FIG 4 is a point cloud processing flow chart in the modeling method of the bulk cargo door machine cabin and the materials in the cabin based on the 3D laser radar;
FIG. 5 is a flow chart of a registration process in a bulk door machine cabin and in-cabin material modeling method based on a 3D laser radar.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
Examples
The working flow of the bulk cargo door machine cabin and the material modeling method in the cabin based on the 3D laser radar is shown in the figure 1, and the processing steps are explained as follows:
the method comprises the following steps that firstly, a first radar is arranged on a revolving body of a gantry crane, a second radar with a self-stabilizing device is arranged at the head of a trunk bridge of a boom of the gantry crane, the second radar is under the action of gravity, and the scanning direction of the second radar is always kept vertically downward under the action of gravity in the operation process of the gantry crane; the first radar and the second radar are both 3D laser radars. And turning to the second step.
On the basis of the above scheme, preferably, the first radar is installed in a place where a scanning operation area near the luffing jib is not blocked.
Secondly, establishing a plurality of coordinate systems including a gantry crane base coordinate system (a global coordinate system), a gantry crane revolving body coordinate system, a first radar coordinate system and a second radar coordinate system; and determining coordinate transformation matrixes of the first radar and the second radar under a gantry crane revolving body coordinate system to obtain the accurate position of the first radar on the revolving body and the estimated position of the second radar on the revolving body, and finishing the initial calibration of the first radar and the second radar. And (6) turning to the third step.
On the basis of the above scheme, preferably, the second step specifically includes:
s2-1, establishing a plurality of coordinate systems including a gantry crane base coordinate system, a gantry crane revolving body coordinate system, a first radar coordinate system and a second radar coordinate system; the gantry crane base coordinate system takes the rotation center of the gantry crane as an origin, is vertical to the horizontal plane and upwards as a Z axis, is vertical to the wharf shoreline and points to the water side as a Y axis, and determines the X axis direction according to the right-hand rule; the origin of the gantry crane revolving body coordinate system, the Z axis and the base coordinate system are superposed, the Y axis always points to the direction of the arm support, and the X axis direction is determined according to the right hand rule;
s2-2, only one rotation angle theta is different between the gantry crane revolving body coordinate system and the gantry crane base, the value of the rotation angle theta is obtained through a sensor which is installed on a revolving body and comprises an absolute value encoder, and a transformation matrix of the value of the rotation angle theta is recorded as TWR；
S2-3, the first radar coordinate system and the second radar coordinate system are clearly defined in the product, and the accurate positions of the first radar and the second radar on the revolving body are unknown, so that a coordinate transformation matrix of the two radars under the revolving body coordinate system needs to be determined in a proper way; and obtaining a coordinate transformation matrix for determining the first radar and the second radar under a coordinate system of a revolving body of the portal crane by preliminarily calibrating the first radar and the second radar so as to obtain the accurate position of the first radar on the revolving body and the estimated position of the second radar on the revolving body.
On the basis of the above scheme, preferably, step S2-3 specifically includes:
s2-3-1, limited by the wharf field condition, the traditional method of calibrating by calibration objects such as calibration plates is not feasible, so the external reference of the first radar is calibrated by using the inherent object on the wharf, and the coordinate transformation matrix T of the first radar relative to the gantry crane revolving body coordinate system is determinedRL1＝[R t]；
S2-3-2, the second radar is installed at the head of the trunk of the; roughly determining a coordinate transformation matrix T of a coordinate system of a second radar and a gantry crane revolving body according to a sensor comprising an absolute value encoder on the gantry craneRL2。
On the basis of the above scheme, preferably, step S2-3-1 specifically includes:
the method comprises the steps that a first radar carries out 3D scanning on a wharf scene to obtain calibration point cloud, and features are extracted from the calibration point cloud for calibration;
extracting mutually vertical ground and electric room plane under the gantry crane from the calibration point cloud, determining three rotational degrees of freedom, and obtaining a coordinate transformation matrix TRL1The rotating part of (a), denoted as R; according to the mechanical size of the gantry crane, selecting a plurality of points with known coordinates on the gantry crane structure from the calibration point cloud, according to the relative relation between the points in the calibration point cloud and the actual scene, calculating an average value P1 of the selected points, obtaining a corresponding point P2 of P1 in the actual scene through rotation and translation transformation, and calculating a translation degree of freedom t by using the average value:
P2＝R*P1+t
thereby obtaining a translation portion t of the coordinate transformation matrix; determining a coordinate transformation matrix T of the first radar relative to a gantry crane rotor coordinate systemRL1＝[R t]。
And thirdly, scanning the operation area by the first radar and the second radar simultaneously according to the operation instruction to obtain a first initial point cloud and a second initial point cloud. And turning to the fourth step.
On the basis of the above scheme, preferably, the third step specifically includes:
the first radar and the second radar scan the working area at the same time to respectively obtain a point cloud, and the two point clouds are converted into a global coordinate system of the gantry crane according to a coordinate transformation matrix chain of the first radar and the second radar; let any coordinate in two point clouds be PL1、PL2The coordinates of the two are respectively P under the global coordinate system of the gantry craneW1、PW2Then, there are:
PW1＝TWR*TRL1*PL1
PW2＝TWR*TRL2*PL2
weighing two point clouds P under global coordinate systemW1And PW2The first initial point cloud and the second initial point cloud are respectively.
Due to TRL2The two point clouds are not accurate values, the two point clouds after transformation can not be directly merged, and the first initial point cloud and the second initial point cloud need to be registered to obtain a transformation matrix T between the first initial point cloud and the second initial point cloud12。
Fourthly, registering the first initial point cloud and the second initial point cloud to obtain a transformation matrix T12And correcting the second initial point cloud to obtain the accurate position of the second radar on the revolving body, and combining the first initial point cloud and the corrected second initial point cloud to obtain the final point cloud.
On the basis of the above scheme, preferably, the fourth step specifically includes:
s4-1, preprocessing the first initial point cloud and the second initial point cloud, wherein the preprocessing comprises removing irrelevant point clouds of the first initial point cloud and the second initial point cloud in a direct filtering mode, carrying out radius filtering, and removing noise points to obtain two corresponding filtering point clouds; continuing to execute step S4-2;
s4-2, in order to accelerate the processing speed of the subsequent algorithm and improve the consistency of the geometrical distribution of the point clouds, performing down-sampling processing on the two filtered point clouds through voxel filtering to obtain a first sparse point cloud and a second sparse point cloud which correspond to the point clouds and have proper scale; continuing to execute step S4-3;
s4-3, carrying out coarse registration on the first sparse point cloud and the second sparse point cloud processed in the step S4-2; the rough registration comprises the steps of calculating FPFH (field-programmable gate hopping) characteristics of two point clouds by taking the first sparse point cloud as a target point cloud and the second sparse point cloud as a source point cloud according to a registration algorithm module provided by a PCL (personal computer) library, setting parameters, and performing registration calculation on the two point clouds to obtain a coordinate transformation matrix T from the second sparse point cloud to the first sparse point cloud1(ii) a And according to T1Performing coordinate transformation on the second sparse point cloud to obtain an updated second initial point cloud; continuing to execute step S4-4;
s4-4, performing straight-through filtering and radius filtering processing on the first sparse point cloud and the updated second initial point cloud again, modifying parameters during straight-through filtering, reducing the range in the X direction, canceling the relaxation of a threshold value, and screening out-25.0 of the first sparse point cloud and the updated second initial point cloud<x<25.0 and yl<y<yl+50.0 range point cloud; continuing to execute step S4-5;
s4-5, performing down-sampling processing on the two-point cloud, and setting the side length of the filter to be lfin(ii) a Continuing to execute step S4-6;
s4-6, carrying out ICP fine registration on the two point clouds; the fine registration comprises the steps of setting parameters by taking the first sparse point cloud as a target point cloud and the updated second initial point cloud as a source point cloud, and carrying out ICP (inductively coupled plasma) registration calculation to obtain a coordinate transformation matrix T2(ii) a And according to T2Performing coordinate transformation on the updated second initial point cloud to obtain a second initial point cloud under a final global coordinate system; continuing to execute step S4-7;
s4-7, transforming the matrix T according to the coordinates1And T2Obtaining a transformation matrix T12＝T1*T2And then correcting the second initial point cloud under the global coordinate system: pW1＝T12*PW2Obtaining a corrected second initial valuePoint cloud; and merging the first initial point cloud and the corrected second initial point cloud to obtain a final point cloud.
On the basis of the above scheme, preferably, step S4-1 specifically includes:
setting the length of the initial filter edge to be lmaxThe expected point cloud size is countminPerforming through filtering processing;
setting the Y value of the wharf coastline as Y in a gantry crane coordinate system for the first initial point cloudlThe Y value is determined, and screening is carried out to obtain-50.0<x<50.0 and yl<y<ylA first initial point cloud in the range of + 50.0;
setting the Y value of the wharf coastline as Y in the gantry crane coordinate system for the second initial point cloudlThe value of Y is determined, and a coordinate transformation matrix T of the second radarRL2Not the exact value, so the filter threshold is relaxed by delta1Screening out-50.0-delta.3.01<x<50.0+δ1And y isl-δ1<y<yl+50.0+δ1A second initial point cloud within the range;
and performing radius filtering on the first initial point cloud and the second initial point cloud which are subjected to the straight-through filtering, and removing noise points to obtain two corresponding filtered point clouds.
On the basis of the above scheme, preferably, step S4-2 specifically includes:
for the first initial point cloud, length of side lcur＝lmaxVoxel filtering is carried out on the point cloud, and the scale of the obtained point cloud is countcurThe following determination is performed:
i. if countcur≥countminIf the point cloud is dense, the voxel filtering process is finished, and the final point cloud scale is countcur；
And ii, vice versa, the current point cloud is over sparse, and the side length is halved, namely lcur＝lcur2.0, carrying out voxel filtering again;
if result countcur<countminContinuing to perform step ii, and repeating so until voxel filteringThe process is ended; otherwise, the existence of a proper filter side length l epsilon (l) is showncur～2lcur) Voxel filtering is carried out according to the side length to obtain the size of countminA first sparse point cloud; specifically, the length of the filter side is finally determined by using the dichotomy thought and continuously reducing the interval through loop iteration, and the length of the filter side is recorded as lfinObtaining a first sparse point cloud;
setting the length of the filter edge to be l for the second sparse point cloudfinAnd carrying out voxel filtering to obtain a second sparse point cloud.
Through the steps, the two sparse point clouds are basically consistent in uniformity degree and moderate in density.
The method adopts the 3D laser radar, can realize rapid modeling of the scene, and has high efficiency, low point cloud distortion and high trueness; an initial transformation matrix of the radar at the trunk like nose is obtained by adopting sensors such as an encoder on the equipment, and the accurate position of the second radar is determined through a coarse registration process and an ICP (inductively coupled plasma) fine registration process, so that the success probability of a registration algorithm is improved; on the other hand, the precision of the registration result is improved; the finally formed point cloud has small dead zone and high quality, and provides powerful guarantee for subsequent algorithm input.
The invention also discloses a computer readable storage medium, wherein a 3D laser radar-based bulk door machine cabin and in-cabin material modeling method program is stored on the storage medium, and when the 3D laser radar-based bulk door machine cabin and in-cabin material modeling method program is executed, the 3D laser radar-based bulk door machine cabin and in-cabin material modeling method is realized.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (10)
1. A bulk door machine cabin and an in-cabin material modeling method based on a 3D laser radar are characterized by comprising the following steps:
s1, mounting the first radar on a revolving body of the gantry crane, mounting the second radar with a self-stabilizing device on the head of a trunk bridge of a boom of the gantry crane, wherein the second radar is under the action of gravity, and the scanning direction of the second radar is always kept vertically downward in the operation process of the gantry crane; the first radar and the second radar are both 3D laser radars;
s2, establishing a plurality of coordinate systems including a gantry crane base coordinate system, a gantry crane revolving body coordinate system, a first radar coordinate system and a second radar coordinate system; determining coordinate transformation matrixes of the first radar and the second radar under a gantry crane revolving body coordinate system to obtain the accurate position of the first radar on a revolving body and the estimated position of the second radar on the revolving body, and finishing the preliminary calibration of the first radar and the second radar;
s3, according to the operation instruction, the first radar and the second radar scan the operation area at the same time to obtain a first initial point cloud and a second initial point cloud;
s4, registering the first initial point cloud and the second initial point cloud to obtain a transformation matrix T12And correcting the second initial point cloud, and combining the first initial point cloud and the corrected second initial point cloud to obtain the final point cloud.
2. The method for modeling the materials in the cabin and the bulk cargo door based on the 3D laser radar of claim 1, wherein the step S1 of mounting the first radar on the revolving body of the door machine specifically comprises:
the first radar is arranged at a place where a scanning operation area near the pitching arm frame is not shielded.
3. The 3D lidar-based bulk door machine cabin and cabin interior material modeling method according to claim 1, wherein the step S2 specifically comprises:
s2-1, establishing a plurality of coordinate systems including a gantry crane base coordinate system, a gantry crane revolving body coordinate system, a first radar coordinate system and a second radar coordinate system; the gantry crane base coordinate system takes the rotation center of the gantry crane as an origin, is vertical to the horizontal plane and upwards as a Z axis, is vertical to the wharf shoreline and points to the water side as a Y axis, and determines the X axis direction according to the right-hand rule; the origin of the gantry crane revolving body coordinate system, the Z axis and the base coordinate system are superposed, the Y axis always points to the direction of the arm support, and the X axis direction is determined according to the right hand rule;
s2-2, only one rotation angle theta is different between the gantry crane revolving body coordinate system and the gantry crane base, the value of the rotation angle theta is obtained through a sensor which is installed on a revolving body and comprises an absolute value encoder, and a transformation matrix of the value of the rotation angle theta is recorded as TWR；
S2-3, the first radar coordinate system and the second radar coordinate system are clearly defined in the product, and the coordinate transformation matrix of the first radar and the second radar under the revolving body coordinate system of the gantry crane is obtained by preliminarily calibrating the first radar and the second radar so as to obtain the accurate positions of the first radar and the second radar on the revolving body and the estimated position of the second radar on the revolving body.
4. The 3D lidar-based bulk door machine cabin and cabin interior material modeling method according to claim 3, wherein the step S2-3 specifically comprises:
s2-3-1, calibrating external parameters of the first radar by using the inherent object on the wharf, and determining a coordinate transformation matrix T of the first radar relative to a gantry crane rotator coordinate systemRL1＝[R t]；
S2-3-2, roughly determining a coordinate transformation matrix T of a coordinate system of a second radar and a gantry crane revolving body according to a sensor comprising an absolute value encoder on the gantry craneRL2。
5. The 3D lidar-based bulk door machine cabin and cabin interior material modeling method according to claim 4, wherein the step S2-3-1 specifically comprises:
the method comprises the following steps that a first radar carries out 3D scanning on a wharf scene to obtain calibration point cloud, and features are extracted from the calibration point cloud for calibration;
extracting mutually vertical ground and lower part of gantry crane in calibration point cloudDetermining three rotational degrees of freedom to obtain coordinate transformation matrix TRL1The rotating part of (a), denoted as R; according to the mechanical size of the gantry crane, selecting a plurality of points with known coordinates on the gantry crane structure from the calibration point cloud, according to the relative relation between the points in the calibration point cloud and the actual scene, calculating an average value P1 of the selected points, obtaining a corresponding point P2 of P1 in the actual scene through rotation and translation transformation, and calculating a translation degree of freedom t by using the average value:
P2＝R*P1+t
obtaining a translation part t of the coordinate transformation matrix; determining a coordinate transformation matrix T of the first radar relative to a gantry crane rotor coordinate systemRL1＝[R t]。
6. The method for modeling the materials in the hold and the cabin of the bulk door machine based on the 3D laser radar according to claim 4, wherein the step S3 specifically comprises:
the first radar and the second radar scan the operation area at the same time to respectively obtain a point cloud, and the two point clouds are converted to a coordinate system of a base of the gantry crane according to a coordinate transformation matrix chain of the first radar and the second radar; let any coordinate in two point clouds be PL1、PL2The coordinates of the two are respectively P under the coordinate system of the gantry crane baseW1、PW2Then, there are:
PW1＝TWR*TRL1*PL1
PW2＝TWR*TRL2*PL2
two point clouds P under base coordinate system of door weighing machineW1And PW2The first initial point cloud and the second initial point cloud are respectively.
7. The method for modeling the materials in the hold and the cabin of the bulk door machine based on the 3D laser radar according to claim 6, wherein the step S4 specifically comprises:
s4-1, preprocessing the first initial point cloud and the second initial point cloud, wherein the preprocessing comprises removing irrelevant point clouds of the first initial point cloud and the second initial point cloud in a direct filtering mode, carrying out radius filtering, and removing noise points to obtain two corresponding filtering point clouds; continuing to execute step S4-2;
s4-2, performing down-sampling processing on the two filtered point clouds through voxel filtering to obtain a first sparse point cloud and a second sparse point cloud which correspond to the point clouds and are proper in scale; continuing to execute step S4-3;
s4-3, carrying out coarse registration on the first sparse point cloud and the second sparse point cloud; the rough registration comprises the steps of calculating the FPFH characteristics of two point clouds by taking the first sparse point cloud as a target point cloud and the second sparse point cloud as a source point cloud according to a registration algorithm module provided by a PCL (personal computer) library, setting parameters, and performing registration calculation on the two point clouds to obtain a coordinate transformation matrix T from the second sparse point cloud to the first sparse point cloud1(ii) a And according to T1Performing coordinate transformation on the second sparse point cloud to obtain an updated second initial point cloud; continuing to execute step S4-4;
s4-4, performing straight-through filtering and radius filtering processing on the first sparse point cloud and the updated second initial point cloud again, modifying parameters during straight-through filtering, reducing the range of the X direction, and canceling the relaxation of a threshold value; continuing to execute step S4-5;
s4-5, performing down-sampling processing on the two-point cloud, and setting the side length of a filter to be lfin(ii) a Continuing to execute step S4-6;
s4-6, carrying out ICP fine registration on the two point clouds; the coarse registration comprises the steps of setting parameters by taking the first sparse point cloud as a target point cloud and the updated second initial point cloud as a source point cloud, and carrying out ICP (inductively coupled plasma) registration calculation to obtain a coordinate transformation matrix T2(ii) a And according to T2Performing coordinate transformation on the updated second initial point cloud to obtain a second initial point cloud under a final gantry crane base coordinate system; continuing to execute step S4-7;
s4-7, transforming the matrix T according to the coordinates1And T2Obtaining a transformation matrix T12＝T1*T2And then correcting the second initial point cloud under the portal crane base coordinate system: pW1＝T12*PW2Obtaining a corrected second initial point cloud; correcting the first initial point cloudAnd merging the second initial point clouds to obtain a final point cloud.
8. The 3D lidar-based bulk door machine cabin and cabin interior material modeling method according to claim 7, wherein the step S4-1 specifically comprises:
setting the length of the initial filter edge as lmaxThe expected point cloud size is countminPerforming through filtering processing;
setting the Y value of the wharf coastline as Y in a gantry crane coordinate system for the first initial point cloudlScreening out x is more than-50.0 and less than 50.0 and yl＜y＜ylA first initial point cloud in the range of + 50.0;
setting the Y value of the wharf coastline as Y in the portal crane coordinate system for the second initial point cloudlRelaxing the filtering threshold by delta1Screening out-50.0-delta.3.01＜x＜50.0+δ1And y isl-δ1＜y＜yl+50.0+δ1A second initial point cloud within the range;
and performing radius filtering on the first initial point cloud and the second initial point cloud which are subjected to the straight-through filtering, and removing noise points to obtain two corresponding filtering point clouds.
9. The 3D lidar-based bulk door machine cabin and cabin interior material modeling method according to claim 7, wherein the step S4-2 specifically comprises:
for the first initial point cloud, length of side lcur＝lmaxVoxel filtering is carried out on the point cloud, and the obtained point cloud scale is countcurThe following determination is performed:
i. if countcur≥countminIf the point cloud is dense, the voxel filtering process is ended, and the final point cloud scale is countcur；
And ii, vice versa, the current point cloud is over sparse, and the side length is halved, namely lcur＝lcur2.0, carrying out voxel filtering again;
if result countcur＜countminContinuing to execute the step ii, and repeating the step ii until the voxel filtering process is finished; otherwise, the condition that a proper filtering side length is in the range of l E (l)cur～2lcur) Voxel filtering is carried out according to the side length to obtain the size of countminThe first sparse point cloud is recorded with the filtering side length of lfin；
Setting the length of the filter edge to be l for the second sparse point cloudfinAnd carrying out voxel filtering to obtain a second sparse point cloud.
10. A computer-readable storage medium, wherein a 3D lidar based bulk door machine cabin and in-cabin material modeling method program is stored on the storage medium, and when executed, implements a 3D lidar based bulk door machine cabin and in-cabin material modeling method of any of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210166061.XA CN114581606A (en) | 2022-02-23 | 2022-02-23 | Bulk door machine cabin based on 3D laser radar and modeling method for materials in cabin |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210166061.XA CN114581606A (en) | 2022-02-23 | 2022-02-23 | Bulk door machine cabin based on 3D laser radar and modeling method for materials in cabin |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114581606A true CN114581606A (en) | 2022-06-03 |
Family
ID=81773024
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210166061.XA Pending CN114581606A (en) | 2022-02-23 | 2022-02-23 | Bulk door machine cabin based on 3D laser radar and modeling method for materials in cabin |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114581606A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115146745A (en) * | 2022-09-01 | 2022-10-04 | 深圳市城市公共安全技术研究院有限公司 | Method, device and equipment for correcting point cloud data coordinate point positions and storage medium |
| CN116817904A (en) * | 2023-08-29 | 2023-09-29 | 深圳市镭神智能系统有限公司 | Door machine detecting system |
-
2022
- 2022-02-23 CN CN202210166061.XA patent/CN114581606A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115146745A (en) * | 2022-09-01 | 2022-10-04 | 深圳市城市公共安全技术研究院有限公司 | Method, device and equipment for correcting point cloud data coordinate point positions and storage medium |
| CN115146745B (en) * | 2022-09-01 | 2022-12-02 | 深圳市城市公共安全技术研究院有限公司 | Method, device and equipment for correcting point cloud data coordinate point positions and storage medium |
| CN116817904A (en) * | 2023-08-29 | 2023-09-29 | 深圳市镭神智能系统有限公司 | Door machine detecting system |
| CN116817904B (en) * | 2023-08-29 | 2023-11-10 | 深圳市镭神智能系统有限公司 | Door machine detecting system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114581606A (en) | Bulk door machine cabin based on 3D laser radar and modeling method for materials in cabin | |
| AU2010265789B2 (en) | Autonomous loading | |
| Roman et al. | Application of structured light imaging for high resolution mapping of underwater archaeological sites | |
| CN112363153B (en) | Material pile edge detection method and system | |
| CN108564525A (en) | A kind of 3D point cloud 2Dization data processing method based on multi-line laser radar | |
| CN113748357A (en) | Attitude correction method, device and system of laser radar | |
| CN110741282A (en) | External parameter calibration method and device, computing equipment and computer storage medium | |
| CN113063368A (en) | Linear laser rotary scanning three-dimensional profile measuring method and device | |
| WO2021196938A1 (en) | Automatic container loading and unloading apparatus and method | |
| CN109987519A (en) | A kind of grab bucket ship unloader carries out the method, apparatus and system of ship-discharging operation | |
| JP2020126363A (en) | Image processing system, image processing method, generation method of learnt model, and data set for leaning | |
| CN105427301B (en) | Based on DC component than the extra large land clutter Scene Segmentation estimated | |
| Hurtós et al. | A novel blending technique for two-dimensional forward-looking sonar mosaicing | |
| US20230348237A1 (en) | Mapping of a Crane Spreader and a Crane Spreader Target | |
| CN113932732A (en) | Full-characteristic detection equipment and method for open type freight vehicle | |
| CN116359929A (en) | Cabin positioning and scanning identification method, system and storage medium for cabin materials | |
| CN113483664A (en) | Screen plate automatic feeding system and method based on line structured light vision | |
| CN110329910A (en) | Automatic cabinet gantry crane remote control trolley path planning method | |
| CN115797563A (en) | Whole ship modeling method with cooperation of multiple gate machines | |
| CN113538566B (en) | Cargo ship hatch position acquisition method and system based on laser radar | |
| CN116642468A (en) | Unmanned aerial vehicle aerial photography and unmanned ship based underwater integrated scanning method | |
| CN202048897U (en) | Laser three-dimensional bulk cargo imaging device for ship unloader | |
| CN114791596A (en) | Method and system for calibrating external parameters of waterborne multi-line laser radar | |
| CN114942421A (en) | Omnidirectional scanning multiline laser radar autonomous positioning device and method | |
| CN206266098U (en) | A kind of control device of high pedestal jib crane |
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 |