CN115848878B - AGV-based tobacco frame identification and stacking method and system - Google Patents
AGV-based tobacco frame identification and stacking method and system Download PDFInfo
- Publication number
- CN115848878B CN115848878B CN202310173761.6A CN202310173761A CN115848878B CN 115848878 B CN115848878 B CN 115848878B CN 202310173761 A CN202310173761 A CN 202310173761A CN 115848878 B CN115848878 B CN 115848878B
- Authority
- CN
- China
- Prior art keywords
- agv
- stacking
- frame
- cigarette
- frames
- 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
Landscapes
- Warehouses Or Storage Devices (AREA)
- Manipulator (AREA)
Abstract
The invention provides a smoke frame identification and stacking method and system based on an AGV, wherein the system comprises the following steps: AGV, level and forward install in the middle of the AGV level visual angle shoot stack position cigarette frame first vision sensor, install in AGV fork side and look down perpendicularly shoot the second vision sensor that is located the cigarette frame that AGV fork below is in the cigarette frame and stacks the position, install and be used for detecting AGV fork inclination's inclination sensor in AGV fork side to and receive cigarette frame image data and AGV fork inclination data and carry out the visual identification controller that the cigarette frame stacks the position calculation and obtain the relative position appearance of cigarette frame. According to the invention, the AGV is designed according to the characteristics of the tobacco frames, so that the accurate stacking of the tobacco frames is realized, and the stacking error is less than +/-20 mm.
Description
Technical Field
The invention relates to the technical field of cigarette frame stacking, in particular to a method and a system for identifying and stacking cigarette frames based on an AGV.
Background
In the plane storage field, as no goods shelf is adopted, the ground storage of goods has high requirement on space, the occupied area is large, the warehouse utilization rate is low, and the storage cost is higher and higher, but the planned storage of the warehouse can be caused to be inflexible and changeable by adopting the goods shelf, and a plurality of plane type warehouses belong to the leasing property, or the floor height is insufficient, and the foundation construction such as the ground does not meet the scheme of adopting the goods shelf. More and more customers are proposing to complete multi-tier stacked storage of goods by AGVs, for example, a great deal of application needs have emerged in the tobacco, dairy, beer, gypsum, etc. industries.
Disclosure of Invention
In order to solve the problem that when the tobacco frames are stacked in multiple layers in the existing tobacco industry, the precision and the efficiency of a manual stacking method are low, the invention provides a tobacco frame identification and stacking method and system based on an AGV, which changes the existing manual stacking into automatic completion by the AGV, improves the efficiency of stacking multiple layers of the tobacco frames, improves the stacking precision and avoids the tobacco frames from sliding down.
The specific scheme is as follows:
the invention provides a smoke frame identification and stacking method based on an AGV, which comprises the following steps:
s1: the method comprises the steps of installing a first vision sensor horizontally and positively in the middle of an AGV, shooting a cigarette frame at a stacking position in a horizontal view angle, installing a second vision sensor on the side face of a fork of the AGV, and shooting the cigarette frame at the stacking position vertically downwards after the AGV runs to the stacking position; calibrating installation parameters among the first visual sensor, the second visual sensor and the AGV and transmitting the installation parameters to an AGV controller;
s2: installing an inclination sensor on the side surface of an AGV fork to detect the inclination angle of the AGV fork;
s3: planning a traveling path of an AGV stacking smoke frame as a planning path, suspending traveling at a front point platform of the planned path when the AGV travels, and sending a horizontal visual angle image acquisition command to a visual identification controller through the AGV controller, wherein the visual identification controller sends the acquisition command to a first visual sensor so that the first visual sensor acquires smoke frame image data of a stacking position and sends the smoke frame image data to the visual identification controller; the front point platform is a middle point arranged in an AGV planning path, and the AGV at the position can shoot cigarette frames at a stacking position in a horizontal view angle by using a first visual sensor;
s4: the vision recognition controller filters the smoke frame image data acquired in the step S3, extracts characteristic data describing the stacking position of the smoke frames, calculates the placement angle of the smoke frames at the stacking position and the distance between the smoke frames and a front point platform where the AGV is located by using the characteristic data as a relative pose, and transmits the pose to the AGV controller;
s5: the AGV controller calculates a temporary dynamic path of the AGV reaching a tobacco frame stacking preparation point of the tobacco frame stacking position according to the relative pose in the step S4;
s6: the AGV controller controls the AGV to drive to a smoke frame stacking preparation point according to the new temporary dynamic path;
s7: after the vehicle runs to the smoke frame stacking preparation point in the step S6, smoke frame stacking is carried out, and the AGV controller sends a vertical visual angle image acquisition command to the visual identification controller; when the AGV equipment runs to the smoke frame stacking preparation point, the AGV equipment does not displace, and only the smoke frame stacking is completed by moving a fork on the AGV;
s8: the visual recognition controller sends an acquisition command to the second visual sensor so that the second visual sensor shoots a cigarette frame positioned below the AGV fork and positioned at a cigarette frame stacking position from top to bottom to obtain vertical image data, the vertical image data are transmitted to the visual recognition controller, and the visual recognition controller performs filtering and feature extraction on the vertical image data to finally calculate left and right position deviation between the cigarette frame below the fork and the AGV fork;
s9: the AGV controller calculates left and right adjustment quantity of the AGV fork according to the left and right position deviation data in the S8, and controls the AGV fork to finish left and right adjustment;
s10: and the AGV controller controls the AGV fork to descend to complete the stacking of the cigarette frames.
Preferably, the planned path in S3 is a theoretical path for the AGV to place the individual cigarette frames that are forked from the ground or transport vehicle into the stacking position of the cigarette frames in the stacking warehouse, and the number of layers of cigarette frames stored in the stacking warehouse is not greater than 4.
Preferably, the first vision sensor can move up and down under the control of the AGV, so that the horizontal visual angles of the cigarette frames with different stacking layers can be shot; the second vision sensor moves up and down along with the AGV fork, so that the vertical visual angles of the different stacking layer cigarette frames are shot. Preferably, after the AGV takes the cigarette frame from the ground or the transport vehicle fork, the AGV controller controls the AGV fork to adjust to the horizontal state according to the AGV fork inclination angle value acquired by the inclination angle sensor.
Preferably, the front-point station in S3 is 1m-5m from the stack position of the cigarette box.
Preferably, the tobacco frame characteristic data is data expressing tobacco frame shape size and position information, including: the cigarette frame covers the bowl-shaped supporting leg and the corner point.
Preferably, the stacking precision of the cigarette frame is required to be within +/-20 mm, and the requirement that the cover bowl structure below the supporting legs of the cigarette frame is embedded with the supporting legs of the cigarette frame below after stacking is met.
Preferably, the method for identifying the feature data of the cigarette frame comprises the steps of firstly establishing a feature parameter template of the cigarette frame, wherein the feature parameter template comprises the outline dimension, the upright post width and the upright post height of the cigarette frame and further comprises point cloud feature data of the cigarette frame; the AGV can identify the forked or stacked tobacco frames by the following method, which comprises the following steps:
step a: the vision sensor collects the image data of the smoke block, and the vision recognition controller carries out filtering processing;
step b: the visual recognition controller acquires depth information of a recognition object from the filtered image data;
step c: obtaining point cloud data of the detected object according to the depth data of the identified object;
step d: performing point cloud registration according to the point cloud data of the detected object and the point cloud data in the tobacco frame characteristic parameter template established in advance;
step e: extracting key corner points and edge straight line characteristics of the cloud data of the detected tobacco frame points, calculating basic data of the overall dimension, the upright column width and the height of the tobacco frame, and comparing the basic data with tobacco frame data in the tobacco frame characteristic parameters established in advance;
step f: and calculating reference point coordinates of the tobacco frames according to the extracted key corner points and the edge linear characteristics of the tobacco frames, wherein the calculated coordinates of the tobacco frame stacking preparation points are the relative pose (dx, dy, dtota) of the tobacco frames relative to the vision sensor.
Preferably, when the first layer of cigarette frames stored in the stacking warehouse area are placed, the AGV places according to the set planning path, and the adjustment of the horizontal inclination angle is not needed.
An AGV-based cigarette frame identification and stacking system comprising: the system comprises an AGV, a first visual sensor, a second visual sensor, an inclination sensor and a visual identification controller, wherein the first visual sensor is horizontally and positively arranged in the middle of the AGV and shoots a cigarette frame at a stacking position through a horizontal visual angle, the second visual sensor is arranged on the side surface of the AGV fork and shoots the cigarette frame positioned below the AGV fork at the stacking position through vertical downward overlooking, the inclination sensor is arranged on the side surface of the AGV fork and used for detecting the inclination angle of the AGV fork, the visual identification controller is used for receiving the first visual sensor, the smoke frame image data of the second visual sensor and the AGV fork inclination angle data of the inclination sensor, and calculating the stacking position of the cigarette frame to obtain the relative pose of the cigarette frame; the AGV includes: the AGV controller is used for receiving the relative pose data of the visual recognition controller, and the AGV fork is controlled by the AGV controller.
Preferably, the first and second vision sensors are depth cameras.
The invention provides a tobacco frame identification and stacking method and system based on an AGV, which realize multi-layer and accurate stacking of tobacco frames. According to the invention, the first visual sensor and the second visual sensor are arranged at the specific position of the AGV according to the characteristics of the cigarette frame, and the recognition of the posture and the position of the cigarette frame is realized through shooting at a specific visual angle. Meanwhile, the attitude and the position deviation of the cigarette frames are obtained through comparison with the planned path, and after the adjustment quantity is calculated, the inclination angle sensor adjusts the angle of the AGV fork, so that high-precision cigarette frame stacking is realized. Third, the first visual sensor and the second visual sensor can be adjusted up and down according to the number of layers of the cigarette frame, and data of a horizontal visual angle and a vertical visual angle of the cigarette frame can be obtained in real time. In summary, the method and the system for identifying and stacking the cigarette frames based on the AGVs, which are based on the AGVs, perform AGV stacking design through the structural characteristics of the cigarette frames, save stacking space, ensure stacking stability, identify the characteristics of the cigarette frames through a visual identification method, utilize the bowl-covered supporting legs of the cigarette frames and the protruding cylinders or cuboids or other structures above the supporting legs, and enable the cigarette frames to be more stable through mutual matching and embedding, so that stacking precision reaches +/-20 mm.
Meanwhile, in the feature recognition of the tobacco frames, recognition calibration is performed aiming at the feature that a cover bowl structure below the supporting legs of the tobacco frames is matched with a protruding part structure on the upper parts of the supporting legs of the tobacco frames, so that when two tobacco frames are stacked together, the stability is higher, the stacking of the tobacco frames is assisted, the multi-layer stacking precision of the tobacco frames is improved, the sliding of the tobacco frames is avoided, and the automation of the storage of the tobacco frames is realized.
Drawings
Fig. 1: a smoke frame identification and stacking method flow chart based on AGVs.
Fig. 2: a method flow chart for identifying tobacco frame characteristic data.
Fig. 3: AGV-based smoke frame recognition and stacking system structure diagram.
Fig. 4: and a point cloud data schematic diagram showing the structural characteristics of the smoke frame part.
Fig. 5: relative position effect map of AGVs located in the station foreground and cigarette boxes in the stacking position.
Fig. 6: and (5) a smoke frame effect diagram after stacking.
In the figure: 1: AGVs; 2: a first vision sensor; 3: a second vision sensor; 4: an inclination sensor; 5: AGV fork; 6: planning a path; 7: a temporary dynamic path; 8: an actual cigarette frame; 9: an actual cigarette box stacking preparation point; 10: planning a tobacco frame; 11: planning a smoke frame stacking preparation point; 12: a front-end station; 13: point cloud data of the lower part structure of the smoke frame; 14: point cloud data of a part of frame structure on the smoke frame; 15: bowl-shaped supporting legs of the cigarette frame cover; 16: a tobacco frame edge frame is formed preliminarily according to the point cloud data; 17: and a protruding structure which is embedded in a matched manner with the supporting leg of the upper smoke frame.
Detailed Description
The invention is further described below with reference to the drawings and examples.
As shown in fig. 1 and 5, the present invention provides a smoke frame identifying and stacking method based on an AGV:
s1: the first vision sensor 2 is horizontally and positively arranged in the middle of the AGV1, a cigarette frame at a stacking position is shot at a horizontal view angle, the second vision sensor 3 is arranged on the side surface of a pallet fork of the AGV, and the cigarette frame at the stacking position is shot vertically downwards after the AGV runs to the stacking position; calibrating installation parameters among the first visual sensor 2, the second visual sensor 3 and the AGV1 and transmitting the parameters to an AGV controller;
s2: the inclination sensor 4 is arranged on the side face of the AGV fork to detect the inclination angle of the AGV fork 5;
s3: the method comprises the steps that a traveling path of an AGV stacking smoke frame is planned to serve as a planned path 6, traveling is suspended at a front point platform 12 of the planned path 6 when the AGV travels, a horizontal visual angle image acquisition command is sent to a visual recognition controller through the AGV controller, and the visual recognition controller sends the acquisition command to a first visual sensor to enable the first visual sensor to acquire smoke frame image data of a stacking position and send the smoke frame image data to the visual recognition controller; the front point station 12 is a neutral point set in the planned path of the AGV, at which position the AGV can take a picture of the cigarette frames in the stacking position at a horizontal angle of view using the first vision sensor;
s4: the vision recognition controller filters the smoke frame image data acquired in the step S3, extracts characteristic data describing the stacking position of the smoke frames, calculates the placement angle of the smoke frames at the stacking position and the distance between the smoke frames and the front point platform 12 where the AGV is positioned by using the characteristic data as a relative pose, and transmits the pose to the AGV controller;
s5: the AGV controller calculates a temporary dynamic path of the AGV reaching a tobacco frame stacking preparation point of the tobacco frame stacking position according to the relative pose in the step S4;
s6: the AGV controller controls the AGV to drive to a smoke frame stacking preparation point according to the new temporary dynamic path;
s7: after the vehicle runs to the smoke frame stacking preparation point in the step S6, smoke frame stacking is carried out, and the AGV controller sends a vertical visual angle image acquisition command to the visual identification controller; when the AGV equipment runs to the smoke frame stacking preparation point, the AGV equipment does not displace, and only the smoke frame stacking is completed by moving a fork on the AGV;
s8: the visual recognition controller sends an acquisition command to the second visual sensor 3 to enable the second visual sensor 3 to shoot a cigarette frame positioned below the AGV fork 5 and positioned at a cigarette frame stacking position from top to bottom to obtain vertical image data, the vertical image data are transmitted to the visual recognition controller, and the visual recognition controller filters and extracts features of the vertical image data to finally calculate left and right position deviation between the cigarette frame below the fork and the AGV fork;
s9: the AGV controller calculates left and right adjustment quantity of the AGV fork according to the left and right position deviation data in the S8, and controls the AGV fork to finish left and right adjustment;
s10: the AGV controller controls the AGV fork 5 to descend to complete the stacking of the cigarette frames.
Preferably, the planned path 6 described in S3 is a theoretical path for the AGV to place the individual cigarette frames that are being forked from the ground or transport vehicle into the stacking position of the cigarette frames in the stacking warehouse, and the number of cigarette frames stored in the stacking warehouse is not greater than 4. The tobacco frame stacking position reached by the planning path 6 in the AGV system is a planning tobacco frame stacking preparation point 11, and is different from an actual tobacco frame stacking preparation point 9, errors exist between the planning tobacco frame stacking preparation point 11 and the actual tobacco frame stacking preparation point 9, and if the errors of the planning tobacco frame stacking preparation point 11 and the actual tobacco frame stacking preparation point 9 are large, an alarm is generated, and the error range is 20mm.
Preferably, the first vision sensor can move up and down under the control of the AGV, so that the horizontal visual angles of the cigarette frames with different stacking layers can be shot; the second vision sensor moves up and down along with the AGV fork, so that the vertical visual angles of the different stacking layer cigarette frames are shot.
Preferably, after the AGV takes the cigarette frame from the ground or the transport vehicle fork, the AGV controller controls the AGV fork to adjust to the horizontal state according to the AGV fork inclination angle value acquired by the inclination angle sensor.
Preferably, the front-point station in S3 is 1m-5m from the stack position of the cigarette box.
Preferably, the tobacco frame characteristic data is data expressing tobacco frame shape size and position information, including: the cigarette frame covers the bowl-shaped supporting leg and the corner point.
Preferably, the stacking precision of the cigarette frame is required to be within +/-20 mm, and the requirement that the cover bowl structure below the supporting legs of the cigarette frame is embedded with the supporting legs of the cigarette frame below after stacking is met.
Preferably, as shown in fig. 2, the method for identifying the feature data of the cigarette frame includes firstly establishing a feature parameter template of the cigarette frame, wherein the feature parameter template comprises the outline dimension, the upright column width and the upright column height of the cigarette frame, and further comprises point cloud feature data of the cigarette frame; the AGV can identify the forked or stacked tobacco frames by the following method, which comprises the following steps:
step a: the vision sensor collects the image data of the smoke block, and the vision recognition controller carries out filtering processing;
step b: the visual recognition controller acquires depth information of a recognition object from the filtered image data;
step c: obtaining point cloud data of the detected object according to the depth data of the identified object;
step d: performing point cloud registration according to the point cloud data of the detected object and the point cloud data in the tobacco frame characteristic parameter template established in advance;
step e: extracting key corner points and edge straight line characteristics of the cloud data of the detected tobacco frame points, calculating basic data of the overall dimension, the upright column width and the height of the tobacco frame, and comparing the basic data with tobacco frame data in the tobacco frame characteristic parameters established in advance;
step f: according to the extracted key corner points and the edge straight line characteristics of the tobacco frame, calculating reference point coordinates of the tobacco frame, wherein the calculated coordinates of the tobacco frame stacking preparation points are the relative pose of the tobacco frame relative to the vision sensor, the relative pose can be represented by coordinates (dx, dy, dtheta), an AGV device is used as a coordinate origin, and dtheta is the tobacco frame inclination angle.
As shown in fig. 4, when stacking the cigarette frames, only a part of the structure of the cigarette frames needs to be identified due to the limitation of shooting distance or view angle, for example, when stacking the cigarette frames on the pallet fork to the cigarette frames stored in the stacking warehouse area, only the lower part structure of the cigarette frames on the pallet fork, namely the point cloud data 13 of the lower part structure of the cigarette frames in fig. 4, is required to be obtained, and the size data such as the supporting legs, the width and the like of the cigarette frames can be displayed; for the tobacco frames stored in the stacking warehouse area, only point cloud data of part of frame structures on the tobacco frames are required to be acquired, for example, a protruding structure which is required in the stacking process and is inlaid in cooperation with the supporting legs of the tobacco frames above, namely 14 in fig. 4, is required to perform feature recognition on part of structural features required in the stacking to acquire the point cloud data, so that the distance of shooting the tobacco frames by a camera is not limited, and the calculation amount is saved.
Preferably, when the first layer of cigarette frames stored in the stacking warehouse area are placed, the AGV places according to the set planning path, and the adjustment of the horizontal inclination angle is not needed. To meet the need of stacking or extracting each cigarette frame, the planned paths are generally arranged horizontally, and the AGV devices can pass through the planned paths at intervals between the rows.
As shown in fig. 3, an AGV-based smoke box recognition and stacking system, comprising: the system comprises an AGV, a first visual sensor, a second visual sensor, an inclination sensor and a visual identification controller, wherein the first visual sensor is horizontally and positively arranged in the middle of the AGV and shoots a cigarette frame at a stacking position through a horizontal visual angle, the second visual sensor is arranged on the side surface of the AGV fork and shoots the cigarette frame positioned below the AGV fork at the stacking position through vertical downward overlooking, the inclination sensor is arranged on the side surface of the AGV fork and used for detecting the inclination angle of the AGV fork, the visual identification controller is used for receiving the first visual sensor, the smoke frame image data of the second visual sensor and the AGV fork inclination angle data of the inclination sensor, and calculating the stacking position of the cigarette frame to obtain the relative pose of the cigarette frame; the AGV includes: the AGV controller is used for receiving the relative pose data of the visual recognition controller, and the AGV fork is controlled by the AGV controller.
Preferably, the first and second vision sensors are depth cameras.
The height of the cigarette frame in the embodiment of the invention is 1400-1500mm.
It should be noted that the above-described embodiments will enable those skilled in the art to more fully understand the invention, but do not limit it in any way. Therefore, although the present invention has been described in detail with reference to the drawings and examples, it will be understood by those skilled in the art that the present invention may be modified or equivalent, and in all cases, all technical solutions and modifications which do not depart from the spirit and scope of the present invention are intended to be included in the scope of the present invention.
Claims (11)
1. A smoke frame identification and stacking method based on AGV is characterized in that,
s1: the method comprises the steps of installing a first vision sensor horizontally and positively in the middle of an AGV, shooting a cigarette frame at a stacking position in a horizontal view angle, installing a second vision sensor on the side face of a fork of the AGV, and shooting the cigarette frame at the stacking position vertically downwards after the AGV runs to the stacking position; calibrating installation parameters among the first visual sensor, the second visual sensor and the AGV and transmitting the installation parameters to an AGV controller;
s2: installing an inclination sensor on the side surface of an AGV fork to detect the inclination angle of the AGV fork;
s3: planning a traveling path of an AGV stacking smoke frame as a planning path, suspending traveling at a front point platform of the planned path when the AGV travels, and sending a horizontal visual angle image acquisition command to a visual identification controller through the AGV controller, wherein the visual identification controller sends the acquisition command to a first visual sensor so that the first visual sensor acquires smoke frame image data of a stacking position and sends the smoke frame image data to the visual identification controller; the front point platform is a middle point arranged in an AGV planning path, and the AGV at the position can shoot cigarette frames at a stacking position in a horizontal view angle by using a first visual sensor;
s4: the vision recognition controller filters the smoke frame image data acquired in the step S3, extracts characteristic data describing the stacking position of the smoke frames, calculates the placement angle of the smoke frames at the stacking position and the distance between the smoke frames and a front point platform where the AGV is located by using the characteristic data as a relative pose, and transmits the pose to the AGV controller;
s5: the AGV controller calculates a temporary dynamic path of the AGV reaching a tobacco frame stacking preparation point of the tobacco frame stacking position according to the relative pose in the step S4;
s6: the AGV controller controls the AGV to drive to a smoke frame stacking preparation point according to the new temporary dynamic path;
s7: after the vehicle runs to the smoke frame stacking preparation point in the step S6, smoke frame stacking is carried out, and the AGV controller sends a vertical visual angle image acquisition command to the visual identification controller; when the AGV equipment runs to the smoke frame stacking preparation point, the AGV equipment does not displace, and only the smoke frame stacking is completed by moving a fork on the AGV;
s8: the visual recognition controller sends an acquisition command to the second visual sensor so that the second visual sensor shoots a cigarette frame positioned below the AGV fork and positioned at a cigarette frame stacking position from top to bottom to obtain vertical image data, the vertical image data are transmitted to the visual recognition controller, and the visual recognition controller performs filtering and feature extraction on the vertical image data to finally calculate left and right position deviation between the cigarette frame below the fork and the AGV fork;
s9: the AGV controller calculates left and right adjustment quantity of the AGV fork according to the left and right position deviation data in the S8, and controls the AGV fork to finish left and right adjustment;
s10: and the AGV controller controls the AGV fork to descend to complete the stacking of the cigarette frames.
2. The method of claim 1, wherein the planned path in S3 is a theoretical path for the AGV to place the individual cigarette frames from the ground or transport vehicle into the stacking position of the cigarette frames in the stacking warehouse, and the number of layers of the cigarette frames stored in the stacking warehouse is not greater than 4.
3. The method for identifying and stacking cigarette frames based on the AGVs according to claim 2, wherein the first vision sensor can move up and down under the control of the AGVs, so that the horizontal visual angles of the cigarette frames with different stacking layers can be shot; the second vision sensor moves up and down along with the AGV fork, so that the vertical visual angles of the different stacking layer cigarette frames are shot.
4. The method for recognizing and stacking cigarette frames based on the AGVs according to claim 2, wherein the AGV controller controls the AGVs to be adjusted to a horizontal state according to the AGV fork inclination angle value collected by the inclination sensor after the AGVs fork the cigarette frames from the ground or the transport vehicles.
5. The AGV-based cigarette frame recognition and stacking method according to claim 2, wherein the front point station is 1m-5m from the cigarette frame stacking position in S3.
6. The AGV-based cigarette frame identification and stacking method of claim 2, wherein the cigarette frame characteristic data is data expressing cigarette frame shape size and position information, comprising: the cigarette frame covers the bowl-shaped supporting leg and the corner point.
7. The AGV-based cigarette frame identification and stacking method according to claim 2, wherein the error in the accuracy of stacking the cigarette frames is within ±20mm, and the requirement that the cover bowl structure below the supporting legs of the cigarette frames is embedded with the supporting legs of the lower cigarette frames after stacking is satisfied.
8. The AGV-based tobacco frame identification and stacking method according to claim 2 is characterized in that the tobacco frame characteristic data identification method comprises the steps of firstly establishing a characteristic parameter template of a tobacco frame, wherein the characteristic parameter template comprises the outline dimension, the upright post width and the upright post height of the tobacco frame, and further comprises point cloud characteristic data of the tobacco frame; the AGV can identify the forked or stacked tobacco frames by the following method, which comprises the following steps:
step a: the vision sensor collects the image data of the smoke block, and the vision recognition controller carries out filtering processing;
step b: the visual recognition controller acquires depth information of a recognition object from the filtered image data;
step c: obtaining point cloud data of the detected object according to the depth data of the identified object;
step d: performing point cloud registration according to the point cloud data of the detected object and the point cloud data in the tobacco frame characteristic parameter template established in advance;
step e: extracting key corner points and edge straight line characteristics of the cloud data of the detected tobacco frame points, calculating basic data of the overall dimension, the upright column width and the height of the tobacco frame, and comparing the basic data with tobacco frame data in the tobacco frame characteristic parameters established in advance;
step f: and calculating reference point coordinates of the tobacco frames according to the extracted key corner points and the edge linear characteristics of the tobacco frames, wherein the calculated coordinates of the tobacco frame stacking preparation points are the relative pose (dx, dy, dtota) of the tobacco frames relative to the vision sensor.
9. The method for identifying and stacking cigarette frames based on AGVs according to claim 2 wherein the AGVs are placed according to the set planned path without adjusting the horizontal inclination angle when the first layer of cigarette frames stored in the stacking warehouse area are placed.
10. An AGV-based cigarette frame identification and stacking system based on the method of claim 1, comprising: the system comprises an AGV, a first visual sensor, a second visual sensor, an inclination sensor and a visual identification controller, wherein the first visual sensor is horizontally and positively arranged in the middle of the AGV and shoots a cigarette frame at a stacking position through a horizontal visual angle, the second visual sensor is arranged on the side surface of the AGV fork and shoots the cigarette frame positioned below the AGV fork at the stacking position through vertical downward overlooking, the inclination sensor is arranged on the side surface of the AGV fork and used for detecting the inclination angle of the AGV fork, the visual identification controller is used for receiving the first visual sensor, the smoke frame image data of the second visual sensor and the AGV fork inclination angle data of the inclination sensor, and calculating the stacking position of the cigarette frame to obtain the relative pose of the cigarette frame; the AGV includes: the AGV controller is used for receiving the relative pose data of the visual recognition controller, and the AGV fork is controlled by the AGV controller.
11. The AGV-based fume frame identification and stacking system of claim 10, wherein said first and second vision sensors are depth cameras.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310173761.6A CN115848878B (en) | 2023-02-28 | 2023-02-28 | AGV-based tobacco frame identification and stacking method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310173761.6A CN115848878B (en) | 2023-02-28 | 2023-02-28 | AGV-based tobacco frame identification and stacking method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115848878A CN115848878A (en) | 2023-03-28 |
| CN115848878B true CN115848878B (en) | 2023-05-26 |
Family
ID=85659218
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310173761.6A Active CN115848878B (en) | 2023-02-28 | 2023-02-28 | AGV-based tobacco frame identification and stacking method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115848878B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008230736A (en) * | 2007-03-19 | 2008-10-02 | Daifuku Co Ltd | Article carrying device |
| CN109437058A (en) * | 2018-12-18 | 2019-03-08 | 云南昆船智能装备有限公司 | One kind being applied to high lifting V shape AGV pallet fork leveling device |
| CN110001661A (en) * | 2019-04-15 | 2019-07-12 | 安徽意欧斯物流机器人有限公司 | A kind of binocular vision navigation fork-lift type AGV control system and method |
| WO2019163378A1 (en) * | 2018-02-23 | 2019-08-29 | コニカミノルタ株式会社 | Forklift image processing device and control program |
| CN110884901A (en) * | 2019-10-12 | 2020-03-17 | 浙江大华机器人技术有限公司 | Deviation rectifying method, device and equipment for AGV butt joint of stacking vehicle and storage medium |
| CN112125226A (en) * | 2020-09-16 | 2020-12-25 | 中建材创新科技研究院有限公司 | Visual detection system and method for laser navigation forklift |
| CN114772142A (en) * | 2021-11-29 | 2022-07-22 | 苏州百年职业学院 | Intelligent logistics storage system based on visual analysis and material scheduling method |
| CN115346211A (en) * | 2022-08-31 | 2022-11-15 | 中电科机器人有限公司 | Visual recognition method, visual recognition system and storage medium |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9715232B1 (en) * | 2016-09-19 | 2017-07-25 | X Development Llc | Using planar sensors for pallet detection |
| CN106672859B (en) * | 2017-01-05 | 2018-11-09 | 深圳市有光图像科技有限公司 | A kind of method and fork truck based on fork truck visual identity pallet |
| US10576526B2 (en) * | 2018-07-03 | 2020-03-03 | Komatsu Industries Corporation | Workpiece conveying system, and workpiece conveying method |
| US11353858B2 (en) * | 2019-01-18 | 2022-06-07 | United States Postal Service | Systems and methods for automated guided vehicle control |
| US20210114826A1 (en) * | 2019-10-16 | 2021-04-22 | Symbotic Canada, Ulc | Vision-assisted robotized depalletizer |
| CN112678724B (en) * | 2019-10-18 | 2022-06-10 | 北京极智嘉科技股份有限公司 | Intelligent forklift and control method thereof |
| CN113050636A (en) * | 2021-03-15 | 2021-06-29 | 广东省农业科学院 | Control method, system and device for autonomous tray picking of forklift |
| CN112660686B (en) * | 2021-03-17 | 2021-06-01 | 杭州蓝芯科技有限公司 | Depth camera-based material cage stacking method and device, electronic equipment and system |
| CN215679136U (en) * | 2021-09-27 | 2022-01-28 | 东风本田发动机有限公司 | Navigation system and AGV dolly |
| CN113998626A (en) * | 2021-09-27 | 2022-02-01 | 诺力智能装备股份有限公司 | AGV attitude adjusting method applied to tray recognition/positioning |
| CN115123839B (en) * | 2022-09-02 | 2022-12-09 | 杭叉集团股份有限公司 | AGV-based bin stacking method, device, equipment and storage medium |
-
2023
- 2023-02-28 CN CN202310173761.6A patent/CN115848878B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008230736A (en) * | 2007-03-19 | 2008-10-02 | Daifuku Co Ltd | Article carrying device |
| WO2019163378A1 (en) * | 2018-02-23 | 2019-08-29 | コニカミノルタ株式会社 | Forklift image processing device and control program |
| CN109437058A (en) * | 2018-12-18 | 2019-03-08 | 云南昆船智能装备有限公司 | One kind being applied to high lifting V shape AGV pallet fork leveling device |
| CN110001661A (en) * | 2019-04-15 | 2019-07-12 | 安徽意欧斯物流机器人有限公司 | A kind of binocular vision navigation fork-lift type AGV control system and method |
| CN110884901A (en) * | 2019-10-12 | 2020-03-17 | 浙江大华机器人技术有限公司 | Deviation rectifying method, device and equipment for AGV butt joint of stacking vehicle and storage medium |
| CN112125226A (en) * | 2020-09-16 | 2020-12-25 | 中建材创新科技研究院有限公司 | Visual detection system and method for laser navigation forklift |
| CN114772142A (en) * | 2021-11-29 | 2022-07-22 | 苏州百年职业学院 | Intelligent logistics storage system based on visual analysis and material scheduling method |
| CN115346211A (en) * | 2022-08-31 | 2022-11-15 | 中电科机器人有限公司 | Visual recognition method, visual recognition system and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 基于双目视觉的无人引导运输装置的研究;吴年祥;吴卫东;;阴山学刊(自然科学版)(04);44-47 * |
| 多目视角定位协同引导自动运输车研究;曾群生等;《计算机工程与应用》;第54卷(第11期);260-264+270 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115848878A (en) | 2023-03-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109436820B (en) | Destacking method and destacking system for goods stack | |
| EP3586314B1 (en) | Improved forest surveying | |
| US9227323B1 (en) | Methods and systems for recognizing machine-readable information on three-dimensional objects | |
| US8150658B2 (en) | Three-dimensional object recognition system and inventory system using the same | |
| CN112849898B (en) | Self-driven robot and carrying method thereof | |
| CN109160452A (en) | Unmanned transhipment fork truck and air navigation aid based on laser positioning and stereoscopic vision | |
| US20140267703A1 (en) | Method and Apparatus of Mapping Landmark Position and Orientation | |
| US20180143624A1 (en) | Method and industrial truck for ascertaining and representing the position of storage bins in a warehouse | |
| CN114035584B (en) | Method for detecting obstacle by robot, robot and robot system | |
| KR20170016010A (en) | Lost vehicle recovery utilizing associated feature pairs | |
| JP2017534836A (en) | Combination of stereo processing and structured light processing | |
| EP3745085A1 (en) | Multi-device visual navigation method and system in variable scene | |
| CN111824668B (en) | Robot and robot-based container storage and retrieval method | |
| CN112660686B (en) | Depth camera-based material cage stacking method and device, electronic equipment and system | |
| US11591197B2 (en) | Load handling module for a material handling vehicle | |
| CN114241269B (en) | A collection card vision fuses positioning system for bank bridge automatic control | |
| CN115848878B (en) | AGV-based tobacco frame identification and stacking method and system | |
| CN115582827A (en) | Unloading robot grabbing method based on 2D and 3D visual positioning | |
| CN114170521A (en) | Forklift pallet butt joint identification positioning method | |
| CN116374474B (en) | Picking intelligent decision-making system based on machine vision | |
| CN115289966A (en) | Goods shelf detecting and positioning system and method based on TOF camera | |
| US20220355474A1 (en) | Method and computing system for performing robot motion planning and repository detection | |
| CN116309882A (en) | Tray detection and positioning method and system for unmanned forklift application | |
| CN117897348A (en) | Goods shelf positioning method, goods shelf butt joint method, device, equipment and medium | |
| CN112611344B (en) | Autonomous mobile flatness detection method, device and 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 |