CN105993393A - Picking method for flower harvesting robot - Google Patents
Picking method for flower harvesting robot Download PDFInfo
- Publication number
- CN105993393A CN105993393A CN201510389142.6A CN201510389142A CN105993393A CN 105993393 A CN105993393 A CN 105993393A CN 201510389142 A CN201510389142 A CN 201510389142A CN 105993393 A CN105993393 A CN 105993393A
- Authority
- CN
- China
- Prior art keywords
- target
- picking
- flower
- targets
- air suction
- 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
Landscapes
- Manipulator (AREA)
Abstract
Provided is a picking method for a flower harvesting robot. The method comprises steps: through step control, moving to a picking region; through a visual sensor, observing the picking region, and through a color filter algorithm, rapidly realizing to obtain a target in a complex background, to obtain a plurality of targets; through a dimension learning algorithm, screening the plurality of targets; using a stereoscopic vision detection system, through left and right eye cameras, obtaining flower image information, and extracting and matching information, to realize position posture detection, generating a position posture database of targets which can be picked; using multi-arm robot parallel synergic movement control algorithm to carry out coordination control among multi-parallem mechanisms, to geneate targets in the position posture database of targets which can be picked, and diving work, performing path optimization on the targets which obtain work division, finding the shortest path of mechanical arm movement, as motion planning requirement of a single mechanical arm. The method overcomes problems that crop targets cannot be effectively identified and picked in complex environment of agricultural work.
Description
Technical field
The present invention relates to a kind of picking robot picking method being applicable to the crops such as flower, belong to agricultural technology processing equipment field.
Background technology
Along with world population ages's trend is obvious, rural laborer is progressively run off, and human cost has become the key factor of restriction agricultural development.China is large agricultural country, therefore the needs of agricultural machinery are the highest, pluck labour force used by operation simultaneously and account for the 30% to 50% of whole production process effect labour force, and the current harvesting of China mainly also relies on manually, therefore about comes the most urgent to the demand of picking robot.
Picking robot is the most rare in the flower field with higher economic worth, such as Flos Jasmini Sambac etc., its flower is plucked has bigger market, artificial cost of plucking has become the principal element of its product cost composition, the most therefore in the urgent need to by carrying out autonomous picked technology research reduction product cost.But complex by its target environment, how the difficult problems such as single goal is less, improve target location accuracy and picking efficiency becomes the matter of utmost importance developing flower picking robot.
The present invention can be according to the difference of target property, and adaptation completes the autonomous harvesting of crops many to Cotton Gossypii, mushroom etc. simultaneously, and can be applicable in the operations such as industry automatic sorting.
Summary of the invention
Solved by the invention technical problem is that: complete the flower preliminary extraction under the complex backgrounds such as grove by color filter, it is achieved the filtration to complex job background；Utilize size learning method, determine the relative position relation with target according to stereoscopic vision detection system, target shared pixel and characteristic in divisible region in stereoscopic vision detection system can be plucked by analyzing, determine the threshold value of object filtering.Use stereoscopic vision detection system, by detection, coupling and position reconstruction to target signature information, obtain the target positional information relative to stereoscopic vision detection system, and then set up the position and attitude data base of target, overcome the problem that under agricultural operation complex environment, crop target cannot effectively be identified and be plucked.
Meanwhile, controlled by multi-parallel mechanism synergy movement, can realize plucking multiobject simultaneously, improve work efficiency further, solve current picking robot efficiency and the artificial difficult problem that there is bigger gap.
The present invention is the picking method of flower picking robot, and its step is as follows:
A. relevant range is moved to by step motion control；
B. by vision sensor observed object region, and quickly realize under complex background the acquisition to target by color filter algorithm, obtain multiple target；
C. by size learning algorithm, the multiple targets obtaining step b are screened；
D. stereoscopic vision detection system is utilized, obtained the image information of flower by left and right mesh camera, and the information in two cameras is extracted, mated, carry out light path and inversely settle accounts, realizing the target after screening step c and carry out position and attitude detection, generation can pluck object pose data base；
E. use many mechanical arms parallel connection synergy movement control algolithm, carry out the inter-agency coordination of multi-parallel and control, the generation of step d can be plucked the target in object pose data base and quickly divide the work and path optimization, obtain motion planning requirement；
F. the motion planning requirement obtained according to step e, controls mechanical arm and completes to hold together the bag of target, and by end sickle fly-cutting target；
G. Built-in air suction formula device in end sickle, opens air suction type device while cutting；
H. air suction type device suction arranges and substantially keeps consistent with target weight, and getter device is directly connected in storage box, completes the storage to flower.
The picking method of described flower picking robot, plucking target is Flos Jasmini Sambac or Cotton Gossypii.
The picking method of described flower picking robot, it is characterised in that step b uses color filter to carry out image segmentation, by analysis background grove and flower discrimination on color characteristics, utilizes color threshold to filter, completes the filtration to grove.
The picking method of described flower picking robot, it is characterized in that step c uses size learning method, relative position relation according to stereoscopic vision detection system Yu target, target shared pixel and characteristic in divisible region in detector can be plucked by analyzing, determine the threshold value of object filtering, it is achieved pluck object filtering.
The picking method of described flower picking robot, it is characterized in that step d carries out synchronizing location for the target after screening, utilize stereoscopic vision detection system that multiple target is carried out synchro measure, realize the disposable covering operation that can pluck target location attitude, and Generate Target Position attitude information data base.
The picking method of described flower picking robot, it is characterized in that the object pose information database that the target location attitude information data base that step e utilizes step d to obtain obtains, by to multi-parallel mechanism synergy movement control algolithm, the motion path of a plurality of mechanical arm is planned, avoid mutually colliding, and be optimized analysis with efficiency for design load.
The picking method of described flower picking robot, it is characterised in that target is completed after bag holds together and cut completing step f by step g, is drawn into collecting pipe by air suction type device by flower, and includes in storage bin and store.
The present invention gives and a kind of can complete noncooperative target feature extraction this to flower and the method for pose measurement under complex background, the flower preliminary extraction under the complex backgrounds such as grove is completed by color filter, realize the filtration to complex job background, utilize size learning method, relative position relation according to stereoscopic vision detection system Yu target, target can be plucked in pixel shared by stereoscopic vision detection system and the characteristic in divisible region by analyzing, determine the threshold value of object filtering.Using stereoscopic vision detection system, by detection, coupling and position reconstruction to target signature information, acquisition target, relative to the positional information of detector, solves the difficult problem that noncooperative target posture information is difficult under non-structure environment measure.Contribute to advancing agricultural AUTONOMOUS TASK technology development.
The present invention use image segmentation algorithm split from non-structure environment by Flos Jasmini Sambac, and utilize size learning method to utilize stereoscopic vision to complete the measurement to object pose to plucking after target is screened, it is effectively increased the efficiency to object pose information measurement.
Use multi-parallel mechanism synergy movement control algolithm, most have planning by avoidance and path, farthest ensure execution efficiency, efficiently solve tradition picking robot inefficiency and cannot be carried out a difficult problem for engineer applied.
The present invention uses air suction type device to realize the automatic of target after plucking and obtains and collect, improve system work efficiency further.Project completes the autonomous harvesting of crops many to Cotton Gossypii, mushroom etc. by adaptation, and can be applicable in the operations such as industry automatic sorting.
Accompanying drawing explanation
Fig. 1 is the harvesting flow process of flower picking robot；
Fig. 2 is flower picking robot conceptual scheme.
1-stereoscopic vision detection system, the many mechanical arms of 2-, 3-sickle and air suction type device, 4-storage bin.
Detailed description of the invention
With detailed description of the invention, technical scheme is described in further details below in conjunction with the accompanying drawings.Obviously, described embodiment is only a part of embodiment of the present invention rather than whole embodiments.Based on embodiments of the invention, the every other embodiment that those skilled in the art are obtained under not making creative work premise, broadly fall into the scope of protection of present invention.
Embodiment 1:
The present invention is the picking method of a kind of Flos Jasmini Sambac picking robot, as it is shown in figure 1, the present invention to be embodied as step as follows:
A. utilize walking mechanism to realize the movement above jasmine flowering shrubs, and move to pluck region by step motion control；
B. by vision sensor observed object region, and quickly realize under complex background the acquisition to target by color filter algorithm, obtain multiple target；
C. by size learning algorithm, the multiple targets obtaining step b are screened；
D. stereoscopic vision detection system is utilized, obtained the image information of flower by left and right mesh camera, and the information in two cameras is extracted, mated, carry out light path and inversely settle accounts, realizing the target after screening step c and carry out position and attitude detection, generation can pluck object pose data base；
E. many mechanical arms parallel connection synergy movement control algolithm is used, carry out the inter-agency coordination of multi-parallel to control, step d is generated the target can plucked in object pose data base quickly divide the work, and the target after obtaining the division of labor is carried out path optimization, find the shortest path of manipulator motion, as the motion planning requirement of independent mechanical arm；
F. the motion planning requirement obtained according to step e, controls mechanical arm and completes to hold together the bag of target, and by end sickle fly-cutting target；
G. Built-in air suction formula device in end sickle, opens air suction type device while cutting；
H. air suction type device suction arranges and substantially keeps consistent with target weight, and getter device is directly connected in storage box, completes the storage to Flos Jasmini Sambac.
Embodiment 2:
The present invention is the picking method of a kind of cotton picking robot, as it is shown in figure 1, the present invention to be embodied as step as follows:
A. utilize walking mechanism to realize the movement above Cotton Gossypii clump, and move to pluck region by step motion control；
B. by vision sensor observed object region, and quickly realize under complex background the acquisition to target by color filter algorithm, obtain multiple target；
C. by size learning algorithm, the multiple targets obtaining step b are screened；
D. stereoscopic vision detection system is utilized, obtained the image information of flower by left and right mesh camera, and the information in two cameras is extracted, mated, carry out light path and inversely settle accounts, realizing the target after screening step c and carry out position and attitude detection, generation can pluck object pose data base；
E. many mechanical arms parallel connection synergy movement control algolithm is used, carry out the inter-agency coordination of multi-parallel to control, step d is generated the target can plucked in object pose data base quickly divide the work, and the target after obtaining the division of labor is carried out path optimization, find the shortest path of manipulator motion, as the motion planning requirement of independent mechanical arm；
F. the motion planning requirement obtained according to step e, controls mechanical arm and completes to hold together the bag of target, and by end sickle fly-cutting target；
G. Built-in air suction formula device in end sickle, opens air suction type device while cutting；
H. air suction type device suction arranges and substantially keeps consistent with target weight, and getter device is directly connected in storage box, completes the storage to Cotton Gossypii.
Described above to the disclosed embodiments, makes those skilled in the art be capable of or uses the present invention.Multiple amendment to these embodiments is apparent from for the patented technology personnel of this area, and generic principles defined herein can realize without departing from the present invention in other embodiments.Therefore, the present invention is not intended to be limited to the embodiments shown herein, and is to fit to the widest range consistent with principles disclosed herein and features of novelty.
Claims (8)
1. a picking method for flower picking robot, its step is as follows:
A. move to pluck region by step motion control；
B. pluck region by vision sensor observation, and quickly realized under complex background target by color filter algorithm
Obtain, obtain multiple target；
C. by size learning algorithm, the multiple targets obtaining step b are screened；
D. utilize stereoscopic vision detection system, obtained the image information of flower by left and right mesh camera, and in two cameras
Information carries out extracting, mating, and carries out light path and inversely settles accounts, it is achieved the target after screening c carries out position and attitude detection, generates
Attitude data storehouse, target location can be plucked；
E. use many mechanical arms parallel connection synergy movement control algolithm, carry out the inter-agency coordination of multi-parallel and control, step d is generated
The target can plucked in object pose data base is divided the work, and the target after obtaining the division of labor is carried out path optimization, finds machinery
The shortest path of arm motion, as the motion planning requirement of independent mechanical arm；
F. the motion planning requirement obtained according to step e, controls mechanical arm and completes to hold together the bag of target, and by end sickle
Fly-cutting target；
G. Built-in air suction formula device in end sickle, while cutting, opens air suction type device；
H. air suction type device suction arranges and substantially keeps consistent with target weight, and getter device is directly connected in storage box, completes
Storage to flower.
2. the picking method of picking robot as claimed in claim 1, it is characterised in that plucking target is Flos Jasmini Sambac or Cotton Gossypii.
3. the picking method of picking robot as claimed in claim 1, it is characterised in that step b uses color filter to carry out figure
As segmentation, by analysis background grove and flower discrimination on color characteristics, utilize color threshold to filter, complete grove
Filtration.
4. the picking method of picking robot as claimed in claim 1, it is characterised in that step c uses size learning method, root
According to the relative position relation of stereoscopic vision detection system Yu target, target shared pixel and can in detector can be plucked by analyzing
The characteristic of cut zone, determines the threshold value of object filtering, it is achieved pluck object filtering.
5. the picking method of picking robot as claimed in claim 1, it is characterised in that step d is entered for the target after screening
Row synchronizes location, utilizes Binocular Stereo Vision System that multiple target is carried out synchro measure, it is achieved to plucking target location attitude
Disposable covering operation, and Generate Target Position attitude information data base.
6. the picking method of picking robot as claimed in claim 1, it is characterised in that step e utilizes the mesh that step d obtains
Cursor position attitude information data base, by multi-parallel mechanism synergy movement control algolithm, entering the motion path of a plurality of mechanical arm
Professional etiquette is drawn, it is to avoid mutually collide, and is optimized analysis with efficiency for design load.
7. the picking method of picking robot as claimed in claim 1, it is characterised in that step g completes step f to target
Complete after bag holds together and cut, by air suction type device flower is drawn into collecting pipe, and include in storage bin and store.
8. a flower picking robot, it is characterised in that including: stereoscopic vision detection system (1), multi-robot arm (2),
End sickle and air suction type device (3), storage box (4), wherein Built-in air suction formula device (3), air-breathing in end sickle
Device (3) is directly connected in storage box (4).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510389142.6A CN105993393A (en) | 2015-07-03 | 2015-07-03 | Picking method for flower harvesting robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510389142.6A CN105993393A (en) | 2015-07-03 | 2015-07-03 | Picking method for flower harvesting robot |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105993393A true CN105993393A (en) | 2016-10-12 |
Family
ID=57082401
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510389142.6A Pending CN105993393A (en) | 2015-07-03 | 2015-07-03 | Picking method for flower harvesting robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105993393A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106973631A (en) * | 2017-04-27 | 2017-07-25 | 中国人民解放军防空兵学院 | One kind automation picker system |
| WO2018133240A1 (en) * | 2017-01-23 | 2018-07-26 | 东莞理工学院 | Automated agricultural pollination method based on micro flying robot |
| CN109739133A (en) * | 2019-01-08 | 2019-05-10 | 太原工业学院 | Tomato picking robot system and its control method based on radar fix |
| CN109978829A (en) * | 2019-02-26 | 2019-07-05 | 深圳市华汉伟业科技有限公司 | A kind of detection method and its system of object to be detected |
| CN110744570A (en) * | 2019-11-27 | 2020-02-04 | 佛山科学技术学院 | Pineapple picking machine and picking method thereof |
| CN113728806A (en) * | 2021-11-05 | 2021-12-03 | 季华实验室 | Fruit picking robot control method and device, electronic equipment and storage medium |
| CN114260894A (en) * | 2021-12-22 | 2022-04-01 | 上海科技大学 | Full-automatic cotton picking robot and control method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5119392B2 (en) * | 2007-02-26 | 2013-01-16 | 井関農機株式会社 | Fruit harvesting robot and strawberry cultivation facility |
| CN103348819A (en) * | 2013-07-12 | 2013-10-16 | 中南林业科技大学 | Oil-tea camellia fruit picking method and oil-tea camellia fruit picking robot using same |
| CN103950033A (en) * | 2014-04-18 | 2014-07-30 | 南京农业大学 | Mechanical arm and end effector of fruit picking robot and fruit picking method |
| CN104050668A (en) * | 2014-06-17 | 2014-09-17 | 安徽农业大学 | Object recognition method applied to green tea tender shoots and based on binocular vision technology |
| CN104067781A (en) * | 2014-06-16 | 2014-10-01 | 华南农业大学 | Virtual robot and real robot integration based picking system and method |
| CN204132001U (en) * | 2014-09-25 | 2015-02-04 | 甘肃农业大学 | A kind of pepper picking machine people based on machine vision |
-
2015
- 2015-07-03 CN CN201510389142.6A patent/CN105993393A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5119392B2 (en) * | 2007-02-26 | 2013-01-16 | 井関農機株式会社 | Fruit harvesting robot and strawberry cultivation facility |
| CN103348819A (en) * | 2013-07-12 | 2013-10-16 | 中南林业科技大学 | Oil-tea camellia fruit picking method and oil-tea camellia fruit picking robot using same |
| CN103950033A (en) * | 2014-04-18 | 2014-07-30 | 南京农业大学 | Mechanical arm and end effector of fruit picking robot and fruit picking method |
| CN104067781A (en) * | 2014-06-16 | 2014-10-01 | 华南农业大学 | Virtual robot and real robot integration based picking system and method |
| CN104050668A (en) * | 2014-06-17 | 2014-09-17 | 安徽农业大学 | Object recognition method applied to green tea tender shoots and based on binocular vision technology |
| CN204132001U (en) * | 2014-09-25 | 2015-02-04 | 甘肃农业大学 | A kind of pepper picking machine people based on machine vision |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018133240A1 (en) * | 2017-01-23 | 2018-07-26 | 东莞理工学院 | Automated agricultural pollination method based on micro flying robot |
| CN106973631A (en) * | 2017-04-27 | 2017-07-25 | 中国人民解放军防空兵学院 | One kind automation picker system |
| CN109739133A (en) * | 2019-01-08 | 2019-05-10 | 太原工业学院 | Tomato picking robot system and its control method based on radar fix |
| CN109978829A (en) * | 2019-02-26 | 2019-07-05 | 深圳市华汉伟业科技有限公司 | A kind of detection method and its system of object to be detected |
| CN110744570A (en) * | 2019-11-27 | 2020-02-04 | 佛山科学技术学院 | Pineapple picking machine and picking method thereof |
| CN113728806A (en) * | 2021-11-05 | 2021-12-03 | 季华实验室 | Fruit picking robot control method and device, electronic equipment and storage medium |
| CN113728806B (en) * | 2021-11-05 | 2022-01-18 | 季华实验室 | Fruit picking robot control method and device, electronic equipment and storage medium |
| CN114260894A (en) * | 2021-12-22 | 2022-04-01 | 上海科技大学 | Full-automatic cotton picking robot and control method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105993393A (en) | Picking method for flower harvesting robot | |
| Ling et al. | Dual-arm cooperation and implementing for robotic harvesting tomato using binocular vision | |
| Chen et al. | Three-dimensional perception of orchard banana central stock enhanced by adaptive multi-vision technology | |
| Zhao et al. | A review of key techniques of vision-based control for harvesting robot | |
| Davidson et al. | Proof-of-concept of a robotic apple harvester | |
| CN104239886B (en) | The extracting method on lawn and background line of demarcation based on graphical analysis | |
| Yuan et al. | An autonomous pollination robot for hormone treatment of tomato flower in greenhouse | |
| CN101356877A (en) | Cucumber picking robot system and picking method in greenhouse | |
| CN103049912A (en) | Random trihedron-based radar-camera system external parameter calibration method | |
| Zhaoxin et al. | Design a robot system for tomato picking based on YOLO v5 | |
| Ji et al. | Research on key technology of truss tomato harvesting robot in greenhouse | |
| CN113226013A (en) | Saffron crocus collection device | |
| JP4961555B2 (en) | Method for locating target part of plant, target part locating device by the method, and working robot using the device | |
| Yoshida et al. | A tomato recognition method for harvesting with robots using point clouds | |
| CN113920190A (en) | Ginkgo flower spike orientation method and system | |
| CN115560754A (en) | Visual navigation method based on weed removal | |
| Rao | Design of automatic cotton picking robot with Machine vision using Image Processing algorithms | |
| Rajendran et al. | Towards autonomous selective harvesting: A review of robot perception, robot design, motion planning and control | |
| Jin et al. | Far-near combined positioning of picking-point based on depth data features for horizontal-trellis cultivated grape | |
| Wang et al. | A research for intelligent cotton picking robot based on machine vision | |
| Parsa et al. | Modular autonomous strawberry picking robotic system | |
| Chen et al. | Research on a parallel robot for tea flushes plucking | |
| Au et al. | The Monash Apple retrieving system: a review on system intelligence and apple harvesting performance | |
| Liu et al. | The Vision-Based Target Recognition, Localization, and Control for Harvesting Robots: A Review | |
| Feng et al. | Design and test of harvesting robot for table-top cultivated strawberry |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161012 |
|
| RJ01 | Rejection of invention patent application after publication |