CN114779692A - Linear sliding table type weeding robot and control method thereof - Google Patents
Linear sliding table type weeding robot and control method thereof Download PDFInfo
- Publication number
- CN114779692A CN114779692A CN202210409421.4A CN202210409421A CN114779692A CN 114779692 A CN114779692 A CN 114779692A CN 202210409421 A CN202210409421 A CN 202210409421A CN 114779692 A CN114779692 A CN 114779692A
- Authority
- CN
- China
- Prior art keywords
- weeding
- transmission shaft
- actuator
- robot
- carrying platform
- 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
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
- G05B19/042—Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
- G05B19/0423—Input/output
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01M—CATCHING, TRAPPING OR SCARING OF ANIMALS; APPARATUS FOR THE DESTRUCTION OF NOXIOUS ANIMALS OR NOXIOUS PLANTS
- A01M21/00—Apparatus for the destruction of unwanted vegetation, e.g. weeds
- A01M21/02—Apparatus for mechanical destruction
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/25—Pc structure of the system
- G05B2219/25257—Microcontroller
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mechanical Engineering (AREA)
- Insects & Arthropods (AREA)
- Pest Control & Pesticides (AREA)
- Wood Science & Technology (AREA)
- Zoology (AREA)
- Environmental Sciences (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Soil Working Implements (AREA)
Abstract
The invention provides a linear sliding table type weeding robot and a control method thereof, relating to the technical field of intelligent agricultural machinery, and comprising the following steps: the device comprises an intelligent controller, a linear sliding table, a transmission shaft rotating mechanism, a device carrying platform, a solar energy storage and energy supply system, a transmission shaft lifting control mechanism, a transmission shaft, an actuator carrying platform, a weeding actuator, a coding motor, a Mecanum wheel, a platform lifting support and a camera group; the weeding robot solves the technical problems that most existing weeding robots can only pull out weeds and are limited in the aspects of freedom degree, efficiency, inter-plant weeding and the like.
Description
Technical Field
The invention relates to the technical field of intelligent agricultural machinery, in particular to a linear sliding table type weeding robot and a control method thereof.
Background
As the most important part of human history, agriculture is always valued by people, with the rapid development of modern science and technology, a large amount of mechanical equipment is applied to each link in daily agricultural production, so that the time and labor cost are greatly saved, and modern agricultural machinery is gradually developed towards intellectualization, refinement and greening.
Weeds have great influence on the growth of crops, the weeds have strong life, are easy to reproduce and have high growth speed, compete for nutrients and living space with the crops, spread plant diseases and insect pests, influence the growth and development of the crops, reduce the yield and quality of the crops if the weeds are not prevented and controlled in time, bring great loss to agricultural production and have negative influence on economic return of farmers.
Common weeding methods are: firstly, artificial weeding can accurately identify weeds and accurately remove the weeds, but a large amount of manpower and time investment cannot meet the requirements of the modern society; biological weeding, stable and durable effect and no pollution, but strict application environment, harsh conditions and difficult large-scale popularization; chemical weeding, such as common herbicide paraquat, is efficient and timely, but easily causes toxic substances to diffuse, pollutes the environment and damages human health.
Mechanical weeding is a weeding method widely accepted by people after chemical agent weeding, and has important research significance and wide application prospect. At present, China is actively promoting intelligent agriculture, and with the continuous development of modern scientific technology and artificial intelligence, most of the existing weeding robots can only pull out weeds, and are limited in the aspects of freedom, efficiency, inter-plant weeding and the like.
The linear sliding table type weeding robot provided by the invention can efficiently weed among plants in multiple degrees of freedom and green environment.
Disclosure of Invention
Technical problem to be solved
Aiming at the defects of the prior art, the invention provides a linear sliding table type weeding robot, which solves the problems that most of the existing weeding robots in the background art can only pull out weeds and are limited in the aspects of freedom degree, efficiency, inter-plant weeding and the like.
(II) technical scheme
In order to achieve the purpose, the invention is realized by the following technical scheme: a linear slide table type weeding robot comprising: the device comprises an intelligent controller 1, a linear sliding table 2, a transmission shaft rotating mechanism 3, a device carrying platform 4, a solar energy storage and energy supply system 5, a transmission shaft lifting control mechanism 6, a transmission shaft 7, an actuator carrying platform 8, a weeding actuator 9, a coding motor 10, a Mecanum wheel 11, a platform lifting support 12 and a camera group 13;
the intelligent controller 1 is arranged in front of the carrying platform 4;
four platform lifting brackets 12 are respectively arranged at the bottoms of four corners of the device carrying platform 4;
a Mecanum wheel 11 is arranged at the bottom of each platform lifting support 12, and each Mecanum wheel is driven and controlled by a coding motor 10;
the camera group 13 is arranged below the intelligent controller 1 and on the side surface of the device carrying platform 4, and is connected with the intelligent controller 1 to acquire real-time data;
the solar energy storage and supply system 5 is arranged above the carrying platform 4;
the linear sliding table 2 is arranged below the carrying platform 4;
the transmission shaft rotating mechanism 3 is arranged on the working position of the linear sliding table 2;
the transmission shaft 7 is arranged below the transmission shaft rotating mechanism 3;
the transmission shaft lifting control mechanism 6 is arranged at the middle end of the transmission shaft and controls the transmission shaft 7 to move up and down;
the actuator carrying platform 8 is arranged below the transmission shaft;
the weeding actuator 9 is installed below the actuator mounting platform 8.
Preferably, the linear slide 2 includes: a working position 2.1, a spindle rotating motor 2.2 and a screw rod 2.3; the working position 2.1 is arranged on the screw rod 2.3; and the spindle rotating motor 2.2 receives a control command from the intelligent controller 1 and controls the working position 2.1 to move to a corresponding position.
Preferably, the transmission shaft 7 includes: a main transmission shaft 7.1, a limiter 7.2 and an auxiliary transmission shaft 7.3.
Preferably, the transmission shaft lifting control mechanism 6 consists of a ball screw pair, a limiter 7.2 and a hydraulic electric brake, controls the auxiliary transmission shaft 7.3 fixed by the limiter 7.2 to lift, and simultaneously is connected to the actuator carrying platform 8 so as to adjust the height of the actuator carrying platform 8.
Preferably, the weeding actuator 9 includes: a cutter head steering motor 9.1, a weeding actuating device 9.2, a cutter 9.3 and an actuator camera 9.4;
a tool bit steering motor 9.1 controls the tool 9.3 to rotate and adjusts the normal direction of the tool surface;
the weeding actuating device 9.2 is fixedly provided with a cutter 9.3;
and the actuator camera 9.4 is connected with the intelligent controller 1 to control the tool bit steering motor to rotate.
Preferably, a light source group comprising a plurality of LED lamp beads is further installed below the device carrying platform 4; LED lamp pearl evenly distributed carries on the platform at the device.
The invention also provides a linear sliding table type solar weeding robot control method, which is based on the weeding robot as claimed in any one of the preceding claims, and comprises the following steps:
constructing a weed identification system based on a deep neural network identification technology;
starting real-time positioning and map construction in an operation system of the weeding robot;
acquiring the current position, distance and environmental information data of a weeding execution part of the weeding robot;
drawing a grid map in a weeding robot operating system, sensing the environment by utilizing an SLAM algorithm and a visual odometer of an RGB-D depth camera, and fusing to generate a three-dimensional point cloud map;
correcting the three-dimensional point cloud map when a path forms a loop based on an RTK-GNSS technology, and calculating the path of the robot;
calculating the current position of the weeding robot, matching coordinates in the three-dimensional point cloud map, and controlling the intelligent weeding robot to move to a target position according to the given target point position;
in the walking process of the weeding robot, the camera group collects real-time data and transmits the real-time data to the intelligent controller, and when the intelligent controller catches weeds with the characteristics of the weed recognition system, the position of the weeds is determined, and the position data of the weeds is transmitted back to the intelligent controller;
the intelligent controller issues control instructions to the linear sliding table, the transmission shaft rotating mechanism, the transmission shaft lifting mechanism and the weeding actuator, the weeding actuator is moved to a working position required by weeding work and weeding operation is executed, meanwhile, the central control table controls the coding motor to stop rotating so that the Mecanum wheel stops rotating, the robot is static, and the weeding action is finished;
when the weeding action at one position is finished, the linear sliding table, the transmission shaft rotating mechanism, the transmission shaft lifting mechanism and the weeding actuator return to the initial position, and the weeding robot continues to advance and continue to acquire the next weed data.
Preferably, the weed identification system is constructed based on the deep neural network identification technology, and comprises the following steps:
acquiring an initial detection image data set, and shooting weed pictures in real time or acquiring various weed pictures with different types, colors and sizes from the Internet, wherein each image comprises weed targets and the backgrounds of the images can be different;
carrying out data preprocessing on the initial detection image data set to obtain a training image for target detection, marking weeds to be identified, and creating different weed categories to be used as final image contrast identification;
randomly rotating, randomly cutting, horizontally turning, vertically turning, deforming and scaling, performing color transformation on the images in the initial detection image data set, and recombining the images to obtain a target combined image;
and training a preset neural network model based on the training image, and extracting a plurality of characteristic images of the training image by the neural network training to obtain a corresponding target detection model.
Advantageous effects
The invention provides a linear sliding table type weeding robot and a control method thereof. The method has the following beneficial effects:
the combination of the linear sliding table, the transmission shaft rotating mechanism, the transmission shaft lifting mechanism and the weeding actuator carried on the weeding machine realizes the accurate interplant weeding with multiple degrees of freedom, and has convenient use and high efficiency.
Drawings
Fig. 1 is a perspective view of a linear sliding table type weeding robot provided by the invention;
FIG. 2 is a front view of a bottom structure of a linear sliding table type weeding robot provided by the invention;
FIG. 3 is a front view of a top view of a linear sliding table type weeding robot provided by the present invention;
fig. 4 is a schematic perspective view of a linear sliding table of the linear sliding table type weeding robot provided by the invention;
fig. 5 is a schematic perspective view of a weeding actuator of a linear sliding table type weeding robot provided by the invention;
FIG. 6 is a flow chart of a method for controlling a linear sliding table type weeding robot provided by the invention;
in the figure: the device comprises an intelligent controller 1, a linear sliding table 2, a transmission shaft rotating mechanism 3, a device carrying platform 4, a solar energy storage energy supply system 5, a transmission shaft lifting control mechanism 6, a transmission shaft 7, an actuator carrying platform 8, a weeding actuator 9, a coding motor 10, a Mecanum wheel 11, a platform lifting support 12, a camera group 13, a working position 2.1, a main shaft rotating motor 2.2, a screw rod 2.3, a main transmission shaft 7.1, a limiter 7.2, an auxiliary transmission shaft 7.3, a tool bit steering motor 9.1, a weeding actuator 9.2, a tool bit 9.3 and an actuator camera 9.4.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
As shown in fig. 1 to 5, a linear-sliding-table type weeding robot comprises: the device comprises an intelligent controller 1, a linear sliding table 2, a transmission shaft rotating mechanism 3, a device carrying platform 4, a solar energy storage and energy supply system 5, a transmission shaft lifting control mechanism 6, a transmission shaft 7, an actuator carrying platform 8, a weeding actuator 9, a coding motor 10, a Mecanum wheel 11, a platform lifting support 12 and a camera group 13;
the intelligent controller 1 is arranged in front of the carrying platform 4;
four platform lifting brackets 12 are respectively arranged at the bottoms of four corners of the device carrying platform 4;
a Mecanum wheel 11 is arranged at the bottom of each platform lifting support 12, and each Mecanum wheel is driven and controlled by a coding motor 10;
the weeding robot realizes flexible maneuvering of all-directional movement, and is suitable for agricultural environment by adopting Mecanum wheels to drive and control through a steering motor and a coding motor and coordinating the speed and the rotation angle of each wheel during steering;
the camera group 13 is arranged below the intelligent controller 1 and on the side surface of the device carrying platform 4, and is connected with the intelligent controller 1 to acquire real-time data;
the solar energy storage and supply system 5 is arranged above the carrying platform 4;
the solar energy storage and supply system 5 comprises: the solar energy panel, the light source converter, the sensor, the voltage boosting and stabilizing circuit and the lens coupling system; the electric energy collected by the solar panel is charged to the storage battery through the controller; the electric energy of the storage battery supplies power to all parts of the weeding robot through an inverter.
The solar energy is used for driving the vehicle to move, so that the vehicle is not limited by insufficient electric quantity of a wired power supply and a storage battery;
the linear sliding table 2 is arranged below the carrying platform 4;
the transmission shaft rotating mechanism 3 is arranged on the working position of the linear sliding table 2;
the transmission shaft 7 is arranged below the transmission shaft rotating mechanism 3;
the transmission shaft lifting control mechanism 6 is arranged at the middle end of the transmission shaft and controls the transmission shaft 7 to move up and down;
the actuator carrying platform 8 is arranged below the transmission shaft;
the weeding actuator 9 is installed below the actuator carrying platform 8.
Preferably, the linear slide 2 includes: a working position 2.1, a spindle rotating motor 2.2 and a screw rod 2.3; the working position 2.1 is arranged on the screw rod 2.3; and the spindle rotating motor 2.2 receives a control command from the intelligent controller 1 and controls the working position 2.1 to move to a corresponding position.
Preferably, the transmission shaft 7 includes: a main transmission shaft 7.1, a limiter 7.2 and an auxiliary transmission shaft 7.3.
Preferably, the transmission shaft lifting control mechanism 6 consists of a ball screw pair, a limiter 7.2 and a hydraulic electric brake, controls the lifting of the auxiliary transmission shaft 7.3 fixed by the limiter 7.2, and simultaneously connects the auxiliary transmission shaft to the actuator carrying platform 8 so as to adjust the height of the actuator carrying platform 8; the intelligent control module 1 is combined to control the front and back movement of the vehicle body, so that the weeding actuator 9 moves in a three-dimensional space with multiple degrees of freedom.
Preferably, the weeding actuator 9 includes: a cutter head steering motor 9.1, a weeding actuating device 9.2, a cutter 9.3 and an actuator camera 9.4;
the tool bit steering motor 9.1 controls the tool 9.3 to rotate and adjusts the normal direction of the tool face;
the weeding actuating device 9.2 is fixedly provided with a cutter 9.3;
and the actuator camera 9.4 is communicated with the intelligent controller 1 to control the tool bit steering motor to rotate.
Preferably, a light source group comprising a plurality of LED lamp beads is further installed below the device carrying platform 4; LED lamp pearl evenly distributed is at device carrying platform.
When the device works at night or the illumination intensity is weak, the intelligent controller 1 issues a control instruction to the light source group fixed under the device carrying platform 4 according to the data transmitted back by the photosensitive sensor, controls the number of the LED lamps to be lightened, and provides a supplementary light source for the device.
The intelligent controller 1 executes a weed identification and removal algorithm in an off-line mode, the intelligent controller 1 is controlled by an upper computer and a lower computer, the lower computer receives data returned by a sensor and a camera, then the data are sent to an upper computer through a CANbus protocol, the upper computer calculates and judges the state of the device through an algorithm set in advance on the upper computer after receiving the data, and then the data are sent to the lower computer through the CANbus protocol to be used by the lower computer to control a weeding actuator to weed.
The working principle is as follows: when the linear sliding table weeding robot walks in a working area, the intelligent controller identifies weeds through the camera group and sends weed position posture information and crop position posture information to the intelligent controller; the intelligent control module sends an instruction, and the direction and the angle of the transmission shaft are adjusted through the main shaft rotating mechanism; after the angle and the direction of the transmission shaft are adjusted, the lifting control mechanism adjusts the position of the actuator in the vertical direction; the camera of the actuator judges whether the vehicle body is in a proper distance position or not and sends information to the intelligent control module, and the intelligent control module controls the vehicle body to move back and forth to a proper position; after the actuator moves to a proper position, the mechanical weeding actuator is adjusted by the attitude of the actuator and the angle adjusting mechanism, the normal direction of the knife face is enabled, the intelligent control module sends an instruction to clamp the knife head, and weed removal is carried out by the weeding action.
As shown in fig. 6, an embodiment of the present invention further provides a method for controlling a linear sliding table type solar weeding robot, based on the weeding robot as claimed in any one of the preceding claims, the method including:
s1, constructing a weed identification system based on the deep neural network identification technology;
s2, starting real-time positioning and map construction in the weeding robot operating system;
s3, acquiring the current position, distance and environmental information data of the weeding execution part of the weeding robot;
s4, drawing a grid map in the weeding robot operating system, sensing the environment by utilizing an SLAM algorithm and a visual odometer of an RGB-D depth camera, and fusing to generate a three-dimensional point cloud map;
s5, correcting the three-dimensional point cloud map when a path forms a loop based on an RTK-GNSS technology, and calculating the path of the robot;
s6, calculating the current position of the weeding robot, matching the coordinates in the three-dimensional point cloud map, and controlling the intelligent weeding robot to move to the target position according to the given target point position;
s7 the weeding robot walks, the camera group collects real-time data and transmits the real-time data to the intelligent controller, and when the intelligent controller catches weeds with characteristics consistent with those of the weed recognition system, the position of the weeds is determined, and the position data of the weeds is transmitted back to the intelligent controller;
s8 the intelligent controller issues control instructions to the linear sliding table, the transmission shaft rotating mechanism, the transmission shaft lifting mechanism and the weeding actuator, the weeding actuator is moved to a working position required by weeding work and weeding operation is executed, meanwhile, the center console controls the coding motor to stop rotating the Mecanum wheel, the robot is static, and the weeding action is completed;
s9 when the weeding action at one position is finished, the linear sliding table, the transmission shaft rotating mechanism, the transmission shaft lifting mechanism and the weeding actuator return to the initial position, and the weeding robot continues to move forward and continue to collect the weed data at the next position.
Preferably, the weed identification system is constructed based on the deep neural network identification technology, and comprises the following steps:
acquiring an initial detection image data set, and shooting weed pictures in real time or acquiring various weed pictures with different types, colors and sizes from the internet, wherein each image comprises weed targets and the backgrounds of the images can be different;
carrying out data preprocessing on the initial detection image data set to obtain a training image for target detection, marking weeds to be identified, and creating different weed categories to be used as final image contrast identification;
randomly rotating, randomly cutting, horizontally turning, vertically turning, deforming and scaling, performing color transformation on the images in the initial detection image data set, and recombining the images to obtain a target combined image;
and training a preset neural network model based on the training image, and extracting a plurality of characteristic images of the training image by the neural network training to obtain a corresponding target detection model.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (8)
1. A linear slider type weeding robot, comprising: the device comprises an intelligent controller (1), a linear sliding table (2), a transmission shaft rotating mechanism (3), a device carrying platform (4), a solar energy storage and energy supply system (5), a transmission shaft lifting control mechanism (6), a transmission shaft (7), an actuator carrying platform (8), a weeding actuator (9), a coding motor (10), a Mecanum wheel (11), a platform lifting support (12) and a camera group (13);
the intelligent controller (1) is arranged in front of the carrying platform (4);
four platform lifting brackets (12) are respectively arranged at the bottoms of four corners of the device carrying platform (4);
the bottom of each platform lifting support (12) is provided with a Mecanum wheel (11), and each Mecanum wheel is driven and controlled by an encoding motor (10);
the camera group (13) is arranged below the intelligent controller (1) and on the side surface of the device carrying platform (4);
the solar energy storage and supply system (5) is arranged above the carrying platform (4);
the linear sliding table (2) is arranged below the carrying platform (4);
the transmission shaft rotating mechanism (3) is arranged on a working position of the linear sliding table (2);
the transmission shaft (7) is arranged below the transmission shaft rotating mechanism (3);
the transmission shaft lifting control mechanism (6) is arranged at the middle end of the transmission shaft and controls the transmission shaft (7) to move up and down;
the actuator carrying platform (8) is arranged below the transmission shaft;
the weeding actuator (9) is arranged below the actuator carrying platform (8).
2. A linear slider-type herbicidal robot as set forth in claim 1, wherein: the linear sliding table (2) comprises: a working position (2.1), a spindle rotating motor (2.2) and a screw rod (2.3);
the working position (2.1) is arranged on the screw rod (2.3);
and the spindle rotating motor (2.2) receives a control command obtained by the intelligent controller (1) and controls the working position (2.1) to move to a corresponding position.
3. A linear slide table type weeding robot according to claim 2, wherein: the transmission shaft (7) comprises: a main transmission shaft (7.1), a limiter (7.2) and an auxiliary transmission shaft (7.3).
4. A linear slide table type weeding robot according to claim 3, wherein: the transmission shaft lifting control mechanism (6) consists of a ball screw pair, a limiter (7.2) and a hydraulic electric appliance brake, controls the auxiliary transmission shaft (7.3) with the posture fixed by the limiter (7.2) to lift, and simultaneously is connected to the actuator carrying platform (8), so that the height of the actuator carrying platform (8) is adjusted.
5. A linear slide table type weeding robot according to claim 4, wherein: the weeding actuator (9) comprises: a cutter head steering motor (9.1), a weeding actuating device (9.2), a cutter (9.3) and an actuator camera (9.4);
a tool bit steering motor (9.1) controls the tool (9.3) to rotate and adjusts the normal direction of the tool surface;
the weeding actuating device (9.2) is fixedly provided with a cutter (9.3);
the actuator camera (9.4) is communicated with the intelligent controller (1) to control the tool bit to rotate to the motor.
6. A linear slide table type weeding robot according to claim 1, wherein: a light source group comprising a plurality of LED lamp beads is also arranged below the device carrying platform (4); LED lamp pearl evenly distributed is at device carrying platform.
7. A control method of a linear sliding table type solar weeding robot based on the weeding robot as claimed in any one of claims 1-6, wherein the method comprises the following steps:
constructing a weed identification system based on a deep neural network identification technology;
starting real-time positioning and map construction in an operation system of the weeding robot;
acquiring the current position, distance and environmental information data of a weeding execution part of the weeding robot;
drawing a grid map in a weeding robot operating system, sensing the environment by utilizing an SLAM algorithm and a visual odometer of an RGB-D depth camera, and fusing to generate a three-dimensional point cloud map;
correcting the three-dimensional point cloud map when a path forms a loop based on an RTK-GNSS technology, and calculating the path of the robot;
calculating the current position of the weeding robot, matching coordinates in the three-dimensional point cloud map, and controlling the intelligent weeding robot to move to a target position according to the given target point position;
in the walking process of the weeding robot, the camera group collects real-time data and transmits the real-time data to the intelligent controller, and when the intelligent controller catches weeds with the characteristics of the weed recognition system, the position of the weeds is determined, and the position data of the weeds is transmitted back to the intelligent controller;
the intelligent controller issues control instructions to the linear sliding table, the transmission shaft rotating mechanism, the transmission shaft lifting mechanism and the weeding actuator, the weeding actuator is moved to a working position required by weeding work and weeding operation is executed, meanwhile, the central control table controls the coding motor to stop rotating so that the Mecanum wheel stops rotating, the robot is static, and the weeding action is finished;
when the weeding action at one position is finished, the linear sliding table, the transmission shaft rotating mechanism, the transmission shaft lifting mechanism and the weeding actuator return to the initial position, and the weeding robot continues to advance and continue to acquire the next weed data.
8. The linear sliding table type solar weeding robot control method according to claim 7, wherein the weed identification system is constructed based on the deep neural network identification technology, and comprises the following steps:
acquiring an initial detection image data set, and shooting weed pictures in real time or acquiring various weed pictures with different types, colors and sizes from the Internet, wherein each image comprises weed targets and the backgrounds of the images can be different;
carrying out data preprocessing on the initial detection image data set to obtain a training image for target detection, marking weeds to be identified, and creating different weed categories to be used as final image contrast identification;
randomly rotating, randomly cutting, horizontally turning, vertically turning, deforming and scaling, performing color transformation on the images in the initial detection image data set, and recombining the images to obtain a target combined image;
and training a preset neural network model based on the training image, and extracting a plurality of characteristic graphs of the training image through the neural network training to obtain a corresponding target detection model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210409421.4A CN114779692A (en) | 2022-04-19 | 2022-04-19 | Linear sliding table type weeding robot and control method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210409421.4A CN114779692A (en) | 2022-04-19 | 2022-04-19 | Linear sliding table type weeding robot and control method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114779692A true CN114779692A (en) | 2022-07-22 |
Family
ID=82430133
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210409421.4A Pending CN114779692A (en) | 2022-04-19 | 2022-04-19 | Linear sliding table type weeding robot and control method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114779692A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115413472A (en) * | 2022-09-22 | 2022-12-02 | 珠海格力电器股份有限公司 | Weeding machine control method and device, storage medium and weeding machine |
| CN117649409A (en) * | 2024-01-29 | 2024-03-05 | 深圳市锐健电子有限公司 | Automatic limiting system, method, device and medium for sliding table based on machine vision |
-
2022
- 2022-04-19 CN CN202210409421.4A patent/CN114779692A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115413472A (en) * | 2022-09-22 | 2022-12-02 | 珠海格力电器股份有限公司 | Weeding machine control method and device, storage medium and weeding machine |
| CN117649409A (en) * | 2024-01-29 | 2024-03-05 | 深圳市锐健电子有限公司 | Automatic limiting system, method, device and medium for sliding table based on machine vision |
| CN117649409B (en) * | 2024-01-29 | 2024-04-02 | 深圳市锐健电子有限公司 | Automatic limiting system, method, device and medium for sliding table based on machine vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114779692A (en) | Linear sliding table type weeding robot and control method thereof | |
| CN110421580B (en) | Intelligent field robot and working method thereof | |
| CN101412217B (en) | Autonomous field robot | |
| CN102239756B (en) | Intelligent farming robot in greenhouse | |
| CN202818996U (en) | Grapevine pruning robot system | |
| CN209527208U (en) | A kind of strawberry picking machine | |
| CN110419315B (en) | Foreign matter removing device and control method thereof | |
| CN205983216U (en) | Agricultural information acquisition robot of new forms of energy | |
| CN109964905B (en) | Self-propelled targeting pesticide application robot based on fruit tree identification and positioning and control method thereof | |
| WO2022021804A1 (en) | Underwater robot device and underwater regulation and control management optimization system and method | |
| CN113057154A (en) | Greenhouse liquid medicine spraying robot | |
| CN113396650B (en) | Intelligent vegetable inter-plant weeding device and control method | |
| CN107509399A (en) | A kind of green intelligent weed-eradicating robot | |
| CN109964675A (en) | Vine beta pruning robot device | |
| CN207269304U (en) | A kind of green intelligent weed-eradicating robot | |
| CN109964558A (en) | A kind of Small plastic shed inside intelligence weeding gapper and application method | |
| CN113467453A (en) | Inspection robot and method for controlling inspection robot to run based on fuzzy PID | |
| CN111750777A (en) | Self-propelled crop phenotype high-flux detection device | |
| Hu et al. | Design and experiment of a new citrus harvesting robot | |
| CN115648163A (en) | Weeding robot with weed recognition function | |
| CN115407771A (en) | Crop monitoring method, system and device based on machine vision | |
| CN210900392U (en) | Foreign matter removing device | |
| CN114700927A (en) | Combined weeding robot with parallel mechanical arms and flexible mechanical arm and control method | |
| Tang et al. | A new robot system for harvesting cucumber | |
| CN114700969A (en) | Weeding robot based on XY I-shaped sliding table parallel flexible manipulator and control method |
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 |