CN109122633B - Plant protection unmanned aerial vehicle accurate variable spraying device for neural network decision-making and control method - Google Patents
Plant protection unmanned aerial vehicle accurate variable spraying device for neural network decision-making and control method Download PDFInfo
- Publication number
- CN109122633B CN109122633B CN201810660859.3A CN201810660859A CN109122633B CN 109122633 B CN109122633 B CN 109122633B CN 201810660859 A CN201810660859 A CN 201810660859A CN 109122633 B CN109122633 B CN 109122633B
- Authority
- CN
- China
- Prior art keywords
- neural network
- spraying
- information
- controller
- unmanned aerial
- 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
- 238000005507 spraying Methods 0.000 title claims abstract description 106
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 65
- 238000000034 method Methods 0.000 title claims abstract description 40
- 239000007788 liquid Substances 0.000 claims abstract description 68
- 239000003814 drug Substances 0.000 claims abstract description 60
- 239000007921 spray Substances 0.000 claims abstract description 16
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 11
- 230000006835 compression Effects 0.000 claims abstract description 8
- 238000007906 compression Methods 0.000 claims abstract description 8
- 241000607479 Yersinia pestis Species 0.000 claims description 37
- 201000010099 disease Diseases 0.000 claims description 36
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 36
- 238000010586 diagram Methods 0.000 claims description 28
- 238000012549 training Methods 0.000 claims description 25
- 210000002569 neuron Anatomy 0.000 claims description 24
- 238000009826 distribution Methods 0.000 claims description 20
- 241000238631 Hexapoda Species 0.000 claims description 17
- RZVHIXYEVGDQDX-UHFFFAOYSA-N 9,10-anthraquinone Chemical compound C1=CC=C2C(=O)C3=CC=CC=C3C(=O)C2=C1 RZVHIXYEVGDQDX-UHFFFAOYSA-N 0.000 claims description 15
- 230000007613 environmental effect Effects 0.000 claims description 11
- 230000003321 amplification Effects 0.000 claims description 8
- 230000010365 information processing Effects 0.000 claims description 8
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 5
- 238000012377 drug delivery Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000013135 deep learning Methods 0.000 claims description 3
- 210000002364 input neuron Anatomy 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000001105 regulatory effect Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 4
- 239000010914 pesticide waste Substances 0.000 abstract description 2
- 229940079593 drug Drugs 0.000 description 3
- 238000005259 measurement Methods 0.000 description 2
- 238000012271 agricultural production Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 230000008021 deposition Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000003595 mist Substances 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Classifications
-
- 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
- A01M7/00—Special adaptations or arrangements of liquid-spraying apparatus for purposes covered by this subclass
- A01M7/0089—Regulating or controlling systems
-
- 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
- A01M7/00—Special adaptations or arrangements of liquid-spraying apparatus for purposes covered by this subclass
- A01M7/0025—Mechanical sprayers
- A01M7/0032—Pressure sprayers
-
- 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
- A01M7/00—Special adaptations or arrangements of liquid-spraying apparatus for purposes covered by this subclass
- A01M7/0082—Undercarriages, frames, mountings, couplings, tanks
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B64—AIRCRAFT; AVIATION; COSMONAUTICS
- B64D—EQUIPMENT FOR FITTING IN OR TO AIRCRAFT; FLIGHT SUITS; PARACHUTES; ARRANGEMENT OR MOUNTING OF POWER PLANTS OR PROPULSION TRANSMISSIONS IN AIRCRAFT
- B64D1/00—Dropping, ejecting, releasing, or receiving articles, liquids, or the like, in flight
- B64D1/16—Dropping or releasing powdered, liquid, or gaseous matter, e.g. for fire-fighting
- B64D1/18—Dropping or releasing powdered, liquid, or gaseous matter, e.g. for fire-fighting by spraying, e.g. insecticides
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Pest Control & Pesticides (AREA)
- Insects & Arthropods (AREA)
- Wood Science & Technology (AREA)
- Zoology (AREA)
- Environmental Sciences (AREA)
- Aviation & Aerospace Engineering (AREA)
- Mechanical Engineering (AREA)
- Catching Or Destruction (AREA)
Abstract
The invention discloses a plant protection unmanned aerial vehicle accurate variable spraying device for neural network decision-making, wherein a brushless direct current motor is connected with the end part of a horn and is connected with a rotor blade, a medicine box is connected with a spray head through a compression-resistant water pipe, the compression-resistant water pipe is connected with a liquid pump, and the spray head sprays liquid medicine under the action of the liquid pump; the positioning device is arranged in the middle of the horn, the controller and the mini-environment information collector are further arranged at the top of the horn, and the controller is further connected with the positioning device, the mini-environment information collector and the hyperspectral camera arranged at the bottom end of the brushless direct current motor through the multi-channel data collector. Compared with the prior art, the control method for the precise variable spraying device of the plant protection unmanned aerial vehicle, which is based on the neural network decision, comprehensively considers various factors influencing the spraying quantity during the spraying operation of the plant protection unmanned aerial vehicle, and has the advantages of precisely adjusting the spraying flow according to the prescription value, and the like, effectively reducing pesticide waste and achieving the best control effect through precise variable spraying.
Description
Technical Field
The invention relates to the technical field of agricultural plant protection, in particular to a plant protection unmanned aerial vehicle accurate variable spraying device for neural network decision-making and a control method.
Background
The plant protection unmanned aerial vehicle of prior art sets up the spraying volume in advance through the pest and disease circumstances of analysis whole operation plot when carrying out the operation, in the actual flight operation, controls instantaneous flow according to unmanned aerial vehicle flight speed and height to guarantee the homogeneity of spraying.
However, the distribution conditions of the same land block and the disease and insect pests are different, and meanwhile, when the plant protection unmanned aerial vehicle is in spraying operation, the sprayed fog drops are evaporated, attached to the surface of a machine body or scattered in the air and the like due to the influences of factors such as an operation environment, a rotor wing wind field and the like, in addition, the existing technical system of spraying operation of the plant protection unmanned aerial vehicle is imperfect and has no unified standard, unmanned aerial vehicles of different plant protection unmanned aerial vehicle companies are different in structural parameters, and under the condition that the operation environment parameters, flight parameters, prescription values and the like are the same, the spraying operation fog drop deposition conditions of the plant protection unmanned aerial vehicles of different models are also different, so that the method for deciding the spraying quantity of the unmanned aerial vehicle in the prior art cannot meet the requirements of accurate agricultural production.
Disclosure of Invention
The invention mainly aims to provide a neural network decision-making plant protection unmanned aerial vehicle accurate variable spraying device capable of accurately adjusting spraying flow, and also provides a control method of the plant protection unmanned aerial vehicle accurate variable spraying device using the neural network decision-making, aiming at accurately adjusting spraying flow on line according to the severity degree of disease and insect pests in an operation area, flight parameters, plant protection unmanned aerial vehicle specification parameters, environmental factors and the like, effectively reducing pesticide waste and accurate variable spraying and achieving the best control effect.
In order to achieve the aim, the accurate variable spraying device of the plant protection unmanned aerial vehicle for neural network decision-making comprises a plurality of machine arms, wherein the end parts of the machine arms are connected with brushless direct current motors, the top ends of the brushless direct current motors are connected with rotor blades, a medicine box is connected below a machine body formed by the machine arms, the bottom of the medicine box is connected with a spray head through a compression-resistant water pipe, the compression-resistant water pipe is connected with a liquid pump, and the spray head is arranged on a spray rod below the machine body; the intelligent robot arm comprises a robot arm body, wherein a plurality of positioning devices are arranged in the middle of the robot arm body, a controller and a micro environment information collector for collecting environment parameters are arranged at the top of the robot body, and the controller is further connected with the positioning devices, the micro environment information collector and a hyperspectral camera arranged at the bottom end of the brushless direct current motor through a multi-channel data collector.
Preferably, a drive amplification module is mounted on the top of the machine body and is respectively connected with the controller and the liquid pump.
Preferably, the liquid pump is communicated with a filter arranged at the bottom of the medicine box through a pipeline.
Preferably, the pressure-resistant water pipe is further provided with an overflow valve, a flow sensor and a pressure sensor in sequence along the flowing direction of the liquid medicine, and the overflow valve, the flow sensor and the pressure sensor are respectively connected with the controller.
Preferably, a liquid level sensor for detecting the liquid level of the liquid medicine is arranged at the top of the inner wall surface of the medicine box and is connected with the controller.
The invention also provides a control method using the spraying device, which comprises the following steps:
s1: the hyperspectral camera is arranged at the lower part of the horn to collect crop image information of an operation land block, the positioning device acquires position information in real time, the mini-type environment information collector arranged at the top of the machine body is used for collecting environment parameters in the field, and the hyperspectral camera, the positioning device and the mini-type environment information collector respectively transmit collected data to the controller; the flow sensor, the pressure sensor and the liquid level sensor feed back the flow, the pressure value and the residual liquid medicine to the controller in real time;
s2: the controller processes the acquired data through the multichannel data acquisition device, and an information processing and decision module in the controller processes the information such as images and the like by using a trained neural network module and generates a spatial distribution diagram of the plant diseases and insect pests;
s3: the controller generates a spraying operation prescription chart of a target land block according to the image processing result and by combining with the crop disease and insect pest control expert module to output a spraying medicine value, and the unmanned aerial vehicle flight control module autonomously plans a route by combining with the operation prescription chart;
s4: the controller processes the collected flight parameters, environment parameters, real-time information of the plant protection unmanned aerial vehicle, prescription value information, liquid level, flow, pressure and other information, calculates each decision quantity through a trained neural network module and obtains the optimal spraying quantity; when a prescription map of an operation land block is acquired before operation, the controller extracts current position information and compares the current position information with the prescription map, and judges the current position of the unmanned aerial vehicle so as to acquire information such as a prescription value of the current position;
s5: the variable spraying control module of the controller generates corresponding duty ratio pulse square waves by using a closed-loop PID algorithm according to the received optimal spraying amount information, the driving amplification module is used for driving the liquid pump to work after amplifying, meanwhile, the flow sensor and the pressure sensor feed back actual flow and pressure values in real time, and the flow is accurately regulated again by the controller; the liquid level sensor monitors the residual quantity of the liquid medicine in the medicine box in real time, feeds back the residual quantity of the liquid medicine to the flight control module, and makes a return operation in time when the liquid medicine is lower than a certain set value.
Preferably, in the step S1, the width of the hyperspectral camera collecting operation land is 10 meters, the length of the hyperspectral camera collecting operation land is matched with the flying speed of the plant protection unmanned aerial vehicle, and an operation area distribution information graph is obtained by combining the geographic position information obtained by the positioning device;
the controller processes the image information in the step S2, carries out convolution operation on the image information as input quantity to filter useless information in the image, extracts the characteristics of color, shape, texture and the like of the preprocessed image through an hidden layer of the neural network module, and obtains a disease, insect and pest distribution map of the operation land block through operation output;
in the step S3, a database is established according to various disease and pest control experiences in different crop growth periods and a large amount of field control experimental data, a spraying prescription value is output by the crop disease and pest control expert module according to the spatial distribution situation of the disease and pest after image recognition, then an operation land prescription map is generated by combining the distribution situation of the disease and pest of the operation land and the spraying quantity, the prescription map is obtained before operation, and the flight control module autonomously generates an optimal spraying operation route according to the prescription map;
the spraying amount information, unmanned aerial vehicle specification parameters, flight parameters and environment parameters output by the crop disease, insect and pest control expert module in the step S4 are transmitted to the controller through the multi-channel data acquisition unit; when the prescription diagram is obtained before operation, the prescription diagram is imported into the controller, and in the process of plant protection operation of the plant protection unmanned aerial vehicle, the position of the plant protection unmanned aerial vehicle relative to the prescription diagram is determined by comparing the current position information with the position information of the prescription diagram so as to obtain the spraying amount.
Preferably, the neural network module in the step S4 is a back propagation neural network module, the neural network includes an input layer, an implicit layer and an output layer, sample data required for training the neural network is valid test data of a large number of related tests, before training the neural network, weights and bias are randomly initialized, a range of each weight is set in (-1, 1), an error function e is set, a training maximum learning number and a calculation precision value epsilon are set, a selected sample is used as an input quantity X, and a desired output corresponding to the sample is defined:
the number of the input amounts is as follows: x (k) = (X) 1 (k)，x 2 (k)，…x n (k))
Outputting an expected value: y is Y
And respectively calculating the input and the output of each neuron of the hidden layer, wherein the input quantity of the hidden layer is as follows:
wherein w is i (k) Connecting weights between each input neuron and the hidden layer; b i Neuron bias values for hidden layers; n is the number of input variables;
hidden layer neuron output:
wherein w is o (k) The connection weight between the neuron and the hidden layer is output; b is the output layer neuron bias value;
calculating multiple hidden layers of the neural network to obtain an output value, and calculating the partial derivative delta of the error function on the neurons of the output layer according to the expected value of the output quantity o (k) Wherein the global error is:
in the neural network training process, the weight w of each layer of neurons of the hidden layer is adjusted according to the accumulated error i (k) Thereby reducing global errors, namely:
eta in the formula is learning rate;
adjusting the weight according to the weight adjusting formula of each hidden layer neuron output layer, namely:
when all variables of the previous training period are smaller than a certain designated threshold value, the neural network completes a learning period, the learning process is completed according to the set learning period, and the neural network module after deep learning training can output a reasonable result after model operation according to input information.
Preferably, the step S5 is to build the neural network module meeting the requirements for training in the controller through programming, the information such as the environmental temperature and humidity, wind speed and wind direction information, flight parameters and prescription value in operation are processed and then used as the input quantity of the neural network, the optimal spraying quantity is output after the operation of the neural network, the embedded system can transmit the optimal spraying quantity to the variable spraying controller, the variable spraying control module generates the pulse square wave with corresponding duty ratio according to the received spraying quantity information and drives the liquid pump after amplifying, meanwhile, the flow sensor and the pressure sensor of the spraying loop measure the flow speed and the pressure value of the drug delivery pipeline in real time, and feed back the variable spraying controller module, and combine the deviation among the actual flow speed, the actual pressure value, the theoretical flow speed and the theoretical pressure value to accurately regulate the spraying process, and the liquid level sensor detects the medicine quantity condition in the medicine box, and when the detected medicine liquid is lower than the set value, the flight control module sends a return instruction.
Compared with the prior art, the technical scheme of the invention has the following advantages:
according to the neural network decision-making plant protection unmanned aerial vehicle accurate variable spraying device, when the plant protection unmanned aerial vehicle spraying operation is comprehensively considered, the influence of environmental factors, flight parameters and unmanned aerial vehicle specification parameters on liquid medicine mist drops is comprehensively considered, a large amount of field experiment data are collected to serve as training samples, a decision model is trained, after the unmanned aerial vehicle specification parameters are input in actual operation, the environmental factors, the flight parameters and the medicine application prescription values are monitored in real time through sensors, the obtained information is used as model input quantity, the optimal spraying quantity is obtained through rapid operation, the liquid medicine quantity actually deposited on the surface of crops just meets the prescription value requirement, and the optimal control effect can be achieved.
According to the technical scheme, when the plant protection unmanned aerial vehicle is used for plant protection operation, hyperspectral collection of field disease, insect and pest distribution conditions is completed, an operation prescription chart is generated, decision information collected by various onboard sensors is combined, an optimal spraying amount is rapidly given, and variable spraying operation is guided. Meanwhile, for an operation prescription map of a target land block obtained before operation, the prescription map is imported into an onboard embedded system, when the plant protection unmanned aerial vehicle flies, the operation prescription map is interpreted to obtain a prescription value according to the position information obtained by the real-time positioning module, and an optimal spraying amount is quickly given by combining decision information collected by various onboard sensors to guide variable spraying operation.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained from the structures shown in these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of a plant protection unmanned aerial vehicle precision variable spraying device according to the present invention;
FIG. 2 is a schematic diagram of a system of a plant protection unmanned aerial vehicle precision variable spraying device according to the present invention;
FIG. 3 is a schematic diagram of a spraying operation of the accurate variable spraying device of the plant protection unmanned aerial vehicle;
FIG. 4 is a flow chart of information processing and decision making of the precise variable spraying device of the plant protection unmanned aerial vehicle;
FIG. 5 is a training flow chart of the neural network module of the plant protection unmanned aerial vehicle accurate variable spraying device of the invention.
Reference numerals illustrate:
the achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that, if directional indications (such as up, down, left, right, front, and rear … …) are included in the embodiments of the present invention, the directional indications are merely used to explain the relative positional relationship, movement conditions, etc. between the components in a specific posture (as shown in the drawings), and if the specific posture is changed, the directional indications are correspondingly changed.
In addition, if there is a description of "first", "second", etc. in the embodiments of the present invention, the description of "first", "second", etc. is for descriptive purposes only and is not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions of the embodiments may be combined with each other, but it is necessary to base that the technical solutions can be realized by those skilled in the art, and when the technical solutions are contradictory or cannot be realized, the combination of the technical solutions should be considered to be absent and not within the scope of protection claimed in the present invention.
The invention provides a plant protection unmanned aerial vehicle accurate variable spraying device for neural network decision.
Referring to fig. 1, the plant protection unmanned aerial vehicle accurate variable spraying device for neural network decision-making in the technical scheme of the invention comprises four horizontally arranged horn 17, wherein the end part of the horn 17 is connected with a brushless direct current motor 18, the top end of the brushless direct current motor 18 is connected with a rotor blade 19, a medicine box 8 is connected below a machine body formed by the four horn 17, the bottom of the medicine box 8 is connected with a spray head 13 through a compression-resistant water pipe 12, the compression-resistant water pipe 12 is connected with a liquid pump 10, and the spray head 13 is arranged on a spray boom 14; the top of the machine body is provided with a controller 4 and a micro environment information collector 3 for collecting environment factor data, wherein the micro environment information collector 3 is used for collecting information such as environment temperature, humidity, wind speed, wind direction and the like in the operation process. Two positioning devices 1 are arranged in the middle of the horn 17, wherein the positioning devices 1 adopt a real-time differential positioning method, so that the geographic position of the plant protection unmanned aerial vehicle can be accurately provided, and the positioning accuracy can reach the centimeter level. The controller 4 is connected with the positioning device 1, the mini-environment information collector 3 and the hyperspectral camera 6 which is arranged at the bottom end of the brushless direct current motor 18 and acquires the image information of the operation land by continuous shooting through the multi-channel data collector 2. The positioning device 1, the micro-environmental information collector 3 and the hyperspectral camera 6 transmit the collected position information, environmental parameters and image information to the controller 4 to serve as reference information for guiding the spraying operation subsequently.
In addition, the machine body of the plant protection unmanned aerial vehicle precise variable spraying device for neural network decision-making in the technical scheme of the invention is provided with a driving amplification module 5 which is respectively connected with a controller 4 and a liquid pump 10, the liquid pump 10 is communicated with a filter 9 arranged at the bottom of a medicine box 8 through a pipeline, a pressure-resistant water pipe 12 is sequentially provided with an overflow valve 11, a flow sensor 15 and a pressure sensor 16 along the flowing direction of medicine liquid, the two are respectively connected with the controller 4, and in addition, the top of the inner wall surface of the medicine box 8 is provided with a liquid level sensor 7 for detecting the liquid level of the medicine liquid, and the liquid level sensor 7 is connected with the controller 4.
According to the neural network decision-making plant protection unmanned aerial vehicle accurate variable spraying device, the position of the plant protection unmanned aerial vehicle in flight is detected in real time through the positioning device 1, accurate positioning is carried out on the position, the information of the environment temperature, the humidity, the wind speed, the wind direction and the like acquired in the flight process of the plant protection unmanned aerial vehicle is sent to the controller 4 in real time by the miniature environment information acquisition device 3 to be analyzed and processed so as to improve the precision of spraying liquid medicine of the plant protection unmanned aerial vehicle, and the high-spectrum camera 6 shoots a high-definition image on the working environment and sends the high-definition image to the controller 4 to be analyzed and processed. The liquid pump 10 arranged below the medicine box 8 pumps the liquid medicine out of the medicine box 8 through rotation, flows to the overflow valve 11 through the filter 9, one of the liquid medicine flows back into the medicine box 8 through a channel connected with the medicine box 8 through the overflow valve 11, and the other liquid medicine enters the main loop through the flow sensor 15 and the pressure sensor 16 and then flows to the spray head 13 through the spray bars 14 respectively through the plurality of branch pipes to spray downwards. The flow sensor 15 and the pressure sensor 16 are arranged on the main loop, so that accurate measurement of flow and pressure can be effectively realized, and a measurement result is fed back to the controller 4 in real time, meanwhile, the liquid level sensor 7 is arranged inside the medicine chest 8, the residual quantity of liquid medicine can be detected in real time, and when the liquid medicine quantity is lower than a set value, a flight control module in the controller 4 controls the plant protection unmanned aerial vehicle to carry out return medicine feeding.
Referring to fig. 2 to 5, the controller according to the technical scheme of the present invention adopts an embedded chip, and a functional module is disposed in the embedded chip, including an information processing and decision module, a prescription diagram interpretation module, an fly control module, a variable spray control module, and a crop disease and pest control expert module.
Specifically, the information processing and decision module of the technical scheme of the invention can process the acquired image information, identify the disease, pest and crop in the image, obtain the spatial distribution condition of the disease, pest and crop, and then the disease, pest and pest control expert module of the crop gives out the application prescription values of different areas according to the crop and disease, pest and crop distribution condition after the image identification. In the actual spraying operation process, the flight parameters, the specification parameters and the environmental factors of the plant protection unmanned aerial vehicle can have certain influence on the spraying amount, so that the disease and insect pest control effect of a target area is enhanced, and in the spraying operation process, the information processing and decision module takes the received environmental information, the flight parameters, the prescription value made by the crop disease and insect pest control expert module and the specification parameters of the plant protection unmanned aerial vehicle as the input amount of the neural network module, and the optimal spraying amount is calculated and output through the neural network module.
The training of the neural network module of the embodiment finds out the complex relation between each influencing factor and the result by using the existing experimental data, wherein the neural network module mainly comprises an input layer, an hidden layer and an output layer which are of a total of 3 layers, a large amount of image information, a spraying amount decision factor and a prescription value are respectively selected as sample data during the training, the sample data is used as the input amount of each neuron of the input layer, each hidden layer of the neural network module calculates the characteristics of the sample data, the output value of each hidden layer is obtained, the training process parameters are continuously optimized according to the error between the expected value and the actual output value, and the training result is output when the error is smaller than the set threshold value.
When the prescription diagram is obtained before operation, the prescription diagram is imported into the controller, the position of the plant protection unmanned aerial vehicle relative to the prescription diagram is judged according to the collected position information during operation, and the prescription value of the current position is obtained.
The flight control module can automatically control and maintain the normal flight attitude of the plant protection unmanned aerial vehicle, in the plant protection operation process, an operator sends a starting instruction to the plant protection unmanned aerial vehicle through the handheld ground station, the flight control module automatically plans a route according to the field disease and insect pest distribution condition, and when in flight operation, the flight control module is connected with the information processing and decision module and the variable spraying control module through buses, and the flight control module adjusts the flight attitude according to the actual condition of operation.
In the technical scheme of the invention, after the variable spraying control module receives the optimal spraying amount information, a control program can generate a pulse square wave signal with corresponding duty ratio according to the actual flow value, the error between the pressure value and the target value, the pulse square wave signal is amplified by the driving amplification module to drive the liquid pump to work, and the closed loop control is adopted to realize the accurate control of the flow, so that the variable spraying operation can be realized.
Referring to fig. 2 to 5, the technical scheme of the invention further provides a control method of the plant protection unmanned aerial vehicle accurate variable spraying device using neural network decision, comprising the following steps:
step S1: the hyperspectral camera acquires crop image information of an operation land block, the positioning device acquires position information in real time, the mini-environment information acquisition device acquires environmental parameters of the field, and the hyperspectral camera, the positioning device and the mini-environment information acquisition device respectively transmit acquired data to the controller; the flow sensor, the pressure sensor and the liquid level sensor feed back the flow, the pressure value and the residual quantity of the liquid medicine to the controller in real time;
step S2: the controller processes the acquired data through the multichannel data acquisition device, and an information processing and decision module in the controller processes the information such as images and the like by using a trained neural network module and generates a spatial distribution diagram of the plant diseases and insect pests;
step S3: the controller generates a spraying operation prescription chart of the target land according to the image processing result and by combining the crop disease and insect pest control expert module to output the spraying medicine quantity value, and the unmanned aerial vehicle flight control module of the controller autonomously plans the route by combining the operation prescription chart;
step S4: the controller processes the collected flight parameters, environment parameters, real-time information of the plant protection unmanned aerial vehicle, prescription value information, liquid level, liquid medicine flow, pressure and other information, calculates each decision quantity through a trained neural network module and obtains the optimal spraying quantity; when a prescription map of an operation land block is acquired before operation, the controller extracts current position information and compares the current position information with the prescription map, and judges the current position of the unmanned aerial vehicle so as to acquire information such as a prescription value of the current position;
step S5: the variable spraying control module of the controller generates corresponding duty ratio pulse square waves by using a closed-loop PID algorithm according to the received optimal spraying amount information, the liquid pump is driven to work after the corresponding duty ratio pulse square waves are amplified by the driving amplification module, meanwhile, the flow sensor, the pressure sensor and the liquid level sensor feed back actual flow and pressure values and the residual amount of liquid medicine in the medicine chest in real time, and the flow is accurately regulated again through the controller.
Preferably, in the step S1, the width of the hyperspectral camera collecting operation land is 10 meters, the length of the hyperspectral camera collecting operation land is matched with the flight speed of the plant protection unmanned aerial vehicle, and the distribution information diagram of the operation area is obtained by combining the geographical position information obtained by the positioning device;
preferably, in the step S2, the controller processes the image information, performs convolution operation on the image information as an input quantity to filter useless information in the image, extracts characteristics such as color, shape, texture and the like from the preprocessed image through an hidden layer of the neural network module, and obtains a disease, pest and control distribution map of the operation land block through operation output;
preferably, in the step S3, a database is established according to various disease and pest control tests and a large amount of field control experimental data in different crop growth periods, a disease and pest space distribution condition is obtained after image recognition, a spraying prescription value is output by a crop disease and pest control expert module, and then an operation land prescription map is generated by combining the distribution condition of disease and pest of an operation land and the spraying quantity, the prescription map which is obtained before operation is generated, and a controller autonomously generates an optimal spraying operation route according to the prescription map;
preferably, the spraying amount information, the unmanned aerial vehicle specification parameter, the flight parameter and the environmental parameter output by the crop disease and insect pest control expert module in the step S4 are transmitted to the controller through the multi-channel data collector; when the operation obtains the prescription diagram, the prescription diagram is imported into the controller, and when the plant protection unmanned aerial vehicle performs plant protection operation, the position of the plant protection unmanned aerial vehicle relative to the prescription diagram is determined by comparing the current position information with the position information of the prescription diagram so as to obtain the spraying amount.
Preferably, the neural network module in the step S4 is a counter-propagating neural network module, the neural network module includes an input layer, an hidden layer and an output layer, samples for training the neural network module are valid test data of a large number of related tests, before training the neural network module, weights and biases are randomly initialized, the weights are set in a (-1, 1) range, an error function e is set, the maximum training learning times and a calculation precision value epsilon are set, the selected samples are used as input quantity X, and expected outputs corresponding to the samples are defined:
the number of the input amounts is as follows: x (k) = (X) 1 (k)，x 2 (k)，…x n (k))
Outputting an expected value: y is Y
And respectively calculating the input and the output of each neuron of the hidden layer, wherein the input quantity of the hidden layer is as follows:
wherein w is i (k) Connecting weights between each input neuron and the hidden layer; b i Neuron bias values for hidden layers; n is the number of input variables;
hidden layer neuron output:
wherein w is o (k) The connection weight between the neuron and the hidden layer is output; b is the output layer neuron bias value;
calculating multiple hidden layers of the neural network module to obtain an output value, and calculating the partial derivative delta of the error function on the neurons of the output layer according to the expected value of the output quantity o (k) Wherein the global error is:
in the training process of the neural network module, the weight w of each layer of neurons of the hidden layer is adjusted according to the accumulated error i (k) Thereby reducing global errors, namely:
eta in the formula is learning rate;
adjusting the weight according to the weight adjusting formula of each hidden layer neuron output layer, namely:
when all variables of the previous training period are smaller than a certain designated threshold value, the neural network module completes a learning period, the learning process is completed according to the set learning period, and the artificial neural network module after deep learning training can output a reasonable result after model operation according to input information.
Preferably, the controller in the step S5 establishes a model meeting the requirements through programming, processes the received information such as environmental temperature and humidity, wind speed and wind direction information, flight parameters, prescription values and the like during operation, and then outputs an optimal spraying amount through the operation of the neural network module, the embedded system transmits the optimal spraying amount to the variable spraying controller, the variable spraying control module generates a pulse square wave with a corresponding duty ratio according to the received spraying amount information, drives the liquid pump after amplifying, and meanwhile, the flow sensor and the pressure sensor of the spraying loop measure the flow velocity and the pressure value of the drug delivery pipeline in real time, and feed back the flow velocity and the pressure value of the drug delivery pipeline, and match the variable spraying controller module, combine the deviation among the actual flow velocity, the actual pressure value, the theoretical flow velocity and the theoretical pressure value, accurately regulate the spraying process, and the flight control module sends a return instruction when the drug liquid is detected to be lower than the set value through detecting the drug amount of the drug tank.
The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the invention, and all equivalent structural changes made by the specification and drawings of the present invention or direct/indirect application in other related technical fields are included in the scope of the present invention.
Claims (4)
1. The plant protection unmanned aerial vehicle accurate variable spraying device for neural network decision-making is characterized by comprising a plurality of horn, wherein the end parts of the horn are connected with a brushless direct current motor, the top end of the brushless direct current motor is connected with rotor blades, a medicine box is connected below a machine body formed by the horn, the bottom of the medicine box is connected with a spray head through a compression-resistant water pipe, the compression-resistant water pipe is connected with a liquid pump, and the spray head is arranged on a spray rod below the machine body; the device comprises a machine arm, a machine body, a controller, a multi-channel data acquisition unit, a micro environment information acquisition unit, a hyperspectral camera, a brushless direct current motor, a positioning device, a controller and a hyperspectral camera, wherein the positioning device is arranged in the middle of the machine arm;
the top of the machine body is provided with a driving amplification module which is respectively connected with the controller and the liquid pump;
the liquid pump is communicated with a filter arranged at the bottom of the medicine box through a pipeline;
the pressure-resistant water pipe is sequentially provided with an overflow valve, a flow sensor and a pressure sensor along the flowing direction of the liquid medicine, and the pressure sensor is respectively connected with the controller;
a liquid level sensor for detecting the liquid level of the liquid medicine is arranged at the top of the inner wall surface of the medicine box and is connected with the controller;
a control method of a plant protection unmanned aerial vehicle accurate variable spraying device applied to neural network decision-making comprises the following steps:
s1: the hyperspectral camera is arranged at the lower part of the horn to collect crop image information of an operation land block, the positioning device acquires position information in real time, the mini-type environment information collector arranged at the top of the machine body is used for collecting environment parameters in the field, and the hyperspectral camera, the positioning device and the mini-type environment information collector respectively transmit collected data to the controller; the flow sensor, the pressure sensor and the liquid level sensor feed back the flow, the pressure value and the residual quantity of the liquid medicine to the controller in real time;
s2: the controller processes the acquired data through the multichannel data acquisition device, and an information processing and decision module in the controller processes the image information by using a trained neural network module and generates a spatial distribution diagram of plant diseases and insect pests;
s3: the controller generates a spraying operation prescription chart of a target land block according to the image processing result and by combining with a crop disease and insect pest control expert module, and the flight control module of the plant protection unmanned aerial vehicle autonomously plans a route by combining with the operation prescription chart;
s4: the controller processes the collected flight parameters, environment parameters, real-time information of the plant protection unmanned aerial vehicle, prescription value information, liquid level, flow and pressure information, calculates each decision quantity through a trained neural network module and obtains the optimal spraying quantity; when a prescription diagram of an operation land block is acquired before operation, the controller extracts current position information and compares the current position information with the prescription diagram, and judges the current position of the plant protection unmanned aerial vehicle so as to acquire prescription value information of the current position;
s5: the variable spraying control module of the controller generates corresponding duty ratio pulse square waves by using a closed-loop PID algorithm according to the received optimal spraying amount information, the driving amplification module is used for driving the liquid pump to work after amplifying, meanwhile, the flow sensor and the pressure sensor feed back actual flow and pressure values in real time, and the flow is accurately regulated again by the controller; the liquid level sensor monitors the residual quantity of the liquid medicine in the medicine box in real time, feeds back the residual quantity of the liquid medicine to the flight control module, and makes a return operation in time when the liquid medicine is lower than a certain set value.
2. The apparatus of claim 1, wherein,
the width of the hyperspectral camera collecting operation land parcels in the step S1 is 10 meters, the length of the hyperspectral camera collecting operation land parcels is matched with the flying speed of the plant protection unmanned aerial vehicle, and an operation area distribution information graph is obtained by combining the geographic position information obtained by the positioning device;
the controller processes the image information in the step S2, carries out convolution operation on the image information as input quantity to filter useless information in the image, extracts color, shape and texture characteristics of the preprocessed image through an hidden layer of the neural network module, and obtains a disease, insect and pest distribution map of the operation land block through operation output;
in the step S3, a database is established according to various disease and pest control experiences in different crop growth periods and a large amount of field control experimental data, the spatial distribution situation of the disease and pest is obtained after image recognition, a spraying prescription value is output by the crop disease and pest control expert module, then the distribution situation of the disease and pest of an operation plot and the spraying quantity are combined to generate the prescription diagram of the operation plot, the prescription diagram is obtained before operation, and the flight control module autonomously generates an optimal spraying operation route according to the prescription diagram;
the spraying amount information, the plant protection unmanned aerial vehicle specification parameters, the flight parameters and the environment parameters output by the crop disease, insect and pest control expert module in the step S4 are transmitted to the controller through the multi-channel data acquisition device; when the prescription diagram is obtained before operation, the prescription diagram is imported into the controller, and in the process of plant protection operation of the plant protection unmanned aerial vehicle, the position of the plant protection unmanned aerial vehicle relative to the prescription diagram is determined by comparing the current position information with the position information of the prescription diagram so as to obtain the spraying amount.
3. The apparatus of claim 1, wherein the neural network module of the step S4 is a back propagation neural network module, the neural network includes an input layer, an implicit layer, and an output layer, the sample data required for training the neural network is valid test data of a plurality of related tests, weights and biases are randomly initialized before training the neural network, a range of each weight is set within (-1, 1), an error function e is set, a training maximum learning number and a calculation precision value epsilon are set, a selected sample is taken as an input quantity X, and a desired output corresponding to the sample is defined:
input quantity: x (k) = (X) 1 (k)，x 2 (k)，…x n (k))
Outputting an expected value: y is Y
And respectively calculating the input and the output of each neuron of the hidden layer, wherein the input quantity of the hidden layer is as follows:
wherein w is i (k) Connecting weights between each input neuron and the hidden layer; b i Neuron bias values for hidden layers; n is the number of input variables;
hidden layer neuron output:
wherein w is o (k) The connection weight between the neuron and the hidden layer is output; b 0 A neuron bias value for the output layer;
calculating a plurality of hidden layers of the neural network to obtain an output value, and calculating a global error according to an expected value of the output quantity, namely:
in the neural network training processAdjusting the weight w of each layer of neurons of the hidden layer according to the accumulated error i (k) Thereby reducing global errors, namely:
eta in the formula is learning rate;
adjusting the weight according to the weight adjusting formula of each hidden layer neuron output layer, namely:
when all variables of the previous training period are smaller than a certain designated threshold value, the neural network completes a learning period, the learning process is completed according to the set learning period, and the neural network module after deep learning training can output a reasonable result after model operation according to input information.
4. The device of claim 1, wherein the neural network module meeting the requirements is built in the controller through programming in the step S5, the received environmental temperature and humidity, wind speed and wind direction information, flight parameters and prescription value information during operation are processed and then used as input values of the neural network, the neural network is operated and then an optimal spraying amount is output, the embedded system can transmit the optimal spraying amount to the variable spraying control module, the variable spraying control module generates a pulse square wave with corresponding duty ratio according to the received spraying amount information and drives the liquid pump after amplification, meanwhile, the flow sensor and the pressure sensor of the spraying loop measure the flow rate and the pressure value of the drug delivery pipeline in real time, and feed back the variable spraying control module, and accurately adjust the spraying process by combining the deviation among the actual flow rate, the actual pressure value, the theoretical flow rate and the theoretical pressure value, and the liquid level sensor detects the medicine amount in the medicine box, and when the liquid level sensor detects that the liquid medicine is lower than the set value, the flight control module sends a return instruction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810660859.3A CN109122633B (en) | 2018-06-25 | 2018-06-25 | Plant protection unmanned aerial vehicle accurate variable spraying device for neural network decision-making and control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810660859.3A CN109122633B (en) | 2018-06-25 | 2018-06-25 | Plant protection unmanned aerial vehicle accurate variable spraying device for neural network decision-making and control method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109122633A CN109122633A (en) | 2019-01-04 |
| CN109122633B true CN109122633B (en) | 2023-10-03 |
Family
ID=64802281
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810660859.3A Active CN109122633B (en) | 2018-06-25 | 2018-06-25 | Plant protection unmanned aerial vehicle accurate variable spraying device for neural network decision-making and control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109122633B (en) |
Families Citing this family (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109677612A (en) * | 2019-01-17 | 2019-04-26 | 昔阳县农业广播电视学校 | A kind of agricultural planting plant protection drone |
| CN109479862B (en) * | 2019-01-18 | 2023-04-07 | 吉林农业大学 | Variable pesticide spraying system and control method thereof |
| CN109845715B (en) * | 2019-03-06 | 2020-12-18 | 广州极飞科技有限公司 | Pesticide spraying control method, device, equipment and storage medium |
| JP6756389B2 (en) * | 2019-03-08 | 2020-09-16 | 東洋製罐株式会社 | Aircraft discharge system |
| BR112021023280A2 (en) * | 2019-05-20 | 2022-01-04 | Basf Agro Trademarks Gmbh | Crop treatment methods, control device and treatment device |
| CN110435892B (en) * | 2019-07-22 | 2022-03-11 | 北京中航智科技有限公司 | Sprinkler and unmanned aerial vehicle |
| CN110435893A (en) * | 2019-07-22 | 2019-11-12 | 北京中航智科技有限公司 | A kind of flusher and unmanned plane |
| CN110979686A (en) * | 2019-10-31 | 2020-04-10 | 王国瑞 | Intelligent plant protection unmanned aerial vehicle and use method thereof |
| DE102019218188A1 (en) * | 2019-11-25 | 2021-05-27 | Robert Bosch Gmbh | Method of working crops in a field |
| CN110825118B (en) * | 2019-12-23 | 2022-05-27 | 河南大学 | Multi-unmanned aerial vehicle cooperative farmland spraying method based on deep learning algorithm |
| CN111077780A (en) * | 2019-12-23 | 2020-04-28 | 湖北理工学院 | Intelligent window adjusting method and device based on neural network |
| CN111280151B (en) * | 2020-02-06 | 2021-10-26 | 山东农业大学 | Variable pesticide application control method based on cotton growth period recognition |
| CN111880558B (en) * | 2020-07-06 | 2021-05-11 | 广东技术师范大学 | Plant protection unmanned aerial vehicle obstacle avoidance spraying method and device, computer equipment and storage medium |
| CN112565726B (en) * | 2020-12-10 | 2023-02-14 | 广州极飞科技股份有限公司 | Method for determining job prescription chart, job control method and related device |
| CN112977828A (en) * | 2021-04-06 | 2021-06-18 | 中国农业科学院都市农业研究所 | Spraying method and device based on unmanned aerial vehicle, electronic equipment and medium |
| CN115500334A (en) * | 2021-06-07 | 2022-12-23 | 中移(成都)信息通信科技有限公司 | Sprayer, plant disease and insect pest identification method and identification equipment |
| CN113247241A (en) * | 2021-06-29 | 2021-08-13 | 扬州大学 | Unmanned aerial vehicle target-aiming accurate pesticide applying system and pesticide applying method thereof |
| CN113460308A (en) * | 2021-07-30 | 2021-10-01 | 中国农业大学 | Unmanned aerial vehicle variable pesticide application control system and method |
| CN113826598A (en) * | 2021-09-10 | 2021-12-24 | 宁波权智科技有限公司 | Unmanned aerial vehicle pesticide spreading method and device based on neural network |
| CN113895631A (en) * | 2021-12-09 | 2022-01-07 | 山东理工职业学院 | Pesticide spraying control equipment, method and medium based on unmanned aerial vehicle |
| CN114445027B (en) * | 2021-12-27 | 2023-04-07 | 江西省交通工程集团有限公司 | Attendance frequency control method and system for spraying vehicle |
| CN116050842B (en) * | 2023-02-01 | 2023-08-04 | 东华理工大学 | Dynamic control method and system for urban underground engineering construction safety risk |
| CN117153325B (en) * | 2023-10-30 | 2024-01-30 | 佛山科学技术学院 | Anticancer drug effectiveness evaluation method and system based on graph contrast learning |
| CN118124838B (en) * | 2024-05-08 | 2024-07-16 | 杭州而墨农业技术有限公司 | Seedling condition and pest and disease damage early warning patrol unmanned aerial vehicle and method |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103999841A (en) * | 2014-04-18 | 2014-08-27 | 浙江大学 | Automatic targeting and spraying system |
| CN105025262A (en) * | 2015-07-08 | 2015-11-04 | 常州华奥航空科技有限公司 | Large-area agricultural work unmanned plane |
| CN105173085A (en) * | 2015-09-18 | 2015-12-23 | 山东农业大学 | Automatic control system and method for variable pesticide spraying of unmanned aerial vehicle |
| CN106564599A (en) * | 2016-11-22 | 2017-04-19 | 江苏蒲公英无人机有限公司 | Plant protection method for unmanned aerial vehicle based on multispectral remote sensing |
| CN106857467A (en) * | 2017-03-06 | 2017-06-20 | 佳木斯大学 | A kind of brainpower insufflation agricultural chemicals machine and pesticide spraying control method |
| CN206573944U (en) * | 2016-12-21 | 2017-10-20 | 石河子大学 | The accurate pesticide spraying system of rotor wing unmanned aerial vehicle |
| CN108069027A (en) * | 2017-04-28 | 2018-05-25 | 深圳市九天创新科技有限责任公司 | The plant protection system and method for multispectral fixed-wing unmanned plane cooperation plant protection unmanned plane |
| CN108094379A (en) * | 2017-12-26 | 2018-06-01 | 燕山大学 | A kind of fog machine and its paths planning method for being used to perform agriculture automatic spraying task |
| CN208425428U (en) * | 2018-06-25 | 2019-01-25 | 华南农业大学 | The accurate variable-rate spraying device of plant protection drone |
-
2018
- 2018-06-25 CN CN201810660859.3A patent/CN109122633B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103999841A (en) * | 2014-04-18 | 2014-08-27 | 浙江大学 | Automatic targeting and spraying system |
| CN105025262A (en) * | 2015-07-08 | 2015-11-04 | 常州华奥航空科技有限公司 | Large-area agricultural work unmanned plane |
| CN105173085A (en) * | 2015-09-18 | 2015-12-23 | 山东农业大学 | Automatic control system and method for variable pesticide spraying of unmanned aerial vehicle |
| CN106564599A (en) * | 2016-11-22 | 2017-04-19 | 江苏蒲公英无人机有限公司 | Plant protection method for unmanned aerial vehicle based on multispectral remote sensing |
| CN206573944U (en) * | 2016-12-21 | 2017-10-20 | 石河子大学 | The accurate pesticide spraying system of rotor wing unmanned aerial vehicle |
| CN106857467A (en) * | 2017-03-06 | 2017-06-20 | 佳木斯大学 | A kind of brainpower insufflation agricultural chemicals machine and pesticide spraying control method |
| CN108069027A (en) * | 2017-04-28 | 2018-05-25 | 深圳市九天创新科技有限责任公司 | The plant protection system and method for multispectral fixed-wing unmanned plane cooperation plant protection unmanned plane |
| CN108094379A (en) * | 2017-12-26 | 2018-06-01 | 燕山大学 | A kind of fog machine and its paths planning method for being used to perform agriculture automatic spraying task |
| CN208425428U (en) * | 2018-06-25 | 2019-01-25 | 华南农业大学 | The accurate variable-rate spraying device of plant protection drone |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109122633A (en) | 2019-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109122633B (en) | Plant protection unmanned aerial vehicle accurate variable spraying device for neural network decision-making and control method | |
| US11169541B2 (en) | Detecting and following terrain height autonomously along a flight path | |
| CN111990388B (en) | Selective spraying system, ground-air cooperative pesticide application system and cooperative method | |
| US11526997B2 (en) | Targeting agricultural objects to apply units of treatment autonomously | |
| CN103988824A (en) | Automatic targeting and spraying system based on binocular vision technology | |
| CN103999841A (en) | Automatic targeting and spraying system | |
| US11963473B2 (en) | Multiple emitters to treat agricultural objects from multiple payload sources | |
| CN203860304U (en) | Automatic targeting and spraying system | |
| US20210187532A1 (en) | Obscurant emission to assist image formation to automate agricultural management and treatment | |
| US20210055417A1 (en) | Methods for generating treatment prescriptions based on uav-derived plant height data and related crop management systems | |
| CN105936337A (en) | Agricultural unmanned aerial vehicle | |
| US20210186006A1 (en) | Autonomous agricultural treatment delivery | |
| Wang et al. | Design and development of orchard autonomous navigation spray system | |
| Vikram | Agricultural Robot–A pesticide spraying device | |
| CN110515393B (en) | Method, device and system for real-time anti-drift correction of agricultural and forestry spraying of unmanned aerial vehicle | |
| US20210185885A1 (en) | Calibration of systems to deliver agricultural projectiles | |
| Roman et al. | Stereo vision controlled variable rate sprayer for specialty crops: Part II. Sprayer development and performance evaluation | |
| Guo et al. | Spraying coverage path planning for agriculture unmanned aerial vehicles | |
| Agurob et al. | Autonomous Vision-based Unmanned Aerial Spray System with Variable Flow for Agricultural Application. | |
| CN116058353A (en) | Unmanned management system and method based on farmland cultivation | |
| CN110839609A (en) | Plant protection fixed wing unmanned aerial vehicle variable spraying control system | |
| CN208425428U (en) | The accurate variable-rate spraying device of plant protection drone | |
| CN116636518A (en) | Unmanned aerial vehicle pesticide application control method and system | |
| JP2024528460A (en) | Mulch device field treatment | |
| CN115079713A (en) | Unmanned aerial vehicle accurate pesticide application operation method based on flight path optimization |
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 |