CN114467900B - Accurate spraying method of lawn herbicide - Google Patents
Accurate spraying method of lawn herbicide Download PDFInfo
- Publication number
- CN114467900B CN114467900B CN202210146503.4A CN202210146503A CN114467900B CN 114467900 B CN114467900 B CN 114467900B CN 202210146503 A CN202210146503 A CN 202210146503A CN 114467900 B CN114467900 B CN 114467900B
- Authority
- CN
- China
- Prior art keywords
- grid
- spraying
- herbicide
- lawn
- images
- 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
- 239000004009 herbicide Substances 0.000 title claims abstract description 82
- 238000005507 spraying Methods 0.000 title claims abstract description 74
- 230000002363 herbicidal effect Effects 0.000 title claims abstract description 59
- 241000196324 Embryophyta Species 0.000 claims abstract description 42
- 239000000575 pesticide Substances 0.000 claims abstract description 25
- 238000000034 method Methods 0.000 claims abstract description 20
- 238000013528 artificial neural network Methods 0.000 claims abstract description 14
- 238000009333 weeding Methods 0.000 claims abstract description 14
- 230000002068 genetic effect Effects 0.000 claims abstract description 9
- 239000007921 spray Substances 0.000 claims abstract description 9
- 238000012549 training Methods 0.000 claims description 19
- 238000001228 spectrum Methods 0.000 claims description 8
- 239000003814 drug Substances 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 229940079593 drug Drugs 0.000 claims description 2
- 230000000694 effects Effects 0.000 description 2
- 241000404546 Soliva Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 241000894007 species Species 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
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
- A01M21/00—Apparatus for the destruction of unwanted vegetation, e.g. weeds
- A01M21/04—Apparatus for destruction by steam, chemicals, burning, or electricity
- A01M21/043—Apparatus for destruction by steam, chemicals, burning, or electricity by chemicals
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Environmental Sciences (AREA)
- Zoology (AREA)
- Insects & Arthropods (AREA)
- Pest Control & Pesticides (AREA)
- Wood Science & Technology (AREA)
- Toxicology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mechanical Engineering (AREA)
- Catching Or Destruction (AREA)
Abstract
The invention discloses an accurate spraying method of a lawn herbicide, and relates to the technical field of weed identification and accurate weeding. According to the method, for the lawn needing herbicide spraying, the longitude and latitude coordinates of the corners of the lawn are located in advance through RTK, aerial images of the lawn are obtained through an unmanned aerial vehicle according to the located longitude and latitude coordinates, and the aerial images are uploaded to a cloud server. The cloud server divides the aerial images into grids, sequentially inputs the grid images of the lawn into the trained classification neural network to recognize weeds, outputs herbicides used correspondingly by the grid images, draws a weed spraying area map according to the longitudes and latitudes of the grid images, sends the longitudes and latitudes of the grid images and the corresponding herbicides to the pesticide applying robot, the pesticide applying robot loads the corresponding herbicides, calculates the shortest path of the grid area needing spraying through a genetic algorithm, moves the pesticide applying robot into the corresponding grid needing spraying, and sprays the herbicides.
Description
Technical Field
The invention relates to the technical field of weed identification and accurate weeding, in particular to an accurate spraying method of a lawn herbicide.
Background
Lawn greening is one of important marks of city civilization degree and is commonly found in places such as park greening, stadiums, golf courses and the like. The lawn has obvious effects of beautifying the environment, purifying the air, maintaining water and soil and the like, and is difficult to maintain due to serious lawn degradation caused by weed attack. Chemical weeding based on a herbicide is a common mode for preventing and controlling lawn weeds, and the realization of accurate spraying of the herbicide is the key for reducing the dosage of the herbicide and reducing environmental pollution. Now, the application of the weeding robot is wider, but the following problems still exist:
(1) Although some weeding robots adopt a suspension structure to slow down the shake, the method not only increases the complexity of the vehicle body structure, but also cannot completely avoid the problem of image shake;
(2) The moving accuracy of the robot is extremely high in requirements on image acquisition and weed identification, moving errors can cause mutual overlapping or interval of continuous image acquisition, herbicide is repeatedly sprayed to the same area due to image overlapping, and part of weeds are not acquired due to image interval to cause spraying leakage;
(3) The weed identification of the weeding robot does not establish the association between the weed species and the herbicide species, different herbicides have different weed control spectrums, the tolerance conditions of the weeds to the different herbicides are different, and the herbicide spraying without distinguishing the weed control spectrums wastes the herbicide and is usually useless;
(4) The weeding robot has the following single machine: the weed identification algorithm deployment mode of the NVIDIA Jetson embedded system is limited by the computing power of the deployed terminal, and the algorithm performance is difficult to be fully exerted;
(5) The weeding robot needs to completely traverse all the operation areas, respectively executes actions such as image acquisition, weed identification, spraying decision and spraying execution in a unit target area, senses weeds and simultaneously sprays decision by taking a small area as a unit, and the efficiency is low due to the cyclic working mode, so that the requirement of large-scale application cannot be met;
(6) At present, the weeding robot integrates a vehicle body moving mechanism, an image acquisition system, a weed identification and spraying decision system, a spraying control system, a spraying actuator and other functions and modules, and has the advantages of large and complex structure, high failure rate, extremely high manufacturing cost and maintenance cost and difficulty in commercial landing.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides an accurate spraying method of a herbicide for a lawn, which combines an aerial lawn image of an unmanned aerial vehicle, a weed spraying area map drawn by a cloud server and the collaborative spraying of a plurality of pesticide applying robots, thereby greatly improving the efficiency of weed control.
In order to achieve the technical purpose, the invention adopts the following technical scheme: a method for spraying a lawn herbicide specifically comprises the following steps:
(1) Shooting a lawn area through an unmanned aerial vehicle, and uploading a collected lawn image to a cloud server;
(2) Dividing the lawn image acquired in the step (1) into m multiplied by n grids, carrying out image enhancement on each grid image to obtain a training set, carrying out manual classification on the training set according to a weed control spectrum, and taking a herbicide corresponding to the weed control spectrum as a label of the grid image;
(3) Inputting the training set into a classification neural network for training until the cross entropy loss function is converged, and finishing the training of the classification neural network;
(4) For the lawn needing to be sprayed with the herbicide, the longitude and latitude coordinates of corners of the lawn are located in advance through RTK, aerial images of the lawn are obtained through an unmanned aerial vehicle according to the located longitude and latitude coordinates, and the aerial images are uploaded to a cloud server. The cloud server performs grid division on the aerial images, sequentially inputs the grid images of the lawn into a trained classification neural network for weed identification, outputs herbicides used correspondingly by the grid images, and combines longitude and latitude information of the grid images to draw a weed spraying area map;
(5) And (3) issuing the longitude and latitude of the grid image and the corresponding herbicide to a pesticide applying robot, loading the corresponding herbicide by the pesticide applying robot, calculating the shortest path of the grid area needing spraying through a genetic algorithm, and moving the pesticide applying robot to the corresponding grid area needing spraying according to the shortest path to spray the herbicide.
Further, the method for enhancing the image in the step (2) comprises the following steps: rotating the grid image, changing the brightness of the grid image, changing the contrast of the grid image, adding noise points and carrying out fuzzy processing on the grid image.
Further, the training set of step (2) deletes grid images that relate to multiple weeds and are sensitive to different herbicides.
Further, the classification neural network employs VGGNet, resNet, or MobileNet.
Further, the cross entropy loss function L is specifically:
wherein M is the total herbicide class, c is the index of the herbicide class, i is the index of the grid image, p ic The predicted probability that the grid image i belongs to the herbicide class c is given, and N is the total number of grid images.
Further, if spraying of multiple herbicides is involved in the weed spraying map drawn in the step (4), multiple pesticide applying robots are required to be loaded with different herbicides, the shortest path of a grid area required to be sprayed is calculated through a genetic algorithm, and the pesticide applying robots move to the grid areas corresponding to the areas required to be sprayed according to the shortest paths to spray the herbicides.
Further, the spraying range of the pesticide spraying robot is the same as the size of the grid area.
Further, the process of calculating the shortest path of the grid area to be sprayed by the genetic algorithm specifically comprises the following steps:
(a) Coding the grid images to be sprayed in the weed spraying area map, and classifying the grid images sprayed with the same herbicide into one class;
(b) Randomly taking a certain code in the grid images classified into one class as an initial population, taking the initial population as a starting point, acquiring the lengths of all paths in the class, and reserving one path with the shortest path length;
(c) Generating a new population by selecting, crossing or mutating the initial population, taking the new population as a starting point, acquiring the lengths of all paths in the class, and reserving one path with the shortest path length;
(d) And (d) repeating the step (c) until the iteration number of generating a new population reaches a threshold value, and selecting the shortest one of the reserved path lengths as the walking path of the medicine application robot.
Compared with the prior art, the invention has the following beneficial effects:
(1) The unmanned aerial vehicle is used for replacing a traditional weeding robot to collect images, and the quality of image collection and the weed identification effect are ensured by stable flying shooting of the unmanned aerial vehicle;
(2) The weed identification process is deployed on the cloud server, and the performance of weed identification and spraying decision can be effectively improved by means of the computing power of the cloud server;
(3) The spraying of herbicide is carried out through the weeds spraying regional picture of drawing, through the division in grid region, can realize the collaborative spraying of multiple herbicide, and spraying time is short, and the spraying pertinence is strong, improves ruderal control efficiency, can effectively practice thrift the herbicide, and spraying efficiency is higher.
Drawings
FIG. 1 is a flow chart of the method of spraying the lawn herbicide of the present invention.
Detailed Description
The technical solution of the present invention is further explained below with reference to the accompanying drawings.
Fig. 1 is a flow chart of a method for spraying the lawn herbicide, which specifically includes the following steps:
(1) The lawn area is shot by the unmanned aerial vehicle, the unmanned aerial vehicle can accurately acquire aerial images of the lawn area according to the RTK positioning information, and the acquired lawn images are uploaded to the cloud server;
(2) And (2) dividing the lawn image collected in the step (1) into m multiplied by n grids, wherein the values of m and n are determined by the unit coverage area sprayed by the pesticide applying robot, and the physical size corresponding to each grid image after cutting is equal to the spraying coverage area of the pesticide applying robot. The method comprises the steps of carrying out image enhancement on each grid image to expand a training set, wherein a classification neural network needs to have a clear classification rule during training, deleting grid images which relate to various weeds and are sensitive to different herbicides in the training set, carrying out manual classification on the training set according to a weeding spectrum, taking the herbicide corresponding to the weeding spectrum as a label of the grid image, identifying the category of the weeds, and further selecting the herbicide for spraying; the image enhancement method comprises the following steps: rotating the grid image, changing the brightness of the grid image, changing the contrast of the grid image, adding noise points and carrying out fuzzy processing on the grid image.
(3) Inputting the training set into a classification neural network for training until the cross entropy loss function is converged, and finishing the training of the classification neural network; the classification neural network adopts VGGNet, resNet or MobileNet, and the related cross entropy loss function L is specifically as follows:
wherein M is the total herbicide class, c is the index of the herbicide class, i is the index of the grid image, p ic The predicted probability that the grid image i belongs to the herbicide class c is given, and N is the total number of grid images.
(4) For the lawn needing herbicide spraying, the longitude and latitude coordinates of the corners of the lawn are located in advance through RTK, aerial images of the lawn are obtained through the unmanned aerial vehicle according to the located longitude and latitude coordinates, and the aerial images are uploaded to the cloud server. The cloud server performs grid division on the aerial images, the grid images of the lawn are sequentially input into a trained classification neural network for weed identification, the herbicides used correspondingly by the grid images are output, a weed spraying area map is drawn by combining longitude and latitude information of the grid images, and the weed spraying area map comprises a grid area needing herbicide spraying, physical position information corresponding to the grid area and the types of the herbicides sprayed in the grid area. The existing spraying technology mainly directly sprays the herbicide by an unmanned aerial vehicle, and has the problems of short endurance, small drug-loading amount, low efficiency and serious aerial fogdrop drifting.
If spraying of multiple herbicides is involved, multiple pesticide applying robots are needed to load different herbicides, the shortest path of a grid area needing spraying is calculated through a genetic algorithm, and the pesticide applying robots move to the corresponding grid area needing spraying according to the shortest path to spray the herbicides.
(5) And issuing the longitude and latitude of the grid image and the corresponding herbicide to a pesticide applying robot, wherein the spraying range of the pesticide applying robot is the same as the size of the grid area. The pesticide application robot loads corresponding herbicides and calculates the shortest path of a grid area needing spraying through a genetic algorithm, and the pesticide application robot moves to the corresponding grid area needing spraying according to the shortest path to spray the herbicides, and the method specifically comprises the following steps:
(a) Coding the grid images to be sprayed in the weed spraying area map, and classifying the grid images sprayed with the same herbicide into one class;
(b) Randomly taking a certain code in the grid images classified into one class as an initial population, taking the initial population as a starting point, acquiring the lengths of all paths in the class, and reserving one path with the shortest path length;
(c) Generating a new population by selecting, crossing or mutating the initial population, taking the new population as a starting point, acquiring the lengths of all paths in the class, and reserving one path with the shortest path length;
(d) And (c) repeating the step (c) until the number of iterations for generating a new population reaches a threshold value, and selecting the shortest one of the reserved path lengths as the walking path of the drug application robot.
According to the accurate spraying method of the lawn herbicide, the aerial image of the lawn shot by the unmanned aerial vehicle, the weed spraying area drawing drawn by the cloud server and the collaborative spraying of the plurality of pesticide applying robots are combined, so that the accuracy of obtaining the image of the lawn area is improved, meanwhile, the spraying of the herbicide in the lawn area is changed into the gridding spraying, the collaborative spraying of various herbicides can be realized, the spraying time is short, the spraying pertinence is high, the weed control efficiency is improved, the herbicide can be effectively saved, and the spraying efficiency is higher.
The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above-mentioned embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.
Claims (7)
1. The method for spraying the lawn herbicide is characterized by comprising the following steps:
(1) Shooting a lawn area through an unmanned aerial vehicle, and uploading a collected lawn image to a cloud server;
(2) Dividing the lawn image acquired in the step (1) into m multiplied by n grids, carrying out image enhancement on each grid image to obtain a training set, carrying out manual classification on the training set according to a weeding spectrum, and taking a herbicide corresponding to the weeding spectrum as a label of the grid image;
(3) Inputting the training set into a classification neural network for training until the cross entropy loss function is converged, and finishing the training of the classification neural network;
(4) For the lawn needing herbicide spraying, pre-positioning longitude and latitude coordinates of corners of the lawn through RTK, acquiring aerial images of the lawn through an unmanned aerial vehicle according to the positioned longitude and latitude coordinates, uploading the aerial images to a cloud server, carrying out grid division on the aerial images by the cloud server, sequentially inputting grid images of the lawn into a trained classification neural network for weed identification, outputting herbicide used by the grid images, and drawing a weed spraying area map according to the longitude and latitude of the grid images;
(5) The longitude and latitude of the grid image and the corresponding herbicide are issued to a pesticide applying robot, the pesticide applying robot loads the corresponding herbicide, the shortest path of a grid area needing spraying is calculated through a genetic algorithm, and the pesticide applying robot moves to the corresponding grid area needing spraying according to the shortest path to spray the herbicide; the spraying range of the pesticide applying robot is the same as the size of the grid area.
2. The method for spraying the lawn herbicide as claimed in claim 1, wherein the image enhancement method in the step (2) comprises: rotating the grid image, changing the brightness of the grid image, changing the contrast of the grid image, adding noise points and carrying out fuzzy processing on the grid image.
3. The method of applying a lawn herbicide as claimed in claim 1, wherein the training set of step (2) deletes the grid image which relates to a plurality of weeds and is sensitive to different herbicides.
4. The method of spraying a lawn herbicide as claimed in claim 1, wherein the neural network of classification is VGGNet, resNet or MobileNet.
5. Method for spraying a lawn herbicide according to claim 1, wherein said cross entropy loss function L is in particular:
wherein M is the total herbicide class, c is the index of the herbicide class, i is the index of the grid image, p ic The predicted probability that the grid image i belongs to the herbicide class c is given, and N is the total number of grid images.
6. The method for spraying a lawn herbicide as claimed in claim 1, wherein if spraying of a plurality of herbicides is involved in the weed spray pattern drawn in the step (4), the shortest path of a grid area to be sprayed is calculated by a genetic algorithm by using a plurality of pesticide applying robots loaded with different herbicides, and the pesticide applying robots move to the grid area to be sprayed according to the shortest path to perform spraying of the herbicide.
7. The method for spraying the lawn herbicide as claimed in claim 1, wherein the process of traversing the shortest path of the grid area to be sprayed by the genetic algorithm is specifically as follows:
(a) Coding the grid images to be sprayed in the weed spraying area map, and classifying the grid images sprayed with the same herbicide into one class;
(b) Randomly taking a certain code in the grid images classified into one class as an initial population, taking the initial population as a starting point, acquiring the lengths of all paths in the class, and reserving one path with the shortest path length;
(c) Generating a new population by selecting, crossing or mutating the initial population, taking the new population as a starting point, acquiring the lengths of all paths in the class, and reserving one path with the shortest path length;
(d) And (c) repeating the step (c) until the number of iterations for generating a new population reaches a threshold value, and selecting the shortest one of the reserved path lengths as the walking path of the drug application robot.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210146503.4A CN114467900B (en) | 2022-02-17 | 2022-02-17 | Accurate spraying method of lawn herbicide |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210146503.4A CN114467900B (en) | 2022-02-17 | 2022-02-17 | Accurate spraying method of lawn herbicide |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114467900A CN114467900A (en) | 2022-05-13 |
CN114467900B true CN114467900B (en) | 2022-12-27 |
Family
ID=81482319
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210146503.4A Active CN114467900B (en) | 2022-02-17 | 2022-02-17 | Accurate spraying method of lawn herbicide |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114467900B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113142170A (en) * | 2021-03-30 | 2021-07-23 | 宁波市农业科学研究院 | Unmanned aerial vehicle intelligent fixed-point weeding technology suitable for rice field |
CN113159459A (en) * | 2021-05-21 | 2021-07-23 | 南京林业大学 | Multi-forest-area air route scheduling planning method based on fusion algorithm |
CN113349188A (en) * | 2021-05-31 | 2021-09-07 | 南京林业大学 | Lawn and forage grass precise weeding method based on cloud weeding spectrum |
-
2022
- 2022-02-17 CN CN202210146503.4A patent/CN114467900B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113142170A (en) * | 2021-03-30 | 2021-07-23 | 宁波市农业科学研究院 | Unmanned aerial vehicle intelligent fixed-point weeding technology suitable for rice field |
CN113159459A (en) * | 2021-05-21 | 2021-07-23 | 南京林业大学 | Multi-forest-area air route scheduling planning method based on fusion algorithm |
CN113349188A (en) * | 2021-05-31 | 2021-09-07 | 南京林业大学 | Lawn and forage grass precise weeding method based on cloud weeding spectrum |
Also Published As
Publication number | Publication date |
---|---|
CN114467900A (en) | 2022-05-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Singh et al. | An intelligent WSN-UAV-based IoT framework for precision agriculture application | |
CN109254588B (en) | Unmanned aerial vehicle cluster cooperative reconnaissance method based on cross variation pigeon swarm optimization | |
Albani et al. | Field coverage and weed mapping by UAV swarms | |
CN111352417B (en) | Rapid generation method of heterogeneous multi-unmanned aerial vehicle cooperative path | |
CN105425820B (en) | A kind of multiple no-manned plane collaboratively searching method for the moving target with perception | |
CN107977743B (en) | Multi-unmanned aerial vehicle cooperative task allocation method and device | |
CN107886201B (en) | Multi-objective optimization method and device for multi-unmanned aerial vehicle task allocation | |
CN107807665B (en) | Unmanned aerial vehicle formation detection task cooperative allocation method and device | |
CN111609864B (en) | Multi-policeman cooperative trapping task allocation and path planning method under road network constraint | |
CN109557939A (en) | A kind of quick approach to formation control based on pseudo- distributed unmanned plane cluster | |
CN111024080B (en) | Unmanned aerial vehicle group-to-multi-mobile time-sensitive target reconnaissance path planning method | |
CN109197278A (en) | Determination method and device, the determination method of herbal sprinkling strategy of Job Policies | |
CN104808660A (en) | Concave convex mixed complex polygon farmland unmanned aerial vehicle spraying operation flight path planning method | |
CN110262563A (en) | Multiple no-manned plane collaboratively searching mesh calibration method waterborne | |
CN110956132A (en) | Method for intelligently constructing weed control model based on unmanned aerial vehicle cooperation | |
CN107135541A (en) | UWB indoor localization method based on OPTICS Density Clusterings and BP neural network | |
Sharma | Evolution of unmanned aerial vehicles (UAVs) with machine learning | |
CN112001309A (en) | Target searching method, device and equipment based on unmanned aerial vehicle cluster and storage medium | |
CN114467900B (en) | Accurate spraying method of lawn herbicide | |
Gao et al. | Ant colony optimization for UAV-based intelligent pesticide irrigation system | |
CN112203291A (en) | Cluster control method for area coverage and connectivity maintenance based on knowledge embedding | |
Dolias et al. | On the routing of unmanned aerial vehicles (UAVs) in precision farming sampling missions | |
CN116772848A (en) | Green real-time planning method for four-dimensional flight track of aircraft terminal area | |
Manullang et al. | Optimum trajectory of multi-UAV for fertilization of paddy fields using ant colony optimization (ACO) and 2-opt algorithms | |
CN111024081A (en) | Unmanned aerial vehicle group-to-single moving time-sensitive target reconnaissance path planning 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |