Precise decision-making and precise irrigation control method based on mobile irrigation system

By introducing visual detection and sensors into the mobile irrigation system, combined with microcontroller control and smart irrigation system, the coordinated operation and precise displacement of water, fertilizer and medicine are realized, the problems of waste and impact force of irrigation resources are solved, and the accuracy and resources of irrigation are improved. Utilization.

CN116076340BActive Publication Date: 2024-12-03GUANGZHOU UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310113931.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-02-13
Publication Date
2024-12-03
Estimated Expiration
2043-02-13

AI Technical Summary

Technical Problem

The existing mobile irrigation system is difficult to accurately irrigate to designated crops during irrigation, resulting in waste of water and fertilizer resources, and is unable to take into account both fertilization and medicine, ignoring the impact of irrigation impact on crops.

Method used

Using precise decision-making and precise irrigation control methods based on mobile irrigation systems, real-time data is obtained through visual detection and sensors, the microcontroller controls the motor mechanism, combines the smart irrigation system to make decisions, realize the coordinated operation and precise displacement of water, fertilizer and medicine, and optimize the irrigation model and Operation of sprinkler irrigation mechanism.

Benefits of technology

The utilization rate of water and fertilizer resources and drug utilization rate are improved, the impact force of irrigation is optimized, the accuracy and controllability of irrigation are enhanced, and resource waste and crop damage are reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116076340B_ABST
    Figure CN116076340B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of smart agriculture and forestry, and discloses a method for precise decision-making and precise control of irrigation based on a mobile irrigation system, comprising the following steps: a first step: the device starts preprocessing; a second step: the sensor module collects environmental information and builds a current crop growth model; a third step: the main control end processes image information and natural environment data; a fourth step: the hardware execution end tracks sample targets according to the joint decision of the main control and the smart irrigation system; a fifth step: the irrigation operation data is transmitted back to the smart irrigation system for analysis operations, and a future crop growth model is built for user operation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of smart agriculture and forestry, and specifically to a method for precise decision-making and precise irrigation control based on a mobile irrigation system. Background Art

[0002] Innovating agricultural production methods, replacing traditional manual labor with automation, and bringing intelligent technology to agriculture are the only way for future technological development. Traditional mobile irrigation devices are mostly used for large-scale spraying, ignoring the accurate positioning of crops, causing most water and fertilizer resources to infiltrate the soil and nutrients to be lost, which not only increases cultivation expenses but also affects the ecological environment. How to use innovative inventions to improve the water resources of agricultural irrigation devices, the utilization rate of fertilizers and drugs, and the control of the fluid impact force of the irrigation subject are issues that need to be solved urgently.

[0003] In response to the above problems, existing patents have coping strategies, such as "An Intelligent Greenhouse Mobile Precision Irrigation Robot" (Publication No. CN113424735A), which mainly identifies the surrounding area of ​​irrigated soil and collects soil photos to achieve precise irrigation. However, there is no capture and analysis of the historical irrigation conditions of crops in the current area.

[0004] "A multi-channel mobile orchard irrigation and fertilization machine and irrigation and fertilization method" (publication number CN113179733A) is mainly about the precise control of fertilizer ratio and fertilization formula. As can be seen from the above, most of the current patented innovative technologies are based on soil and crop growth conditions and innovative water-fertilizer ratios, and there are few technical studies on optimizing irrigation decisions and irrigation nozzle tracking to achieve mobile irrigation of water, fertilizer and medicine classification.

[0005] Therefore, most of the existing precision identification technologies tend to be reasonable water-fertilizer ratios and soil identification. Their disadvantage is that they do not accurately identify the specific locations of fruits in different growth stages and the specific precision irrigation points and specific decision-making models, which can easily lead to excessive irrigation or waste of irrigation resources. The difficulty lies in the rational construction of the irrigation model and the precise displacement of the sprinkler mechanism. With the development of smart agricultural machinery and big data technology, agricultural irrigation in the new era should be combined with the precise positioning of the sprinkler mechanism and the precise decision-making of the sprinkler method to improve the irrigation level. For this reason, we propose a precise decision-making and precise control method of irrigation based on a mobile irrigation system. Summary of the invention

[0006] 1. Technical issues to be solved

[0007] In view of the deficiencies of the prior art, the present invention provides a precise decision-making and irrigation control method based on a mobile irrigation system, which solves the problem that the current integrated water and fertilizer mobile device is difficult to accurately irrigate the designated crops during sprinkler irrigation, resulting in waste of water and fertilizer resources, and cannot simultaneously take into account fertilization and pesticide application in mobile device operations, while ignoring the impact of irrigation impact on crops when each crop group is irrigated. By introducing an intelligent irrigation system to participate in decision-making, visual detection and sensors in the operating area obtain real-time sample data, and a single-chip microcomputer controls the motor mechanism group to collaboratively implement the decision-making plan in the irrigation system, return the irrigated data and various growth indexes of the crop group in the current area, so as to achieve the effect of improving the utilization rate of water and fertilizer resources and the utilization rate of drugs.

[0008] (II) Technical solution

[0009] To achieve the above-mentioned purpose, the present invention provides the following technical solution: a method for accurate decision-making and accurate irrigation control based on a mobile irrigation system, comprising the following steps:

[0010] Step 1: The device starts pre-processing;

[0011] Step 2: The sensor module collects environmental information and builds a model of current crop growth conditions;

[0012] Step 3: The main control end processes image information and natural environment data;

[0013] Step 4: The hardware execution end tracks the sample target based on the joint decision of the main control and the smart irrigation system;

[0014] Step 5: The irrigation operation data is transmitted back to the smart irrigation system for analysis and operation, and a future crop growth model is constructed for user operation.

[0015] Preferably, the contents of the preprocessing in the first step are as follows: after the whole machine is powered on, the smart irrigation system at the main control end and the user end is initialized, the upper and lower computers and the sensor end perform scanning and preprocessing, and the irrigation mechanism pre-irrigates to ensure the smooth flow of circulation within the mechanism and ensure that each unit operates normally during the operation of the device.

[0016] Preferably, the specific content of the second step is: the sensor measures the real-time distance between the device and the target crop through the camera module and the distance measurement module, the current growth data of the crop is packaged and transmitted in a frame format, and the main control board classifies and processes the transmitted information according to a preset coding table to obtain the data.

[0017] Preferably, the specific content of the third step is: the main control end receives the image, and for specific image data, uses the "RGB model" and "HSV model" and "edge pixel acquisition" methods for real-time processing, and the calculation results include "sample distance", "sample condition" and "sample surrounding environment", and based on the recurrent neural network, associate the changes in the area over the time span, construct real-time sample growth height, growth status and other data curves, and package the sample data and transmit it back to the user-controlled smart irrigation system end;

[0018] At the same time, the physical data such as the sample position and the device relative to the crop are calculated according to the fuzzy adaptive PID and the simulation inference table is queried to defuzzify the displacement parameters required for the relevant approximation, and then the target movement value in the table is transmitted to the lower hardware execution end.

[0019] Preferably, the specific content of the fourth step is: after obtaining the real-time signal and data, the hardware execution end will approach the crop sample and start the operation, the upper three-way irrigation structure will cooperate with the drug ratio scheme to implement fruit-end sprinkler irrigation, and the lower three-way irrigation structure will cooperate with the water-fertilizer ratio scheme to implement root-end sprinkler irrigation, real-time sprinkler volume, and the sprinkler mechanism will return real-time data to the smart irrigation system controlled at the user level.

[0020] Preferably, the specific content of the fifth step is: when the irrigation operation in the area is completed, the irrigation equipment will transmit the operation status in the current area back, and the algorithm decision optimization plan on the smart irrigation system side will be used to construct the historical and future changes of crop growth height, disease symptoms and other comprehensive states in the time dimension based on the neural network algorithm. At the same time, the growth model curves of each region are combined with the living conditions and habits of related plants in the agricultural and forestry big data to calculate the user-level adjustable natural conditions required for the optimal environment of each area at the current time.

[0021] (III) Beneficial effects

[0022] Compared with the prior art, the present invention provides a precise decision-making and precise irrigation control method based on a mobile irrigation system, which has the following beneficial effects:

[0023] 1. The precise decision-making and precise control method of irrigation based on the mobile irrigation system, combined with the control model of the coordinated operation of water, fertilizer and medicine, preferably performs adaptive irrigation operations. This is because the upper and lower three-way control mechanism of water, fertilizer and medicine in the present invention achieves precise displacement based on vision and fuzzy adaptive PID algorithm. Through this optimization scheme, the displacement subject can stably and quickly approach the sample, control the spacing between crop samples, and reasonably make up for the shortcomings of traditional irrigation in controlling the distance between the irrigation subject and the irrigated object.

[0024] 2. The precise decision-making and precise irrigation control method based on the mobile irrigation system reasonably changes the fluid pressure according to the influence of the weight of water and fertilizer on the spraying speed of the fluid, thereby optimizing its impact on the plant. This is because the present invention timely adjusts the irrigation fluid pressure or the nozzle structure of the injection end based on the impact of the fluid weight and the fluid injection direction on the target soil irrigation point and the pressure resistance and accuracy of the drug on the target fruit, thereby optimizing the control of the impact force on the irrigation subject during real-time irrigation at the mobile end.

[0025] 3. The precise decision-making and precise control method of irrigation based on the mobile irrigation system builds a smart irrigation system model required for optimizing the growth of crop water, fertilizer and pesticides to efficiently deepen the overall controllability of crops. This is because the user-controlled smart irrigation system used in the present invention optimizes the decision-making party based on the recurrent neural network, based on historical irrigation experience and the irrigated water, fertilizer and pesticide curves, combined with real-time conditions, to find the optimal irrigation amount, present the overall irrigation distribution of the irrigated area, the irrigation degree of each region and the future growth curve, and enhance the rationality and practicality of the main control decision. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 Schematic diagram of the pre-stage data analysis for the main control of sensor detection;

[0027] Figure 2 It is a schematic diagram of the decision model of the smart irrigation system;

[0028] Figure 3 It is a schematic diagram of the approaching sample flow of the three-way injection mechanism;

[0029] Figure 4 This is a schematic diagram of the entire operation of the irrigation device. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] See also Figure 1-4 , the precise decision-making and precise control method of irrigation based on the mobile irrigation system includes the following steps:

[0032] Step 1: After the whole machine is powered on, the smart irrigation system at the main control end and the user end is initialized, the upper and lower computers and the sensor end perform scanning and pre-processing, and the irrigation mechanism pre-irrigates to ensure the smooth flow of circulation within the mechanism and ensure that all units operate normally during the operation of the device.

[0033] Step 2: The sensor module collects environmental information and builds a model of the current crop growth situation. The sensor measures the real-time distance between the device and the target crop through the camera module and the distance measurement module. The data such as the current growth status of the crop is packaged and transmitted in a frame format. The main control board will classify and process the transmitted information according to the preset coding table. Among them, the methods of identifying data and data transmission are applicable to multiple forms, supporting preset communication methods such as SPI, IIC, and USART. The specific operation is determined by the specific technical implementer.

[0034] Step 3: The main control end processes image information and natural environment data. The main control end receives the image, and for specific image data, uses "RGB model", "HSV model" and "edge pixel acquisition" to process in real time. The calculation results include "sample distance", "sample situation", "sample surrounding environment", etc. Based on the recurrent neural network, the changes in the area over the time span are associated to build real-time sample growth height, growth status and other data curves, package the sample data and transmit it back to the user-controlled smart irrigation system. At the same time, the physical data such as the sample position and the relative position of the device to the crop are calculated according to the fuzzy adaptive PID and the simulation reasoning table is queried to solve the fuzziness and obtain the displacement parameters required for the relevant approximation, and then the target movement value in the table is transmitted to the lower hardware execution end.

[0035] Step 4: The hardware execution end tracks the sample target according to the joint decision of the main control and the smart irrigation system. After obtaining real-time signals and data, the hardware execution end will approach the crop sample and start the operation. The upper three-way irrigation structure will cooperate with the drug ratio scheme to implement fruit-end sprinkler irrigation, and the lower three-way irrigation structure will cooperate with the water-fertilizer ratio scheme to implement root-stem sprinkler irrigation. The real-time sprinkler volume, the area’s sprinkler volume, the sprinkler mechanism’s real-time position and other data are all returned to the user-controlled smart irrigation system. At the same time, during the movement of the mobile device, the device will track the target crop to achieve the effect of "roots follow water", "soil follows fertilizer", and "fruit follows medicine".

[0036] Step 5: The irrigation operation data is transmitted back to the smart irrigation system for analysis and operation, and a future crop growth model is constructed for user operation. When the irrigation operation in the area is completed, the irrigation equipment will transmit the operation status in the current area, and the algorithm decision optimization plan on the smart irrigation system side will be used to construct the historical and future changes of the crop growth height, disease symptoms and other comprehensive states in the time dimension based on the neural network algorithm. At the same time, the growth model curves of each area are combined with the living conditions and habits of the relevant plants in the agricultural and forestry big data to calculate the temperature, humidity and other natural conditions that can be adjusted by the user at the current time for the most suitable environment in each area.

[0037] The input end of the sensor module mainly returns real-time images, and the information obtained is based on the upper and lower computers that cooperate with ESP32 and STM32 to achieve the effect of feedback control. At the same time, the sensor module can also feedback distance data, which is convenient for the sprinkler control mechanism and sample spacing. It should be noted that the image transmission delay of this method can be ignored. The above sensor operates based on Openmv, and the onboard STM32H7 can obtain more than 50 frames of recognition feedback data in a high-speed computing environment. The obtained data is given to the main control chip based on high-speed SPI communication.

[0038] The main control end is based on the STM32F4 series single-chip control, and its 168MHz operating speed ensures the speed and accuracy of PWM control. The data obtained from the input end is processed in real time based on the recurrent neural network, thereby analyzing the target plant type, growth period and other objective conditions. Based on the theoretical judgment of the fluid weight and fluid direction in the main control end and the feedback of the infrared sensor and visual mechanism in real-time operation, it is used to achieve precise control, stable spraying, and reasonable fertilization. The obtained results are combined with the fuzzy adaptive PID algorithm, and the quantification square is the distance between the device and the target crop and fruit, and the corresponding deviation e and deviation increment ec are calculated. The triangular membership function can be used to participate in the fuzzification process, and then the required parameters for movement are obtained by reasoning and defuzzification according to the fuzzy rules. The displacement control of the three-way irrigation mechanism (three-dimensional mobile mechanism) is realized based on the above control scheme, and the speed and direction of the mechanism are adjusted at the same time, so as to reduce the objective errors that the system will encounter during the mobile operation.

[0039] The sprinkler hardware implementation end analyzes and implements the mobile execution mode rehearsed by the above main control. The visual end at the sprinkler head takes real-time photos to obtain the crop situation in the current work area and feedback. At the same time, the three-way sprinkler structure leads the sprinkler head to the designated location. When it reaches the calculated position, it quickly feedbacks the real-time situation and implements the sprinkler action. The spray device follows the target crop fruit, and the fertilization device follows the target root and stem. When the three-way sprinkler mechanism performs the above operations, based on the weight and direction of the fluid, the fluid injection formula controls the speed and irrigation source pressure. By changing the water pressure of the irrigation source or changing the nozzle size of the sprinkler device and the impact flow rate and fertilization direction of the spray and fertilization operation, the impact force on the crop and soil is rationalized.

[0040] The data obtained by the sensor will be transmitted back to the smart irrigation system operated by the user. The irrigation system stores data such as temperature, humidity, weather, crop types, crop distribution, and crop growth in the current working environment. The smart irrigation system is based on a recurrent neural network, and the input parameters are the current growth period of the crop, crop disease, soil hardness parameters (data provided by humidity sensors pre-deployed in the soil), the ratio of water and fertilizer to be irrigated, and the amount of irrigation. The weight file in the neural network training set is changed in real time according to the area captured by the camera in the current mobile state, and the growth optimization plan is combined with big data calculation in the background to present the time-based growth model curve of the irrigation and growth parameters of each area required in the future.

[0041] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A precise decision-making and precise irrigation control method based on a mobile irrigation system, characterized in that: The following steps are involved: Step 1: The device starts pre-processing; Step 2: The sensor module collects environmental information and builds a model of current crop growth conditions; Step 3: The main control end processes image information and natural environment data; Step 4: The hardware execution end tracks the sample target based on the joint decision of the main control and the smart irrigation system; Step 5: The irrigation operation data is transmitted back to the smart irrigation system for analysis and operation, and a future crop growth model is constructed for user operation; The specific content of the third step is: the main control end receives the image, and for specific image data, uses the "RGB model" and "HSV model" and "edge pixel acquisition" methods for real-time processing, and the calculation results include "sample distance", "sample condition" and "sample surrounding environment". Based on the recurrent neural network, the changes in the area over the time span are associated, and the real-time sample growth height and growth status data curves are constructed, and the sample data is packaged and transmitted back to the user-controlled smart irrigation system end; At the same time, the physical data of the sample position and the device relative to the crop position are calculated according to the fuzzy adaptive PID and the simulation inference table is queried to defuzzify the displacement parameters required for the relevant approximation, and then the target movement value in the table is transmitted to the lower hardware execution end; The specific content of the fourth step is: after obtaining real-time signals and data, the hardware execution end will approach the crop sample and start the operation, the upper three-way irrigation structure will cooperate with the drug ratio scheme to implement fruit-end sprinkler irrigation, and the lower three-way irrigation structure will cooperate with the water-fertilizer ratio scheme to implement root-end sprinkler irrigation, and the real-time sprinkler irrigation volume, the sprinkler mechanism will return the real-time data to the smart irrigation system controlled at the user level; The specific content of the fifth step is: when the irrigation operation in the area is completed, the irrigation equipment will report back the operation status in the current area, and the algorithm decision optimization plan on the smart irrigation system side will be used to build the historical and future changes of crop growth height and disease comprehensive status in the time dimension based on the neural network algorithm. At the same time, the growth model curve of each area is combined with the living conditions and habits of related plants in the agricultural and forestry big data to calculate the user-level adjustable natural conditions required for the optimal environment of each area at the current time; The data obtained by the sensor will be transmitted back to the smart irrigation system operated by the user. The smart irrigation system stores the temperature, humidity, weather, crop type, crop distribution, and crop growth data in the current working environment. The smart irrigation system is based on a recurrent neural network. The input parameters are the current growth period of the crop, crop disease conditions, soil hardness parameters, water-fertilizer ratio and irrigation amount. The weight file in the neural network training set is changed in real time according to the area captured by the camera in the current mobile state. In the background, the growth optimization plan is calculated based on big data to present the time-based growth model curve of the irrigation and growth parameters of each area required in the future.

2. The precise decision-making and precise irrigation control method based on the mobile irrigation system according to claim 1, characterized in that: The contents of the preprocessing in the first step are as follows: after the whole machine is powered on, the smart irrigation system at the main control end and the user end is initialized, the upper and lower computers and the sensor end are scanned and preprocessed, and the irrigation mechanism is pre-irrigated to ensure the smooth flow of circulation within the mechanism and to ensure the normal operation of each unit during the operation of the device.

3. The method for accurate decision-making and accurate irrigation control based on a mobile irrigation system according to claim 1, characterized in that: The specific content of the second step is: the sensor measures the real-time distance between the device and the target crop through the camera module and the distance measurement module, the current growth data of the crop is packaged and transmitted in a frame format, and the main control board classifies and processes the transmitted information according to the preset coding table.