Unmanned ground automatic targeting pesticide application vehicle capable of detecting crop canopy pore structure and pest and disease distribution in real time

By using STM32F4 series microcontrollers and fisheye cameras on unmanned spraying trucks, combined with robotic arms and sensors, accurate spraying of crops such as peppers has been solved, and the problems of insufficient intelligence and penetration have been improved. Spraying efficiency and coverage reduce operating costs and protect the environment.

CN119014388AInactive Publication Date: 2024-11-26HAINAN UNIV

Patent Information

Application Number
CN202411219483.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-02
Publication Date
2024-11-26
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing unmanned spraying trucks are not intelligent, have a narrow scope of application, high promotion costs, outstanding air operation endurance, insufficient penetration, resulting in waste of pesticides and environmental pollution, making it difficult to achieve efficient and accurate shrub crops such as peppers. Spray.

Method used

Using STM32F4 series microcontrollers as control system, equipped with fisheye cameras and a variety of sensors, realize mechanical visual analysis and pore distribution detection, combine the robotic arms and universal joints with three servo structures to connect the wheels, design an automated spraying system to support land operations , reduce maintenance costs, and realize remote monitoring and data sharing through the Internet of Things and WeChat applets.

Benefits of technology

Accurate identification and efficient spraying of crops such as peppers are achieved, reducing pesticide waste, improving spraying efficiency and coverage, reducing operation and maintenance costs, and protecting the environment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119014388A_ABST
    Figure CN119014388A_ABST
Patent Text Reader

Abstract

The invention discloses an unmanned ground automatic targeting pesticide application vehicle capable of detecting a crop canopy pore structure and pest and disease damage distribution in real time. The unmanned ground automatic targeting pesticide application vehicle comprises a pore distribution detection system, a targeting control system, a spraying system, a power system and a man-machine interaction interface design. The distribution state of pepper plants is obtained through the sensor group, a plant distribution diagram is obtained after processing, the pest and disease damage distribution condition and the plant morphological characteristics are calculated and uploaded to the cloud server through the Internet of Things module, and real-time data sharing with other upper computers is carried out. Relevant parameters can be accessed through a human-computer interaction interface in the form of a mobile phone or a computer webpage and the like; meanwhile, relevant parameters are transmitted to an airborne control system and are processed to generate a control signal, and a response signal is transmitted to a power and spraying system. The device can accurately spray pesticide in a targeted mode, is suitable for various crops, can reduce pesticide waste and pollution in the pesticide spraying process, and solves the problems that similar products are low in intelligent level, high in cost and the like.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a medicine spraying vehicle, and in particular to an automatic unmanned medicine spraying vehicle. Background Art

[0002] With the advancement of agricultural modernization, the backward status of agricultural production needs to be improved urgently, and the intelligentization and automation of agricultural production is imminent. Most of the current unmanned spraying vehicles use spraying tubes for spraying. This spraying method not only leads to serious waste of pesticides, but also aggravates environmental pollution. The intelligence level of existing spraying vehicles needs to be improved, and most of them still need manual operation and supervision. In addition, the maintenance and management of unmanned vehicles also require certain manpower and material resources, and their manufacturing and maintenance costs are relatively high. Some of the spraying work is also carried out with the help of drones, but due to the high density of shrub crops such as peppers, the penetration of the liquid droplets sprayed by drones is insufficient. At the same time, the rotor wind field generated by the drone greatly promotes the phenomenon of "wind drifting" of drugs, causing pesticides to drift and causing serious impacts on the environment.

[0003] This product uses a robotic arm for targeted spraying, which can not only accurately identify the disease locations of peppers and other plants, but also effectively reduce the waste and drift of pesticides. By being equipped with cameras and a variety of sensors, it can realize real-time feedback on the spraying process and actively adjust the appropriate angles of the main and auxiliary robotic arms to achieve more accurate targeted spraying. It can realize autonomous path planning and transmit the working scene to the mobile phone in real time to realize manual remote supervision. At the same time, it can establish an IoT hosting database through the ESP8266 WIFI module and share data in real time through the WeChat applet. The main monitoring parameters include the remaining amount of medicine in the medicine box, the real-time location of the trolley, the land environment parameters, etc., so as to better ensure the smooth operation of the trolley.

[0004] Although there are some similar products on the market, their technical levels and performances vary. The current unmanned pesticide application machinery is generally divided into drone application and unmanned vehicle application. Although drone spraying has the advantages of high efficiency, saving manpower and material resources, and convenient remote operation, it also has some disadvantages that cannot be ignored:

[0005] (1) Spraying is limited by its load and endurance, and the efficiency of spraying operations on large areas of farmland is low;

[0006] (2) UAV operation requires a high level of technical skills and operating experience, and improper operation is dangerous;

[0007] (3) UAV spraying has a low coverage of the bottom of shrub crops such as peppers, resulting in low pesticide utilization;

[0008] (iv) The cost of drone spraying is relatively high, which is a considerable burden for independent farmers with limited financial resources;

[0009] (V) Pesticides are easily blown away by the wind, causing environmental pollution and damage to other plants, and lack of targeting.

[0010] In addition, most of the unmanned spraying vehicles on the market currently use extensive methods to spray pesticides, which will cause a large amount of pesticide waste and "wind drift". At the same time, the unmanned spraying system has high requirements for the selection and dosage of pesticides. Improper operation will cause environmental pollution or crop damage. Due to the imperfect technology, unmanned vehicles still face challenges in identifying target crops and avoiding spraying in non-target areas, which may cause problems such as mis-spraying or missed spraying.

[0011] Therefore, in line with the country's growing demand for efficient and intelligent pesticide application products for saving resources and protecting the environment in the new era and for new smart agricultural management, the patent of this invention came into being. Summary of the invention

[0012] The purpose of this patent is to overcome the problems of the above-mentioned automated pesticide application products, such as insufficient intelligence, narrow scope of application, high promotion cost, prominent problems of aerial operation endurance and insufficient penetration. It has the following characteristics:

[0013] (i) In terms of intelligence, the present invention adopts a relatively cheap STM32F4 series single-chip microcomputer as the main control system of the car, and uses multiple ultrasonic sensors (distance sensors) as the main detection sensors for the morphology of pepper plants. In addition, in order to improve the monitoring accuracy, a fisheye camera based on Raspberry Pi 4B is also installed for mechanical vision analysis, which is used to detect the planting range, canopy density, canopy size and the direction of travel of the field ridges of pepper plants, and feedback is given to the control board to adjust the posture of the robotic arm and control the car to work autonomously;

[0014] (ii) In terms of applicability, the patent of the present invention adopts a two-degree-of-freedom arm profiling design with a three-servo structure for the main mechanical arm. By controlling the angle of the servo, the main mechanical arm can be adjusted to adapt to different plant canopy forms. In addition, we use universal joints to connect the wheels to form a wheel camber angle of -15°. When the trolley walks on the ridges, the wheels and the ridges achieve the maximum contact area, which not only ensures the working power output of the trolley, but also reduces the travel resistance during the working process, ensuring the trolley's ability to travel between ridges;

[0015] (III) In terms of application and promotion, the design of the patent of the present invention also takes into account the humanized design requirements during use. First, the different operating instructions of the car realize the visual design of the controller and the Internet of Things service. In addition to the controller that comes with the car, the user can also remotely control it through a computer or mobile phone app. Secondly, the working parameters of the car are transmitted to the remote server in real time through the Internet of Things device, which can realize real-time data detection. In addition, the main mechanical arm structure of the car realizes the standardization of the communication interface, the independence of the control program and the modularization of the main structure, which further reduces the cost of promotion and use while also ensuring the simplicity of maintenance;

[0016] (IV) In terms of working endurance, the task of the present invention is to spray on land. In terms of the amount of medicine stored, it is not necessary to consider the influence of the working lift force like unmanned aerial vehicles, and the amount of medicine stored in the medicine storage box can be effectively increased. In addition, the main working system of the car can be separated from the chassis of the car. In the process of upgrading the car, the power system of the car can also be enhanced, such as replacing a gasoline engine or a small diesel engine to ensure the output power and working efficiency of the car.

[0017] (V) In terms of penetration, the main robotic arm of the present invention adopts a three-servo structure, which can better ensure the high conformability of the robotic arm posture, and can make corresponding nozzle positioning points and spraying angles for different crop forms during the adjustment process; in addition, the present invention has an auxiliary robotic arm installed under the lifting platform, and an electric push rod is installed under the auxiliary robotic arm, which can adjust the extension length of the auxiliary robotic arm to better achieve spraying on the back of plant leaves; at the same time, in the actual working process, the auxiliary robotic arm and the main robotic arm can work together to form a "protective cover" to cover the entire canopy, thereby achieving full coverage of the plant and reaching the crown of the crop pests.

[0018] The purpose of the present invention is achieved through the following technical solutions:

[0019] An unmanned ground automatic target spraying vehicle comprises a pore distribution detection system, a target control system, a spraying system, a power system and a human-machine interaction interface design.

[0020] The pore distribution detection system includes an information collection module and an airborne processor. The information collection module includes a fisheye camera for obtaining crop image information, an ultrasonic sensor for measuring the distance between the plant and the trolley, an optical sensor for monitoring the surface characteristics of the plant, a Beidou positioning module for obtaining crop theorem information, and a laser radar for obtaining three-dimensional point cloud information of the plant; the airborne processor is used to process the agricultural information data obtained by the information collection module, and then establish a plant distribution map model through a spatial statistical algorithm; the agricultural information data includes the location and degree of pests and diseases, ridge spacing, ridge length, plant height, and plant growth cycle, etc.;

[0021] The target control system includes a robotic arm module and an airborne control module; the robotic arm module includes a driving servo for driving its reasonable operation and a base, needle bearings and upper and lower arms that support the robotic arm mechanism; the airborne control module is used to receive the plant distribution map model obtained by the airborne processor, calculate the corresponding time signal, servo PWM signal, water pump voltage signal according to the logistic regression algorithm, spatial regression algorithm, etc., and transmit it to the human-machine interface while transmitting it to the spraying system and power system for effective operation.

[0022] The spraying system includes a lifting platform, a medicine box, a water pump, a pipeline system, a nozzle, an anti-drift cover and an electromagnetic valve. The lifting platform cooperates with the mechanical arm to complete the spraying work at different heights; the medicine box is equipped with a perforated partition, which can reduce the unstable walking of the vehicle body and the uneven spraying of pesticides caused by the shaking of the vehicle when walking. There are gaps between the partitions to make the pesticides more evenly mixed, and the medicine box contains a water level sensor for real-time monitoring of the water level of the medicine, a flow rate sensor for real-time monitoring of the flow rate and flow of the medicine, etc., and these data will be fed back to the human-computer interaction interface in real time; the nozzle is set at the top of the main mechanical arm and the auxiliary mechanical arm, and also includes a laser radar sensor for monitoring the distance between the nozzle and the plant, which is electrically connected to the airborne control module; the anti-drift cover adopts a prism structure with a wide top and a narrow bottom, which is innovatively placed on the unmanned vehicle, and can effectively cooperate with the auxiliary mechanical arm to reduce the waste of pesticide drift; the electromagnetic valve is used to control the spray volume of the nozzle, which is set on the nozzle and electrically connected to the airborne control module.

[0023] The power system is mainly composed of a steering adjustment system and a forward and backward system for executing operations. The steering adjustment system includes a rack, a pulley, a connecting rod and a wheel bracket, and is driven by a stepper motor; the forward and backward system for executing operations include four wheels arranged in an outward-facing O-shaped shape, and are driven by a DC brushed motor.

[0024] The human-computer interaction interface mainly includes a real-time monitoring interface and a remote monitoring interface. The real-time monitoring interface includes equipment parameter monitoring, drug management, pest and disease monitoring, and plant status monitoring. Equipment parameter monitoring focuses on real-time data and historical analysis; the drug management interface includes record forms, historical record viewing, and usage suggestions; pest and disease monitoring presents uploads, identification results, and early warning prompts; plant status monitoring displays images, maturity analysis, and management suggestions; the remote control interface is mainly used for emergency stops of the trolley in emergencies.

[0025] The pore detection system comprises the following steps:

[0026] The agricultural information data of the operation area is collected through the information collection module of the small vehicle, and the agricultural information data includes the location and degree of pests and diseases, ridge spacing, ridge length, plant height and plant growth cycle, etc.;

[0027] The location and extent of pests and diseases are processed by the onboard processor to obtain a pest and disease distribution map; the pore distribution and specific location of the plants are obtained by combining the image information obtained by the fisheye camera and the three-dimensional point cloud information obtained by the lidar;

[0028] Combined with spatial statistical algorithms, the ideal grid prescription map of pests and diseases is obtained after linear processing of the pest and disease distribution map, and the ideal grid prescription map of plants is obtained after linear processing of the specific positions of plants, and the two are transmitted together to the airborne control module;

[0029] The onboard control module obtains the location and degree of pests and diseases in each grid, the specific location of the plants, determines a reasonable path planning, generates corresponding work instructions and transmits them to the robotic arm module, spraying system and power system to guide them to complete the target action, spraying task and walking work;

[0030] The onboard control module is provided with a memory, which is used to store the data acquired by the information collection module, mainly a plant distribution map model established based on agricultural information data; in this way, on the one hand, it can be used to make a comprehensive and multi-angle evaluation of the spraying work after spraying, and on the other hand, if a special optimized spraying algorithm is set in the control system, it can be used to optimize the next spraying.

[0031] Furthermore, considering the pore distribution, the spraying amount compensation of each nozzle is:

[0032] q'1=-αρx+q

[0033] Among them, the spraying volume of each nozzle after compensation is q'1, α is the porosity compensation coefficient, and the value range is 0-1; the spraying volume of a single nozzle in the ideal grid prescription map is x, the internal porosity of the grid is ρ, and the spraying volume of each nozzle before compensation is q. The working principle of the above-mentioned unmanned ground automatic target spraying vehicle (hereinafter referred to as the vehicle) is:

[0034] The trolley supports both automatic mode and remote control mode. Before the spraying work begins, the trolley can reach the area to be sprayed through the remote control mode. When entering the ridge, the ultrasonic sensor of the pore distribution detection system carried by the trolley measures the distance between the object and the sensor by sending and receiving ultrasonic signals, and detects the distance between the plants at both ends of the trolley and the trolley in real time. When the distance changes within the pre-set range, it can be considered that the path taken by the trolley is reasonable. When it is detected that the distance between the two ends of the trolley has a large change, it can be considered that it has reached the end of the ridge, and then the turning action is executed. During the movement of the trolley, the Beidou positioning module will feedback the position of the trolley to the human-computer interaction interface in real time. The above is the working principle of the trolley movement.

[0035] During the operation of the trolley, the fisheye camera will accurately record the plants in front, and obtain the location and degree of pests and diseases through the machine vision algorithm, and feed it back to the onboard control module. The onboard control module will output the corresponding PWM signal, time signal, etc. to the spraying module, and spray the medicine according to the calculated drug concentration and volume. At the same time, the control module will also transmit the spatial coordinates of the spraying position to the servo of the robotic arm, the push rod motor on the lifting platform, and the motor on the slide rail. The cooperation of these three enables the fan-shaped cover on the top of the robotic arm to accurately cover the location of the pests and diseases, so as to start spraying. Similarly, the two nozzles installed on the frame will also be combined with the corresponding fan-shaped windshield of the robotic arm to effectively spray the pests and diseases at the bottom of the crop. After spraying the calculated amount of medicine, the sprayer will automatically move forward slowly. If the location of the pests and diseases is not detected on both sides, the sprayer will continue to move forward; but as long as the presence of pests and diseases is detected on one side, the sprayer will stop moving forward. The above is the working principle of the trolley spraying.

[0036] During the operation of the trolley, the equipment parameters of the trolley, including the trolley running speed, the amount of medicine remaining in the medicine box, the spraying speed, the position of the mechanical arm, etc., as well as the pepper position map obtained by the onboard processor will be fed back to the human-computer interaction interface in real time for the operator to make the next plan. If an emergency occurs, the trolley will automatically alarm and stop all actions. There is also a "one-click stop" function in the human-computer interaction interface. When the amount of medicine in the trolley reaches the warning line, the back-end system will remind the operator that the amount of medicine is insufficient and add medicine in time.

[0037] During the movement of the car, the Beidou navigation system on the car will send the latitude and longitude position of the car to the human-computer interaction interface at any time, and display the surrounding map in real time. Through the data returned by the fisheye camera, we can view the surrounding environment of the sprayer at this time on the human-computer interaction interface. After completing the spraying action, the system will record the entire spraying process, drug type, plant pest type, drug dosage, battery consumption and other parameters for reference next time. At the same time, the human-computer interface will also display the weather at this time and place, local weather warnings, and future weather changes. Based on the big data model, the system will recommend the best spraying work for users according to the current weather conditions.

[0038] A preferred embodiment of the present invention, wherein the pore distribution detection system includes an information collection module and an airborne processor. The information collection module includes a fisheye camera for acquiring crop image information, an ultrasonic sensor for measuring the distance between the plant and the trolley, an optical sensor for monitoring the surface characteristics of the plant, a Beidou positioning module for acquiring crop theorem information, and a laser radar for acquiring three-dimensional point cloud information of the plant; the airborne processor is used to process the agricultural information data acquired by the information collection module, and then establish a plant distribution map model through a spatial statistical algorithm; the agricultural information data includes the location and degree of pests and diseases, ridge spacing, ridge length, plant height, and plant growth cycle, etc.;

[0039] Preferably, the operation of the pore detection system comprises the following steps:

[0040] The agricultural information data of the operation area is collected through the information collection module of the small vehicle, and the agricultural information data includes the location and degree of pests and diseases, ridge spacing, ridge length, plant height and plant growth cycle, etc.;

[0041] The onboard Raspberry Pi 4B processor combined with the YOLOv5 visual processing solution monitors the location and extent of pests and diseases to obtain a pest and disease distribution map. The image information obtained by the fisheye camera and the three-dimensional point cloud information obtained by the lidar are combined to obtain the pore distribution and the specific location of the plant.

[0042] Combined with spatial statistical algorithms, the ideal grid prescription map of pests and diseases is obtained after linear processing of the pest and disease distribution map, and the ideal grid prescription map of plants is obtained after linear processing of the specific positions of plants, and the two are transmitted together to the airborne control module;

[0043] The onboard control module obtains the location and degree of pests and diseases in each grid, the specific location of the plants, determines a reasonable path planning, generates corresponding work instructions and transmits them to the robotic arm module, spraying system and power system to guide them to complete the target action, spraying task and walking work;

[0044] A preferred embodiment of the present invention, wherein the target control system includes a robotic arm module and an airborne control module; the robotic arm module includes a driving servo for driving its reasonable operation and a base, a needle bearing and an upper and lower arm that support the robotic arm mechanism; the airborne control module is used to receive the plant distribution map model obtained by the airborne processor, calculate the corresponding time signal, servo PWM signal, water pump voltage signal according to the logistic regression algorithm, the spatial regression algorithm, etc., and transmit it to the human-machine interface while transmitting it to the spraying system and the power system for effective operation.

[0045] Preferably, the onboard control module is provided with a memory, and the memory is used to store the data acquired by the information collection module, mainly a plant distribution map model established based on agricultural information data; in this way, on the one hand, it can be used to make a comprehensive and multi-angle evaluation of the spraying work after spraying, and on the other hand, if a special optimized spraying algorithm is set in the control system, it can be used to optimize the next spraying.

[0046] A preferred solution of the present invention, wherein the spraying system includes a lifting platform, a medicine box, a water pump, a pipeline system, a nozzle, an anti-drift cover and an electromagnetic valve. The lifting platform cooperates with the mechanical arm to complete the spraying work at different heights; the medicine box is equipped with a perforated partition, which can reduce the unstable walking of the vehicle body and the uneven spraying of pesticides caused by the shaking of the vehicle when walking, and there are gaps between the partitions to make the pesticides more evenly mixed, and the medicine box contains a water level sensor for real-time monitoring of the water level of the medicine, a flow rate sensor for real-time monitoring of the flow rate and flow of the medicine, etc., and these data will be fed back to the human-computer interaction interface in real time; the nozzle is set at the top of the main mechanical arm and the auxiliary mechanical arm, and also includes a laser radar sensor for monitoring the distance between the nozzle and the plant, which is electrically connected to the airborne control module; the anti-drift cover adopts a prism structure with a wide top and a narrow bottom, which is innovatively placed on the unmanned vehicle, and can effectively cooperate with the auxiliary mechanical arm to reduce the waste of pesticide drift; the electromagnetic valve is used to control the spray volume of the nozzle, is set on the nozzle, and is electrically connected to the airborne control module.

[0047] Preferably, considering the pore distribution, the spraying amount compensation of each nozzle is:

[0048] q'1=-αρx+q

[0049] Among them, the spraying amount of each nozzle after compensation is q'1, α is the porosity compensation coefficient, and its value range is 0-1; the spraying amount of a single nozzle in the ideal grid prescription map is x, the internal porosity of the grid is ρ, and the spraying amount of each nozzle before compensation is q.

[0050] Preferably, two of the main robotic arms and the auxiliary robotic arms are used, and a lifting platform and a slide rail are used. The lifting platform can raise the robotic arm, and the slide rail can move the robotic arm left and right. With the joint action of the two, the range of movement is wider.

[0051] Preferably, a wide-at-top, narrow-at-bottom prism-type anti-drift shield is provided on the spray head of the trolley robotic arm, which can prevent the pesticide drift of the main robotic arm while forming a linkage with the auxiliary robotic arm to effectively reduce pesticide waste and environmental pollution.

[0052] A preferred embodiment of the present invention is that the power system is mainly composed of a steering adjustment system and a forward and backward system for executing operations. The steering adjustment system includes a rack, a pulley, a connecting rod and a wheel bracket, and is driven by a stepper motor; the forward and backward system for executing operations include four wheels arranged in an outward-facing O-shape, and are driven by a DC brushed motor.

[0053] A preferred embodiment of the present invention, wherein the human-computer interaction interface mainly includes a real-time monitoring interface and a remote monitoring interface, the real-time monitoring interface includes equipment parameter monitoring, drug management, pest and disease monitoring, and plant status monitoring, and the equipment parameter monitoring focuses on real-time data and historical analysis; the drug management interface includes record forms, historical record viewing and usage suggestions; pest and disease monitoring presents uploads, identification results and early warning prompts; plant status monitoring displays images, maturity analysis and management suggestions; the remote control interface is mainly used for emergency stopping of the vehicle in emergency situations.

[0054] Preferably, the pest detection is mainly performed by collecting plant information with a fisheye camera, calculating that the location of the pest is different from the normal location through a machine vision algorithm, and indicating the degree of the pest with different depths of color in the interface, and using a laser radar for comparison. When the results obtained by the two are not much different, the identification is considered to be correct. When the results of the two are greatly different, the identification is continued until the results are not much different.

[0055] Preferably, the plant status detection mainly infers the growth cycle of the plant based on different plant heights to determine parameters such as the robot arm angle and water pump voltage. Compared with the prior art, the present invention has the following beneficial effects:

[0056] 1. The main robotic arm of the patent of this invention adopts a three-servo structure, which can better ensure the high conformability of the robotic arm posture. During the adjustment process, the corresponding nozzle positioning point and spraying angle can be made according to different crop forms; in addition, we have installed a secondary robotic arm under the lifting platform, and at the same time installed an electric push rod under the secondary robotic arm, which can adjust the extension length of the robotic arm to better realize the spraying on the back of the plant leaves; at the same time, in the actual working process, the secondary robotic arm and the main robotic arm can work together to make the spray form a "protective cover" covering the entire canopy, achieving full coverage of the plant and reaching the top of the crop pests.

[0057] 2. The patent of this invention adopts a prism-type anti-drifting shield that is wide at the top and narrow at the bottom. While preventing the main mechanical arm from drifting pesticides, it can better form a linkage with the auxiliary mechanical arm to reduce pesticide waste and protect the environment. The medicine box is equipped with basic monitoring and detection sensors, including a water level sensor to monitor the water level of the medicine, a flow rate sensor to monitor the flow rate and flow of the medicine, etc., and a perforated partition is installed in the medicine box to reduce the problems of unstable walking of the vehicle body and uneven spraying of pesticides caused by the shaking of the trolley when walking. In addition, the gaps between the partitions can make the pesticide mixing more even.

[0058] 3. In the process of spraying, the patent of this invention uses laser radar sensor detection to control the robotic arm and then adjust the nozzle position. The fisheye camera can detect the location of pests and diseases in three-dimensional space. By using PWM control technology, the servos on the robotic arm can cooperate with each other, and finally the nozzle is directly aimed at the location of pests and diseases. In addition, this product can achieve refined management for crops in different growth periods. Through the built-in intelligent analysis system, this product can automatically identify the growth stage of the plant and adjust the spraying plan according to the needs of different stages. For example, in the seedling stage, the car will use a gentle spray method to avoid damage to the seedlings; in the middle and mature stages, the concentration and spraying amount of the liquid medicine will be adjusted according to the growth of the crops and the pest and disease situation to ensure that the crops can grow healthily and improve the yield and quality. In addition, the onboard processor can accurately calculate the amount of pesticides to be applied, the spraying flow rate, etc. according to the growth status of the plants, and can take into account the pores between the plants and make compensation in real time, so that pesticides can be used rationally and pesticide waste and pollution can be reduced. At the same time, during the entire spraying process, this product will record key data in real time, such as spraying volume, spraying time, and pest and disease conditions in the operating area. These data will not only be stored for subsequent analysis and agricultural management, but will also be fed back to the agricultural management system in real time through a specific data transmission system for management personnel to view and analyze at any time.

[0059] 4. The patent of the present invention uses a fisheye camera for path planning and an ultrasonic sensor for obstacle avoidance. The wheels adopt an outward-facing octave layout and use a rack + connecting rod + stepper motor combination for steering. The optimal steering conditions are obtained through reasonable calculation and design. Experimental comparison shows that this method is more stable than the differential steering method under the premise that the medicine box is full. This can make the unmanned ground automatic target spraying vehicle safer and more reliable when operating on complex paths.

[0060] 5. The patent of this invention is based on WeChat applet as the human-computer interaction interface, establishes communication with the back end through WebSocket in real time, receives the position, speed and other information of the unmanned ground automatic target spraying vehicle in real time, and displays it on the interface. The back-end vehicle is equipped with Beidou positioning module and multiple sensors, and uploads the data to the back-end server in real time. The server processes the data in real time and pushes updates to the front end through WebSocket. By calling the API of WeChat applet, the data of each sensor is obtained and displayed on the interface in real time. The chart library is used to display historical data analysis. After receiving the sensor data, it is stored in the database, and an API is provided for the front end to request data. Equipment parameters, use and application conditions are recorded through forms, etc. After receiving the data submitted by the front end, it is stored in the database. Upload crop images, call the image recognition interface for analysis, and display the recognition results. Use machine learning models to analyze the uploaded images, identify pests and diseases, and return results. Call sensors to obtain plant status, use sensor data processing technology, analyze plant status and return results. The human-computer interaction scenarios provided by the patent of this invention include: remote monitoring and real-time adjustment of the trolley through a camera, adjusting the real-time working status of the trolley through remote control intervention, and real-time broadcasting through a voice module. In addition, real-time feedback of parameters such as water level, flow rate, and trolley speed in the medicine box can be uploaded to the cloud through the Internet of Things device to enable users to access data in real time throughout the process.

[0061] 6. The patent of this invention also has powerful data recording and analysis functions, which can record key data of spraying operations in real time, such as spray volume, spray time, crop growth, etc., and store these data in the cloud or local server. Farmers and managers can use data analysis software to conduct in-depth analysis and mining of these data, so as to understand the growth status of crops, the occurrence patterns of pests and diseases, and other information, providing a strong reference for subsequent agricultural management and pest control.

[0062] 7. The patent of this invention realizes the precise spraying and efficient operation of common crop diseases and insect pests through the organic combination of machine vision, sensor technology, autonomous navigation and path planning, further filling the blank field of precise spraying and efficient operation of crops in China, and is an important part of realizing the national agricultural modernization construction and promoting the sustainable development strategy. In addition, this product improves the use efficiency and operation efficiency of pesticides, reduces pesticide waste and environmental pollution, and can make positive contributions to the sustainable development of modern agriculture. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 The overall three-dimensional structure diagram of the unmanned ground automatic target spraying vehicle in the present invention is

[0064] Figure 2The overall front view of the unmanned ground automatic target spraying vehicle in the present invention

[0065] Figure 3 The overall bottom view of the unmanned ground automatic target spraying vehicle in the present invention

[0066] Figure 4 for Figure 1 Schematic diagram of the three-dimensional structure of the robotic arm

[0067] Figure 5 for Figure 1 Exploded view of the robotic arm

[0068] Figure 6 for Figure 1 Schematic diagram of the three-dimensional structure of the steering mechanism

[0069] Figure 7 for Figure 1 Exploded diagram of the steering mechanism

[0070] Figure 8 for Figure 1 Schematic diagram of the steering mechanism

[0071] Fig. 9 The working process of the unmanned ground automatic target spraying vehicle in the present invention is

[0072] Fig.10 for Figure 1 The workflow of the pore distribution detection system

[0073] Fig.11 for Figure 1 Human-computer interaction interface work content

[0074] Fig.12 for Fig.11 How the human-computer interface works DETAILED DESCRIPTION

[0075] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is further described below in conjunction with embodiments and drawings, but the embodiments of the present invention are not limited thereto.

[0076] The unmanned ground automatic target spraying vehicle in this embodiment includes a pore distribution detection system, a target control system, a spraying system,

[0077] Power system and human-computer interface design;

[0078] The pore distribution detection system includes an information collection module and an airborne processor. The information collection module includes a fisheye camera for obtaining crop image information, an ultrasonic sensor for measuring the distance between the plant and the trolley, an optical sensor for monitoring the surface characteristics of the plant, a Beidou positioning module for obtaining crop theorem information, and a laser radar for obtaining three-dimensional point cloud information of the plant; the airborne processor is used to process the agricultural information data obtained by the information collection module, and then establish a plant distribution map model through a spatial statistical algorithm; the agricultural information data includes the location and degree of pests and diseases, ridge spacing, ridge length, plant height, and plant growth cycle, etc.;

[0079] The target control system includes a robotic arm module and an airborne control module; the robotic arm module includes a driving servo for driving its reasonable operation and a base, needle bearings and upper and lower arms that support the robotic arm mechanism; the airborne control module is used to receive the plant distribution map model obtained by the airborne processor, calculate the corresponding time signal, servo PWM signal, water pump voltage signal according to the logistic regression algorithm, spatial regression algorithm, etc., and transmit it to the human-machine interface while transmitting it to the spraying system and power system for effective operation.

[0080] The spraying system includes a lifting platform, a medicine box, a water pump, a pipeline system, a nozzle, an anti-drift cover and an electromagnetic valve. The lifting platform cooperates with the mechanical arm to complete the spraying work at different heights; the medicine box is equipped with a perforated partition, which can reduce the unstable walking of the vehicle body and the uneven spraying of pesticides caused by the shaking of the vehicle when walking. There are gaps between the partitions to make the pesticides more evenly mixed, and the medicine box contains a water level sensor for real-time monitoring of the water level of the medicine, a flow rate sensor for real-time monitoring of the flow rate and flow of the medicine, etc., and these data will be fed back to the human-computer interaction interface in real time; the nozzle is set at the top of the main mechanical arm and the auxiliary mechanical arm, and also includes a laser radar sensor for monitoring the distance between the nozzle and the plant, which is electrically connected to the airborne control module; the anti-drift cover adopts a prism structure with a wide top and a narrow bottom, which is innovatively placed on the unmanned vehicle, and can effectively cooperate with the auxiliary mechanical arm to reduce the waste of pesticide drift; the electromagnetic valve is used to control the spray volume of the nozzle, which is set on the nozzle and electrically connected to the airborne control module.

[0081] The power system is mainly composed of a steering adjustment system and a forward and backward system for executing operations. The steering adjustment system includes a rack, a pulley, a connecting rod and a wheel bracket, and is driven by a stepper motor; the forward and backward system for executing operations include four wheels arranged in an outward-facing O-shaped shape, and are driven by a DC brushed motor.

[0082] The human-computer interaction interface mainly includes a real-time monitoring interface and a remote monitoring interface. The real-time monitoring interface includes equipment parameter monitoring, drug management, pest and disease monitoring, and plant status monitoring. Equipment parameter monitoring focuses on real-time data and historical analysis; the drug management interface includes record forms, historical record viewing, and usage suggestions; pest and disease monitoring presents uploads, identification results, and early warning prompts; plant status monitoring displays images, maturity analysis, and management suggestions; the remote control interface is mainly used for emergency stops of the trolley in emergencies.

[0083] The pore detection system comprises the following steps:

[0084] The agricultural information data of the operation area is collected through the information collection module of the small vehicle, and the agricultural information data includes the location and degree of pests and diseases, ridge spacing, ridge length, plant height and plant growth cycle, etc.;

[0085] The location and extent of pests and diseases are processed by the onboard processor to obtain a pest and disease distribution map; the pore distribution and specific location of the plants are obtained by combining the image information obtained by the fisheye camera and the three-dimensional point cloud information obtained by the lidar;

[0086] Combined with spatial statistical algorithms, the ideal grid prescription map of pests and diseases is obtained after linear processing of the pest and disease distribution map, and the ideal grid prescription map of plants is obtained after linear processing of the specific positions of plants, and the two are transmitted together to the airborne control module;

[0087] The onboard control module obtains the location and degree of pests and diseases in each grid, the specific location of the plants, determines a reasonable path planning, generates corresponding work instructions and transmits them to the robotic arm module, spraying system and power system to guide them to complete the target action, spraying task and walking work;

[0088] The onboard control module is provided with a memory, which is used to store the data acquired by the information collection module, mainly a plant distribution map model established based on agricultural information data; in this way, on the one hand, it can be used to make a comprehensive and multi-angle evaluation of the spraying work after spraying, and on the other hand, if a special optimized spraying algorithm is set in the control system, it can be used to optimize the next spraying.

[0089] Furthermore, considering the pore distribution, the spraying amount compensation of each nozzle is:

[0090] q'1=-αρx+q

[0091] Among them, the spraying amount of each nozzle after compensation is q'1, α is the porosity compensation coefficient, and its value range is 0-1; the spraying amount of a single nozzle in the ideal grid prescription map is x, the internal porosity of the grid is ρ, and the spraying amount of each nozzle before compensation is q.

[0092] See also Figure 4 and Figure 5 The most important mechanism in the target control system, namely the mechanical arm mechanism, includes a main mechanical arm and a secondary mechanical arm, wherein the main mechanical arm is driven by three steering gears 2-3, supported by a base 2-6, a mechanical arm upper arm 2-4, and a mechanical lower arm 2-5. The top steering gear is mainly used to control the nozzle 2-1 to achieve two-dimensional movement in the plane, and the two bottom steering gears are used to assist the nozzle to reach the operating position. Needle bearings 2-7 are installed at the connection between the steering gear and the mechanical arm to reduce friction and make the mechanical arm move more smoothly. The mechanical arm uses a three-steering gear structure to enable the nozzle to flexibly and accurately reach the designated position to start operation.

[0093] Furthermore, the auxiliary robotic arm module 6 is arranged on both sides of the frame; the auxiliary robotic arm is driven by a steering gear and can cooperate with the main robotic arm module.

[0094] See also Figure 6 and Figure 7 The spraying system includes a lifting platform, a medicine box, a water pump, a pipe system, a nozzle, an anti-drift cover and an electromagnetic valve. The upper end of the lifting platform is equipped with two main mechanical arm modules; the medicine box is fixed to the rear end of the trolley; the nozzle is fixed to the top of the main mechanical arm and the auxiliary mechanical arm; the anti-drift cover adopts a prism-like structure with a wide top and a narrow bottom, and is fixed on the nozzle; the electromagnetic valve is used to control the spray volume of the nozzle, is set on the nozzle, and is electrically connected to the onboard control module.

[0095] Furthermore, a reflow groove is installed at the nozzles of the main robotic arm and the auxiliary robotic arm, and an interface specially designed to be connected to the reflow groove is provided in the medicine box.

[0096] See also Figure 6-Figure 8 The most important mechanism in the power system, namely the steering mechanism, is composed of a stepper motor 9-1 driving multiple connecting rods. Among them, the connecting rod 9-2 cooperates with the stepper motor 9-1, which is composed of a slider and a rack connected by a solid state, and is driven by the gear on the stepper motor 9-1 to slide on the slide groove 9-4. The connection between the connecting rod 9-2 and the wheel bracket 9-5 is achieved through the connecting rod 9-3. The whole steering system realizes steering by driving the connecting rod 9-2 to move left and right through the stepper motor 9-1 and then transmitting the force to the wheel bracket 9-5 through the connecting rod 9-3.

[0097] See also Fig.11 Fig.12, the main working contents of the human-computer interaction interface include real-time monitoring interface and remote control interface, among which the purpose of real-time monitoring interface is to display equipment parameters, identify the degree of pests and diseases, manage the amount of drugs, and monitor the status of plants, and the purpose of remote monitoring interface is to handle the control of the trolley in emergency situations. These design contents facilitate the operator to manage equipment and perform operations reasonably and effectively; the human-computer interaction interface establishes communication with the back-end through WebSocket in real time, receives the position, speed and other information of the unmanned ground automatic target spraying vehicle in real time, and displays it on the interface. The trolley on the back end is equipped with Beidou positioning module and multiple sensors, which upload the data to the back-end server in real time. The server processes the data in real time and pushes updates to the front end through WebSocket. By calling the API of WeChat applet, the data of each sensor is obtained and displayed in real time on the interface. Use the chart library to display historical data analysis. After receiving the sensor data, it is stored in the database, and an API is provided for the front end to request data. Equipment parameters, use and application conditions are recorded through forms, etc. After receiving the data submitted by the front end, it is stored in the database. Upload crop images, call the image recognition interface for analysis, and display the recognition results. Use the machine learning model to analyze the uploaded images, identify pests and diseases, and return the results. Call the sensor to obtain the plant status, use the sensor data processing technology to analyze the plant status and return the result.

[0098] In this embodiment, all motors and circuits are turned on and off by the remote control integrated circuit. In addition, the visual acquisition algorithm of the camera is prepared in advance when the action starts. Since the fully automatic spraying function is realized, the priorities of the spraying vehicles have been reasonably set, but the main switch of the remote control switch has the highest priority. Therefore, in the event of an accident, the main switch can be turned off to stop the operation of the spraying vehicle.

[0099] See also Figure 1-Figure 12 The working method of the unmanned ground automatic target spraying vehicle in this embodiment includes the following steps:

[0100] Step 1: The trolley supports both automatic mode and remote control mode. Before the spraying work begins, the trolley can reach the area to be sprayed through the remote control mode.

[0101] Step 2: When entering the ridge, the fisheye camera, ultrasonic sensor, optical sensor, Beidou positioning module and lidar in the pore distribution detection system carried by the vehicle start to power on and obtain external agricultural information data.

[0102] The agricultural information data include the location and extent of pests and diseases, ridge spacing, ridge length, plant height and plant growth cycle, etc.;

[0103] Step 3: Send the agricultural information data to the onboard processor, which processes it to obtain the distribution map of pests and diseases, the distribution of pores and the specific location model of the plants;

[0104] Step 4: Combined with the spatial statistical algorithm, the pest and disease distribution map is linearly processed to obtain the ideal grid prescription map of pests and diseases, and the specific position of the plant is linearly processed to obtain the ideal grid prescription map of the plant. The two are transmitted together to the airborne control module to facilitate the calculation of the optimal path planning and spraying intensity.

[0105] Furthermore, considering the pore distribution, the spraying amount compensation of each nozzle is:

[0106] q'1=-αρx+q

[0107] Among them, the spraying amount of each nozzle after compensation is q'1, α is the porosity compensation coefficient, and its value range is 0-1; the spraying amount of a single nozzle in the ideal grid prescription map is x, the internal porosity of the grid is ρ, and the spraying amount of each nozzle before compensation is q.

[0108] Step 5: The onboard processor calculates the corresponding results based on the obtained data, outputs PWM signals, high and low level signals, time signals, etc., and transmits them to the robotic arm module, spraying system and power system, and the car starts working.

[0109] Step 6: During operation, the trolley transmits all agricultural information data, the working environment of the trolley, local weather conditions, weather warning conditions, future weather changes, etc. to the human-computer interaction interface through a remote transmission module (WIFI module, Bluetooth module, etc.) so that the operator can make the next plan. The above is a preferred embodiment of the present invention, but the embodiment of the present invention is not limited to the above one. Any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principle of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. An unmanned ground automatic target spraying vehicle capable of real-time detection of crop canopy pore structure and pest distribution, comprising a pore distribution detection system, a target control system, a spraying system, a power system and a human-computer interaction interface design: the pore distribution detection system comprises an information collection module and an onboard processor; the target control system comprises a mechanical arm module and an onboard control module; the spraying system comprises a lifting platform, a medicine box, a water pump, a pipeline system, a spray head, an anti-drift cover and an electromagnetic valve; the power system comprises a steering adjustment system and a forward and backward system; the human-computer interaction interface design comprises a real-time monitoring interface and a remote control interface; 2. The unmanned ground automatic target spraying vehicle according to claim 1, Features: The pore distribution detection system includes an information collection module and an onboard processor. The information collection module includes a fisheye camera for obtaining crop image information, an ultrasonic sensor for measuring the distance between the plant and the trolley, an optical sensor for monitoring the surface characteristics of the plant, a Beidou positioning module for obtaining crop theorem information, and a laser radar for obtaining three-dimensional point cloud information of the plant; the onboard processor is used to process the agricultural information data obtained by the information collection module, and then establish a plant distribution map model through a spatial statistical algorithm; 3. The unmanned ground automatic target spraying vehicle according to claim 1, Features: The target control system includes a robotic arm module and an airborne control module; the robotic arm module includes a driving servo for driving its reasonable operation and a base, needle bearings and upper and lower arms that support the robotic arm mechanism; the airborne control module is used to receive the plant distribution map model obtained by the airborne processor, calculate the corresponding time signal, servo PWM signal, water pump voltage signal according to the logistic regression algorithm, spatial regression algorithm, etc., transmit it to the human-machine interaction interface for the user to view, and at the same time pass it to the spraying system and power system for effective operation.

4. The unmanned ground automatic target spraying vehicle according to claim 1, Features: The spraying system includes a lifting platform, a medicine box, a water pump, a pipeline system, a nozzle, an anti-drift cover and an electromagnetic valve. The lifting platform cooperates with the mechanical arm to complete the spraying work at different heights; the medicine box is equipped with a perforated partition, which can reduce the unstable walking of the vehicle body and the uneven spraying of pesticides caused by the shaking of the vehicle when walking. There are gaps between the partitions to make the pesticides more evenly mixed, and the medicine box contains a water level sensor for real-time monitoring of the water level of the medicine, a flow rate sensor for real-time monitoring of the flow rate and flow of the medicine, etc., and these data will be fed back to the human-computer interaction interface in real time; the nozzle is set at the top of the main mechanical arm and the auxiliary mechanical arm, and also includes a laser radar sensor for monitoring the distance between the nozzle and the plant, which is electrically connected to the airborne control module; the anti-drift cover adopts a prism structure with a wide top and a narrow bottom, which is innovatively placed on the unmanned vehicle, and can effectively cooperate with the auxiliary mechanical arm to reduce the waste of pesticide drift; the electromagnetic valve is used to control the spray volume of the nozzle, which is set on the nozzle and electrically connected to the airborne control module.

5. The unmanned ground automatic target spraying vehicle according to claim 1, Features: The power system is mainly composed of a steering adjustment system and a forward and backward system for performing operations. The steering adjustment system includes a rack, a pulley, a connecting rod and a wheel bracket, and is driven by a stepper motor; the forward and backward system for performing operations include four wheels with a wheel camber angle of -15°, and are driven by a DC brush motor.

6. The unmanned ground automatic target spraying vehicle according to claim 1, Features: The human-computer interaction interface mainly includes a real-time monitoring interface and a remote monitoring interface. The real-time monitoring interface includes equipment parameter monitoring, drug management, pest and disease monitoring, and plant status monitoring. Equipment parameter monitoring focuses on real-time data and historical analysis; the drug management interface includes record forms, historical record viewing, and usage suggestions; pest and disease monitoring presents upload, identification results, and early warning prompts; Plant status monitoring displays images, maturity analysis and management recommendations; The remote control interface is mainly used for emergency stopping of the trolley in unexpected situations and operation of the trolley in remote control mode.

7. The pore detection system according to claim 2, The following steps are involved: The agricultural information data of the operation area is collected through the information collection module of the small vehicle, and the agricultural information data includes the location and degree of pests and diseases, ridge spacing, ridge length, plant height and plant growth cycle, etc.; The location and extent of pests and diseases are processed by the onboard processor to obtain a pest and disease distribution map; the pore distribution and specific location of the plants are obtained by combining the image information obtained by the fisheye camera and the three-dimensional point cloud information obtained by the lidar; Combined with spatial statistical algorithms, the ideal grid prescription map of pests and diseases is obtained after linear processing of the pest and disease distribution map, and the ideal grid prescription map of plants is obtained after linear processing of the specific positions of plants, and the two are transmitted together to the airborne control module; The onboard control module obtains the location and degree of pests and diseases in each grid, the specific location of the plants, determines a reasonable path planning, generates corresponding work instructions and transmits them to the robotic arm module, spraying system and power system to guide them to complete the target action, spraying task and walking work; 8. The target control system according to claim 3, It is characterized in that The onboard control module is provided with a memory, which is used to store the data obtained by the information collection module, mainly the plant distribution map model established based on the agricultural information data; in this way, on the one hand, it can be used to make a comprehensive and multi-angle evaluation of the spraying work after spraying, and on the other hand, if a special optimized spraying algorithm is set in the control system, it can be used to optimize the next spraying.

9. According to the pore detection system of claim 8, taking into account the pore distribution, the spraying amount compensation of each nozzle is: q‘ 1 =-αρx+q in, The spraying amount of each nozzle after compensation is q' 1 , α is the porosity compensation coefficient, and its value range is 0~1; the spray volume of a single nozzle in the ideal grid prescription map is x, the internal porosity of the grid is ρ, and the spray volume of each nozzle before compensation is q.

Citation Information

Patent Citations

  • Automatic profile modeling sprayer applicable to citrus orchard terrain

    CN108849830A

  • Unmanned aerial vehicle sugarcane disease and insect pest prevention and control type pesticide spraying method based on accurate meteorological support technology

    CN110050619A

  • Spray drifting prevention device for boom sprayer

    CN115088697A

  • Unmanned aerial vehicle variable spraying system and method capable of adjusting spraying width

    CN116806792A

  • Anti-wear steering mechanism of sightseeing train

    CN219428122U

Cited By

  • Field double-layer pesticide spraying robot applied to potatoes and control method

    CN120021603A

  • Spraying device and spraying control method

    CN120477165A