Operation method of plant protection robot based on environment perception
By installing sensor modules on the plant protection robot and using environmental perception algorithms and machine learning models for data processing and path optimization, the problem of traditional plant protection robots lacking environmental perception and path planning optimization is solved, and efficient and accurate spraying operations and resource utilization are achieved.
Patent Information
- Application Number
- CN202411596746.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-11
- Publication Date
- 2025-03-14
AI Technical Summary
Traditional plant protection robots lack environmental perception capabilities, resulting in poor spraying effect, low resource utilization efficiency, and under-optimized path planning, and unsatisfactory efficiency and results.
By installing multiple sensor modules to collect environmental data in real time, and using environment perception algorithms and machine learning models for data processing and path optimization, dynamically adjusting the job path and spraying volume.
It improves the accuracy and operating efficiency of spraying, reduces the use of pesticides and its impact on the environment, improves resource utilization efficiency, and adapts to complex terrain and real-time environmental changes.
Smart Images
Figure CN119610146A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of agriculture, and in particular to an operation method of a plant protection robot based on environment perception. Background Art
[0002] With the development of global agriculture and the continuous advancement of technology, the demand for efficient and accurate plant protection operations in modern agricultural production is growing. Plant protection operations involve the spraying and management of pesticides to prevent and control crop diseases and pests and ensure the healthy growth of crops. However, traditional plant protection operations usually rely on manual operations or mechanized equipment. These methods are inefficient and labor-intensive when facing large areas of farmland, and are prone to excessive or insufficient use of pesticides, which not only increases production costs, but also causes certain harm to the environment and human health.
[0003] In recent years, plant protection robots, as a new type of agricultural automation equipment, have received widespread attention in plant protection operations. These robots operate through preset paths and fixed spraying patterns. Although they have improved the operation efficiency to a certain extent, they still have the following problems:
[0004] Lack of environmental perception: Traditional crop protection robots usually follow a predetermined path and spray at fixed times and locations, lacking the ability to perceive the real-time environment. This means they cannot dynamically adjust according to actual environmental changes or crop growth conditions, resulting in poor spraying effects.
[0005] Inadequate resource utilization: Due to the lack of real-time perception and dynamic adjustment capabilities, traditional plant protection robots may over-spray pesticides in some areas and under-spray in other areas, resulting in inefficient resource utilization and potential negative impacts on the environment.
[0006] Poorly optimized path planning: The path planning of traditional plant protection robots is relatively simple, usually a preset fixed route, which fails to fully consider the complex terrain and real-time environmental changes in the operating area, resulting in unsatisfactory operating efficiency and results.
[0007] In response to the above problems, researchers and enterprises have begun to explore and develop plant protection robots based on environmental perception and intelligent control. These new plant protection robots collect environmental data in real time by installing multiple sensor modules, and use advanced algorithms for data processing and path optimization to achieve intelligent management and dynamic adjustment of the operation process. Therefore, there is an urgent need for a plant protection robot operation method based on environmental perception to solve the problems of the above traditional plant protection robots. Summary of the invention
[0008] The purpose of the present invention is to overcome the above-mentioned deficiencies in the prior art and to provide an operating method of a plant protection robot based on environmental perception, which can perceive environmental changes in real time, dynamically adjust the operating path and spraying amount, improve operating efficiency and accuracy, and reduce the use of pesticides and their impact on the environment.
[0009] The technical solution adopted by the present invention to solve the above problem is: an operation method of a plant protection robot based on environmental perception, comprising the following steps:
[0010] S1. Install the sensor module: the sensor module includes any one or a combination of a temperature and humidity sensor, a light sensor, a soil moisture sensor and a gas sensor;
[0011] S2, real-time data collection and pre-processing: real-time collection of environmental data, and data cleaning and pre-processing through the data processing unit;
[0012] S3, generate environmental information: based on the preprocessed data, use the environmental perception algorithm to integrate the preprocessed data to generate detailed environmental information of the work area;
[0013] S4, Path Planning: Based on environmental information, use machine learning models to analyze the operating area, determine the optimal path for the plant protection robot, and dynamically generate a path planning solution based on environmental information;
[0014] S5. Execute spraying operation: Based on the path planning scheme, control the plant protection robot to execute spraying operation along the optimized path.
[0015] Preferably: the sensor module described in S1 and the data processing unit transmit data via a wireless communication module, the data cleaning module includes a data cleaning module for removing abnormal data and noise, a data preprocessing module for data normalization processing, and a data fusion module for fusing multiple sensor data, and the wireless communication module adopts Wi-Fi or LoRa communication protocol.
[0016] Preferably, the S2 comprises the following steps:
[0017] S2.1. Abnormal data detection: The DBSCAN clustering algorithm is used to detect and remove abnormal data. The algorithm parameters include a distance threshold of 0.5 and a minimum sample number of 5.
[0018] S2.2. Data normalization: The data is processed using the maximum and minimum normalization method to ensure comparability between different data sources. The normalization range is 0 to 1.
[0019] Preferably, the S3 comprises the following steps:
[0020] S3.1, Data fusion: Use weighted average method to fuse data from different sensors, where the weight coefficients are 0.3 for temperature and humidity sensor, 0.2 for light sensor, 0.3 for soil moisture sensor, and 0.2 for gas sensor;
[0021] S3.2, Model training: A model based on random forest regression is used to train historical environmental data to generate an environmental prediction model. The training parameters include the number of trees as 100, the maximum depth as 30, and the minimum number of sample splits as 2;
[0022] S3.3, real-time prediction: Use the environmental prediction model generated in S3.2 to predict real-time environmental data and generate environmental information of the working area.
[0023] Preferably, the S4 comprises the following steps:
[0024] S4.1, Path Optimization: Genetic algorithm is used to optimize the operation path of the plant protection robot. The genetic algorithm parameters include population size of 50, mutation rate of 0.01, crossover rate of 0.8, and number of iterations of 100 times;
[0025] S4.2. Dynamic adjustment: Based on real-time environmental data and crop growth characteristics, dynamically adjust path planning to cope with environmental changes.
[0026] Preferably: the machine learning model is used to analyze the environmental information of the working area and determine the optimization path, and is trained using a deep neural network algorithm. The training parameters include a learning rate of 0.001, a batch size of 64, a number of training rounds of 50, and a model training data set containing at least 10,000 records.
[0027] Preferably, the spraying operation in S5 is implemented by an intelligent spraying system, and the intelligent spraying system can dynamically adjust the spraying amount and spraying time according to real-time environmental data.
[0028] Preferably: the intelligent spraying system includes a solenoid valve for adjusting the spraying amount, an ultrasonic sensor for detecting the spraying range, a timer module for controlling the spraying time, and an image sensor for real-time monitoring of the spraying effect, and dynamically adjusts the spraying amount and spraying time through real-time environmental data.
[0029] Compared with the prior art, the present invention has the following advantages and effects:
[0030] The present invention uses a variety of sensor modules to collect environmental data in real time, and uses advanced environmental perception algorithms and machine learning models for data processing and path optimization. By real-time monitoring of environmental data and dynamically adjusting the operation path and spraying amount, it is ensured that the plant protection robot operates at the most appropriate time and location, thereby improving the accuracy and efficiency of spraying. Based on the crop growth status and real-time environmental data, the intelligent spraying system can accurately control the amount of pesticide used, reduce the waste and excessive use of pesticides, thereby reducing costs and reducing pollution to the environment. Through environmental perception and path optimization, the problems of over-spraying and under-spraying are effectively avoided, and the utilization efficiency of resources is improved. Intelligent path planning and dynamic adjustment functions enable the plant protection robot to adapt to complex terrain and real-time environmental changes, ensuring the continuity and efficiency of operations. Reducing the amount of pesticide used and the impact on the environment will help achieve sustainable agricultural development and protect the ecological environment and farmland health. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] Figure 1 It is a flow chart of an operation method of a plant protection robot based on environment perception according to an embodiment of the present invention.
[0032] Figure 2 Schematic diagram of the distribution of sensor modules according to an embodiment of the present invention.
[0033] Figure numbers: temperature and humidity sensor 1, light sensor 2, soil moisture sensor 3, gas sensor 4. DETAILED DESCRIPTION
[0034] The present invention will be further described in detail below with reference to the accompanying drawings and by way of examples. The following examples are intended to explain the present invention but the present invention is not limited to the following examples.
[0035] Example: See Figure 1 and Figure 2 In this embodiment, a method for operating a plant protection robot based on environmental perception is specifically used for pest control and crop protection in farmland, aiming to improve the efficiency and accuracy of plant protection and reduce the amount of pesticides used and their impact on the environment. The method specifically includes the following steps:
[0036] S1. Install the sensor module
[0037] Multiple sensor modules are installed on the plant protection robot. These sensor modules include:
[0038] Temperature and humidity sensor 1: used to detect the temperature and humidity of the environment and provide real-time climate data. Temperature and humidity sensor 1 is installed on the top of the robot to ensure that the temperature and humidity changes of the air can be accurately monitored.
[0039] Light sensor 2: used to detect the light intensity of the environment and help evaluate the photosynthesis of crops. Light sensor 2 is arranged on the front and side of the robot to comprehensively monitor the light conditions received by the crops.
[0040] Soil moisture sensor 3: used to monitor soil moisture and ensure the water requirements of crop roots. Soil moisture sensor 3 is buried in the soil around the roots of crops to accurately measure the moisture level of the soil.
[0041] Gas Sensor 4: Used to detect gas components in the air, such as carbon dioxide and ammonia, to help determine the health of crops. Gas Sensor 4 is installed in multiple locations on the robot to ensure comprehensive monitoring of air quality.
[0042] S2. Real-time data collection and preprocessing
[0043] The sensor module collects environmental data in real time at a frequency of 10 times per second. The collected data is transmitted to the data processing unit through the wireless communication module, and the data processing unit performs the following preprocessing steps:
[0044] S2.1, Data cleaning: Use the DBSCAN clustering algorithm to detect and remove abnormal data. The algorithm parameters include a distance threshold of 0.5 and a minimum sample number of 5. By detecting abnormal data, the accuracy and reliability of sensor data are ensured.
[0045] S2.2, Data normalization: The maximum and minimum normalization method is used to process the data to ensure the comparability between different data sources, and the normalization range is 0 to 1. Normalization processing helps to eliminate the dimensional differences of different sensor data and improve the effect of data fusion.
[0046] S3. Generate environmental information
[0047] Based on the preprocessed data, the environmental perception algorithm is used to fuse and analyze the data to generate detailed environmental information of the work area. The specific steps include:
[0048] S3.1, Data fusion: The data from different sensors are fused using the weighted average method, with the weight coefficients being 0.3 for temperature and humidity sensor, 0.2 for light sensor, 0.3 for soil moisture sensor, and 0.2 for gas sensor. The weighted average method can effectively integrate various environmental data and generate accurate environmental information.
[0049] S3.2, Model training: A model based on random forest regression is used to train historical environmental data to generate an environmental prediction model. The training parameters include the number of trees as 100, the maximum depth as 30, and the minimum number of sample splits as 2. The random forest model has strong predictive ability and robustness.
[0050] S3.3, Real-time prediction: Use the environmental prediction model trained in S3.2 to predict the real-time environmental data and generate detailed environmental information of the operation area. The environmental prediction model can be updated in real time to ensure that the environmental information of the operation area is accurate and reliable.
[0051] S4. Path Planning
[0052] Based on the generated environmental information, the machine learning model is used to analyze the operating area and determine the optimal path for the plant protection robot. The path planning scheme is dynamically generated and adjusted in real time to cope with environmental changes. The specific steps include:
[0053] S4.1, Path Optimization: Genetic algorithm is used to optimize the operation path of the plant protection robot. The genetic algorithm parameters include population size of 50, mutation rate of 0.01, crossover rate of 0.8, and number of iterations of 100. The genetic algorithm simulates the biological evolution process to find the optimal path planning solution.
[0054] S4.2, Dynamic Adjustment: Dynamically adjust the path planning according to real-time environmental data and crop growth characteristics to ensure the efficiency and accuracy of the operation. Dynamic adjustment can adapt to environmental changes and improve the adaptability and flexibility of path planning.
[0055] S5. Perform spraying operations
[0056] According to the optimized path planning scheme, the plant protection robot is controlled to perform spraying operations along the optimized path. The spraying operation is achieved through the intelligent spraying system, which includes:
[0057] Solenoid valve: used to adjust the spraying amount. The solenoid valve can accurately control the spraying amount of pesticides to avoid waste.
[0058] Ultrasonic sensor: used to detect the spraying range. Ultrasonic sensors can monitor the spraying range in real time to ensure uniform coverage.
[0059] Timer module: used to control the spraying time. The timer module accurately controls the spraying time according to the set spraying plan.
[0060] Image sensor: used to monitor spraying effects in real time. Image sensors can provide real-time feedback on spraying effects to ensure optimal spraying effects for each area and crop.
[0061] The intelligent spraying system can dynamically adjust the spraying amount and spraying time according to real-time environmental data to ensure the best spraying effect for each area and crop. Through the above steps, the present invention realizes environmental perception and intelligent control of the plant protection robot, improves operating efficiency and accuracy, reduces the amount of pesticide used and its impact on the environment, and has broad application prospects and important practical significance.
[0062] The above contents described in this specification are merely examples of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in similar ways, as long as they do not deviate from the contents of the present specification or exceed the scope defined by the claims, they shall all fall within the protection scope of the present invention.
Claims
1. A method for operating a plant protection robot based on environmental perception, characterized in that: The following steps are involved: S1. Install the sensor module: the sensor module includes any one or a combination of a temperature and humidity sensor, a light sensor, a soil moisture sensor and a gas sensor; S2, real-time data collection and pre-processing: real-time collection of environmental data, and data cleaning and pre-processing through the data processing unit; S3, generate environmental information: based on the preprocessed data, use the environmental perception algorithm to integrate the preprocessed data to generate detailed environmental information of the work area; S4, Path Planning: Based on environmental information, use machine learning models to analyze the operating area, determine the optimal path for the plant protection robot, and dynamically generate a path planning solution based on environmental information; S5. Execute spraying operation: Based on the path planning scheme, control the plant protection robot to execute spraying operation along the optimized path.
2. The operation method of the plant protection robot based on environment perception according to claim 1, characterized in that: The sensor module described in S1 and the data processing unit transmit data via a wireless communication module. The data cleaning module includes a data cleaning module for removing abnormal data and noise, a data preprocessing module for data normalization, and a data fusion module for fusing multiple sensor data. The wireless communication module adopts Wi-Fi or LoRa communication protocol.
3. The operation method of the plant protection robot based on environment perception according to claim 1, characterized in that: The S2 comprises the following steps: S2.
1. Abnormal data detection: The DBSCAN clustering algorithm is used to detect and remove abnormal data. The algorithm parameters include a distance threshold of 0.5 and a minimum sample number of 5. S2.
2. Data normalization: The data is processed using the maximum and minimum normalization method to ensure comparability between different data sources. The normalization range is 0 to 1.
4. The operation method of the plant protection robot based on environment perception according to claim 1, characterized in that: The S3 comprises the following steps: S3.1, Data fusion: Use weighted average method to fuse data from different sensors, where the weight coefficients are 0.3 for temperature and humidity sensor, 0.2 for light sensor, 0.3 for soil moisture sensor, and 0.2 for gas sensor; S3.2, Model training: A model based on random forest regression is used to train historical environmental data to generate an environmental prediction model. The training parameters include the number of trees as 100, the maximum depth as 30, and the minimum number of sample splits as 2; S3.3, real-time prediction: Use the environmental prediction model generated in S3.2 to predict real-time environmental data and generate environmental information of the working area.
5. The operation method of the plant protection robot based on environment perception according to claim 1, characterized in that: The S4 comprises the following steps: S4.1, Path Optimization: Genetic algorithm is used to optimize the operation path of the plant protection robot. The genetic algorithm parameters include population size of 50, mutation rate of 0.01, crossover rate of 0.8, and number of iterations of 100 times; S4.
2. Dynamic adjustment: Based on real-time environmental data and crop growth characteristics, dynamically adjust path planning to cope with environmental changes.
6. The operation method of the plant protection robot based on environment perception according to claim 1, characterized in that: The machine learning model is used to analyze the environmental information of the working area and determine the optimization path. It is trained using a deep neural network algorithm. The training parameters include a learning rate of 0.001, a batch size of 64, and a number of training rounds of 50. The model training data set contains at least 10,000 records.
7. The operation method of the plant protection robot based on environment perception according to claim 1, characterized in that: The spraying operation in S5 is achieved through an intelligent spraying system, which can dynamically adjust the spraying amount and spraying time according to real-time environmental data.
8. The operation method of the plant protection robot based on environment perception according to claim 7, characterized in that: The intelligent spraying system includes a solenoid valve for adjusting the spraying amount, an ultrasonic sensor for detecting the spraying range, a timer module for controlling the spraying time, and an image sensor for real-time monitoring of the spraying effect, and dynamically adjusts the spraying amount and spraying time through real-time environmental data.