Search

CN-122018402-A - Near field communication interaction-based intelligent plant maintenance method and system

CN122018402ACN 122018402 ACN122018402 ACN 122018402ACN-122018402-A

Abstract

The embodiment of the invention relates to the technical field of intelligent gardening and discloses a plant intelligent maintenance method based on near field communication interaction, which comprises the following steps that an intelligent terminal responds to a near field communication tag embedded in a plant maintenance device to establish near field communication connection; the intelligent terminal responds to the near field communication connection established with the near field communication tag to read curing state information from the microcontroller unit, responds to user input by the intelligent terminal and sends execution instructions to the microcontroller unit to control corresponding executors to carry out watering operation according to the curing state information and the user input. The device solves the technical problems of complex configuration, extensive maintenance decision, inconvenient data interaction, high execution risk, poor man-machine cooperativity and the like of the traditional plant maintenance device.

Inventors

  • Xie Jiehang

Assignees

  • 广州丽芳园林生态科技股份有限公司

Dates

Publication Date
20260512
Application Date
20260210

Claims (10)

  1. 1. A plant intelligent maintenance method based on near field communication interaction is characterized by comprising the following steps: Writing, by an intelligent terminal, a set of initial configuration parameters to a microcontroller unit of a plant care device in response to establishing a near field communication connection with a near field communication tag embedded in the plant care device, wherein the plant care device is disposed in a pot, the initial configuration parameters including a plant variety identifier, a pot volume, a matrix type identifier, and a security policy parameter; Periodically collecting, by the microcontroller unit of the plant care device, environmental data of a respective plant from an environmental sensor, wherein collecting the environmental data includes executing within a wake-up period of the microcontroller unit; Generating, by the microcontroller unit, a corresponding plant care decision based on the collected environmental data and the stored initial configuration parameters, wherein generating the corresponding plant care decision comprises: constructing a maintenance feature vector based on the environmental data; Inputting the maintenance feature vector to a pre-trained machine learning model stored in the microcontroller unit to obtain an initial recommended amount of water and an initial urgency; Performing constraint optimization on the initial suggested water quantity and the initial emergency degree based on the safety strategy parameters so as to determine a final water quantity and a final emergency degree; Reading maintenance state information from the microcontroller unit by the intelligent terminal in response to the near field communication connection established with the near field communication tag, and displaying the maintenance state information on the intelligent terminal of the user; And the intelligent terminal responds to the user input and sends an execution instruction to the micro-controller unit so as to control the corresponding executor to carry out watering operation according to the maintenance state information and the user input.
  2. 2. The intelligent maintenance plant method based on near field communication interaction of claim 1, wherein the establishing, by the intelligent terminal, a near field communication connection in response to the near field communication tag embedded in the plant maintenance device comprises: Triggering the microcontroller unit of the plant maintenance device to enter a communication mode through a near field communication tag so as to wake up the microcontroller unit; establishing a corresponding temporary communication session, wherein establishing the corresponding temporary communication session comprises: the intelligent terminal sends a session request frame to the microcontroller unit, wherein the session request frame comprises a terminal random number, a time stamp and intelligent terminal identification information; Generating a device random number and an incremental session counter by the microcontroller unit in response to the session request frame, and transmitting the device random number, the session counter and a device state abstract to the intelligent terminal; A unique session identifier for identifying a temporary communication session is commonly constructed by at least one of the intelligent terminal and the microcontroller unit based on the terminal nonce, the device nonce, and the session counter, wherein all communication frames during the temporary communication session carry the unique session identifier to prevent cross-session replay attacks.
  3. 3. The intelligent maintenance method of plants based on near field communication interaction of claim 1, wherein the environment sensor comprises a water content sensor, a temperature sensor, a conductivity sensor and an illumination sensor, and the periodically collecting environmental data of the corresponding plants from the environment sensor comprises: the microcontroller unit is awakened at regular time according to a preset sampling period, and raw data are acquired from the water content sensor, the temperature sensor, the conductivity sensor and the illumination sensor during the awakening period; The method comprises the steps of performing multi-factor compensation on an original reading of a water content sensor, wherein the compensation quantity is calculated based on synchronously acquired temperature data, conductivity data and a compensation calculation formula, and the compensation calculation formula is as follows: Wherein, the In order to correct the volume water content after correction, In order to achieve the original volume water content, For the temperature compensation coefficient to be a function of the temperature, For salinity compensation coefficient, EC is conductivity; t is the detection temperature; When multi-frequency measurement is used, a double-frequency differential estimation method is adopted to process the data of the water content sensor so as to inhibit coupling interference of soil salinity on water content measurement; Based on the compensated and/or processed sensor data, and combining the stored effective volume of the pot, the time and water quantity of last watering and the type identifier of the matrix, constructing a multidimensional feature vector for intelligent irrigation decision, wherein the multidimensional feature vector is used for representing the current water demand environment and the historical state of plants; the multidimensional feature vector is: ; wherein f is a multidimensional feature vector, In order to correct the volume water content after correction, The water content change rate is represented by T, the temperature parameter is represented by T, the electrical conductivity parameter is represented by EC, the illumination intensity is represented by L, Is the effective volume of the basin device, For the time stamp of the last time of watering, For the water quantity of the last time, Is of the matrix type.
  4. 4. The intelligent plant maintenance method based on near field communication interaction according to claim 3, wherein the pre-trained machine learning model is a lightweight multi-layer perceptron neural network with quantized weights, and the model weight data size is not more than 4 kilobytes; the multi-layer perceptron neural network includes: An input layer for receiving the multi-dimensional feature vector; at least one hidden layer for extracting a nonlinear characterization configured with a nonlinear activation function; The output layer is used for obtaining intermediate output, and the intermediate output is converted into an original watering emergency score and an original suggested water quantity through Sigmoid and Softplus respectively; the constraint optimization of the initial suggested amount of water and the initial urgency based on the safety policy parameters to determine a final amount of water and a final urgency includes: Determining corresponding predicted water quantity according to the initial suggested water quantity and a watering prediction formula, wherein the watering prediction formula is as follows: wherein, among them, In order to predict the water content of the water, For the water content of the current volume, The amount of water is initially suggested; for the effective volume of the basin device, Is the moisture absorption conversion coefficient; determining a dynamic target water content in a preset safe water content interval based on the initial urgency, the safety strategy parameters and a target mapping formula, wherein the target mapping formula is as follows: ; Wherein, the For the target water content to be the same as the water content, For a permanent wilting point, For the lower safety margin of the limit, For the water-holding capacity in the field, U is the initial emergency degree; as a lower limit of the target value, Is the target upper limit; and determining the optimized water quantity by taking the initial suggested water quantity as an optimized starting point and minimizing an objective function, wherein the objective function is as follows: Q is more than or equal to 0, wherein, The intensity is penalized for the water cut, Punishing intensity for risk of overstock; and constraining the watering amount obtained by the optimization solution to a feasible area defined by a preset safety strategy to determine the final watering amount.
  5. 5. The intelligent maintenance method of plants based on near field communication interactions of claim 4, wherein the over-casting penalty function is configured as a piecewise quadratic function, the over-casting penalty function being: Wherein, the In order to penalize the coefficients, Is a safe threshold of water pouring quantity, and , The amount of water is initially suggested; The determining the optimized water quantity by minimizing the objective function further comprises: calculating theoretical optimal watering quantity without penalty items; determining an optimized water quantity based on the theoretical optimal water quantity, a dynamic safe water quantity threshold value and an objective function; Applying the dynamic safe watering amount threshold value to the optimized watering amount as a hard upper limit constraint, and ensuring non-negative performance through a cutting function, wherein the cutting function is as follows: wherein the clipping function limits the input value to a lower limit of 0 and an upper limit Between them.
  6. 6. The intelligent maintenance method for plants based on near field communication interaction according to claim 1, wherein the reading maintenance status information from the micro controller unit comprises: The method comprises the steps of reading current maintenance state data locally calculated and stored by a micro-controller unit, wherein the current maintenance state data comprise current soil volume moisture content and change rate thereof, watering emergency degree, final watering quantity, device state abstract and maintenance history abstract, the final watering quantity is obtained after the micro-controller unit is subjected to safety constraint optimization, the device state abstract comprises electric quantity information, sensor abnormality marks and alarm grades, and the maintenance history abstract comprises a time stamp of last watering, watering quantity and log abstract of a designated item.
  7. 7. The intelligent plant curing method based on near field communication interaction according to claim 1, further comprising, after controlling the corresponding actuator to perform a watering operation according to the curing state information and a user input: Writing feedback data into the microcontroller unit by the intelligent terminal through the near field communication connection, wherein the feedback data comprises execution records, execution time stamps, actual execution water quantity, user confirmation marks and safety configuration parameter updating data of the watering operation, and the safety configuration parameter updating data comprises a permanent wilting point, field water holding capacity, an upper limit safety margin and a lower limit safety margin; Validating and storing, by the microcontroller unit, the feedback data to a local non-volatile memory; and updating the internal self-adaptive model parameters by the microcontroller unit based on the actual execution water quantity, the environmental data before and after watering and the time interval.
  8. 8. The intelligent maintenance method of plants based on near field communication interactions of claim 7, wherein the internal adaptive model parameters include an estimated effective water holding rate and an estimated transpiration rate, and the updating the internal adaptive model parameters includes: Calculating an updated estimated effective water content according to a first update formula, wherein the first update formula is as follows: Wherein, the To estimate the effective water holding rate before updating, For the updated estimated effective water holding rate, In order to achieve an effective water holding rate learning rate, In order to actually perform the watering amount, For the cumulative evapotranspiration volume calculated based on the estimated evapotranspiration rate during the interval between two watering, For the net change in soil volume moisture content observed during the interval, The effective volume of the basin; calculating an updated estimated rate of evaporation according to a second updated formula, the second updated formula being: Wherein, the To estimate the rate of evaporation before the update, For the updated estimated rate of evaporation, In order for the rate of transpiration to learn, For the currently observed volume water content of the soil, For the target soil volume moisture content, For clipping functions, for limiting the deviation value to a predetermined limit And (3) inner part.
  9. 9. The intelligent maintenance method of plants based on near field communication interactions as set forth in claim 7, wherein said intelligent maintenance method further comprises: Calculating the recovery amount of the water content of the soil volume in the first sensor sampling period after the watering operation; if the recovery quantity is smaller than the expected influence threshold value of the water casting quantity, judging that ventilation or permeation abnormality of the root zone exists, and improving the corresponding alarm grade; monitoring the change trend of the conductivity of the soil, judging that salt damage or fertilizer damage risks exist if the conductivity is detected to continuously increase in a preset period, and outputting corresponding risk grades and release suggestions; And/or, the intelligent maintenance method further comprises the following steps: Calculating a prediction error according to a deviation calculation formula in the next sensor sampling period after the watering operation is executed The deviation calculation formula is as follows: Wherein, the For one sampling period after watering The measured actual soil volume moisture content, To water the quantity based on execution Effective volume of basin Moisture absorption conversion coefficient The predicted water content after watering; When the prediction error And triggering the model retraining mark when the preset threshold value is exceeded.
  10. 10. Plant intelligent maintenance system based on near field communication interaction, characterized by comprising: The intelligent terminal is used for responding to the establishment of near field communication connection with a near field communication tag embedded in the plant maintenance device, and writing a set of initial configuration parameters into a microcontroller unit of the plant maintenance device, wherein the plant maintenance device is arranged in a pot, and the initial configuration parameters comprise a plant variety identifier, a pot volume, a matrix type identifier and a security policy parameter; The system comprises a microcontroller unit, an acquisition module, a control module and a control module, wherein the microcontroller unit is used for periodically acquiring environmental data of corresponding plants from an environmental sensor by the plant maintenance device, and the acquisition of the environmental data is performed in a wake-up period of the microcontroller unit; The generation module is used for generating corresponding plant maintenance decisions by the microcontroller unit based on the acquired environment data and the stored initial configuration parameters, wherein the generation module comprises the steps of constructing maintenance feature vectors based on the environment data, inputting the maintenance feature vectors into a pre-trained machine learning model stored in the microcontroller unit to acquire initial suggested water quantity and initial emergency degree, and carrying out constraint optimization on the initial suggested water quantity and the initial emergency degree based on the safety strategy parameters to determine final water quantity and final emergency degree; the communication module is used for responding to the near field communication connection established with the near field communication tag by the intelligent terminal, reading maintenance state information from the microcontroller unit and displaying the maintenance state information on the intelligent terminal of the user; And the operation module is used for responding to the user input by the intelligent terminal and sending an execution instruction to the microcontroller unit so as to control the corresponding executor to carry out watering operation according to the maintenance state information and the user input.

Description

Near field communication interaction-based intelligent plant maintenance method and system Technical Field The invention relates to the technical field of intelligent gardening, in particular to a plant intelligent maintenance method and system based on near field communication interaction. Background Currently, existing home plant maintenance mostly relies on experience or continuous networking equipment (Wi-Fi/BLE gateway). The WiFI scheme is easy to be overcast or lack water, the cost and maintenance (distribution network scheme, battery replacement and privacy protection) of the BLE gateway scheme are relatively difficult, and meanwhile, the mobility of the traditional threshold method is poor and the calibration cost is high due to different varieties/matrixes/basin geometries. Therefore, designing a solution for intelligent maintenance is a technical problem to be solved by those skilled in the art. Disclosure of Invention Aiming at the defects, the embodiment of the invention discloses a plant intelligent maintenance method based on near field communication interaction, which can realize the accuracy and low power consumption of plant maintenance and reduce maintenance cost. The first aspect of the embodiment of the invention discloses a plant intelligent maintenance method based on near field communication interaction, which comprises the following steps: Writing, by an intelligent terminal, a set of initial configuration parameters to a microcontroller unit of a plant care device in response to establishing a near field communication connection with a near field communication tag embedded in the plant care device, wherein the plant care device is disposed in a pot, the initial configuration parameters including a plant variety identifier, a pot volume, a matrix type identifier, and a security policy parameter; Periodically collecting, by the microcontroller unit of the plant care device, environmental data of a respective plant from an environmental sensor, wherein collecting the environmental data includes executing within a wake-up period of the microcontroller unit; Generating, by the microcontroller unit, a corresponding plant care decision based on the collected environmental data and the stored initial configuration parameters, wherein generating the corresponding plant care decision comprises: constructing a maintenance feature vector based on the environmental data; Inputting the maintenance feature vector to a pre-trained machine learning model stored in the microcontroller unit to obtain an initial recommended amount of water and an initial urgency; Performing constraint optimization on the initial suggested water quantity and the initial emergency degree based on the safety strategy parameters so as to determine a final water quantity and a final emergency degree; Reading maintenance state information from the microcontroller unit by the intelligent terminal in response to the near field communication connection established with the near field communication tag, and displaying the maintenance state information on the intelligent terminal of the user; And the intelligent terminal responds to the user input and sends an execution instruction to the micro-controller unit so as to control the corresponding executor to carry out watering operation according to the maintenance state information and the user input. As an optional implementation manner, in the first aspect of the embodiment of the present invention, the establishing, by the intelligent terminal, a near field communication connection in response to the near field communication tag embedded in the plant care device includes: Triggering the microcontroller unit of the plant maintenance device to enter a communication mode through a near field communication tag so as to wake up the microcontroller unit; establishing a corresponding temporary communication session, wherein establishing the corresponding temporary communication session comprises: the intelligent terminal sends a session request frame to the microcontroller unit, wherein the session request frame comprises a terminal random number, a time stamp and intelligent terminal identification information; Generating a device random number and an incremental session counter by the microcontroller unit in response to the session request frame, and transmitting the device random number, the session counter and a device state abstract to the intelligent terminal; A unique session identifier for identifying a temporary communication session is commonly constructed by at least one of the intelligent terminal and the microcontroller unit based on the terminal nonce, the device nonce, and the session counter, wherein all communication frames during the temporary communication session carry the unique session identifier to prevent cross-session replay attacks. In a first aspect of the embodiment of the present invention, the environment sensor includes a water content sensor, a temperat