A monitoring and management system and method for an autonomous harvester based on artificial intelligence

Abstract

The present invention relates to the technical field of harvester operation monitoring and management, and discloses an artificial intelligence-based automatic driving harvester monitoring and management system and method; the system comprises a harvesting operation area geographic data processing module, a harvesting operation area dangerous object monitoring execution module, and an automatic driving harvester harvesting operation path management module; a metal detector is used to dynamically collect object monitoring signal parameters in the harvesting operation area, and a signal value comparison is performed with the dangerous object signal parameters in the harvesting operation area based on an artificial intelligence intelligent recognition algorithm to determine whether there are dangerous objects in the harvesting operation area, thereby achieving accurate online identification of dangerous objects in the harvesting operation area and improving the safety of the harvesting operation of the automatic driving harvester; a position sensor is used to accurately obtain the spatial position of dangerous objects in the harvesting operation area and efficiently clean up the dangerous objects, thereby reducing the safety risk of the harvesting operation of the automatic driving harvester.

Description

Technical Field

[0001] The present invention relates to the technical field of harvester operation monitoring and management, and specifically to an automatic driving harvester monitoring and management system and method based on artificial intelligence. Background Art

[0002] A harvester is a machine that harvests crops in an integrated manner. It completes harvesting and threshing at one time, and collects the grains into storage bins, and then transports the grains to transport vehicles through conveyor belts. It can also be harvested manually, and the straw of crops such as rice and wheat is laid in the field, and then picked up and threshed by grain harvesting machinery. Crop harvesting machinery that harvests grains and straw of cereal crops such as rice and wheat. It includes harvesters, reapers, balers, grain combine harvesters and grain threshers. Grain harvesting machinery is developed on the basis of various harvesting and threshing tools. With the development of smart agriculture, the self-driving harvester is a new type of intelligent agricultural machinery, and its appearance represents an important progress in agricultural mechanization and intelligence. Compared with traditional manual harvesting and manually driven harvesters, it can complete harvesting operations more accurately and efficiently, greatly improving the quality and efficiency of agricultural production. The working principle of the self-driving harvester is mainly divided into two parts: perception and control. The perception part uses a variety of sensors such as lidar, cameras, and GPS to perceive the surrounding environment. Among them, the laser radar can obtain the three-dimensional information of field crops, and can realize accurate distance measurement and terrain detection; the camera can perform image recognition and positioning, and convert the crop information into digital signals to input the control system; GPS is used to locate and measure the speed and direction of the vehicle and other related parameters. The control part is composed of a computer and an intelligent control system. Through the information input of the above sensors, the vehicle's driving, turning, stopping, harvesting and other operations are autonomously controlled. The control system of the automatic driving harvester not only has the ability of intelligent judgment and decision-making, but also can autonomously plan the path, adjust the speed, and realize obstacle avoidance functions to ensure the smooth completion of the entire operation process. However, the existing automatic driving harvester operation process cannot detect dangerous objects in the harvesting operation area, thereby causing damage to the automatic driving harvester, and cannot flexibly adjust the harvesting operation path spacing parameters according to the specifications and models of the automatic driving harvester, which reduces the safety of the automatic driving harvester and the efficiency of agricultural harvesting operations.

[0003] The Chinese invention patent application with publication number CN107111843A discloses a grain management system having the combine harvester and a management server. The combine harvester is provided with: a grain measuring unit, which outputs a measured value related to the composition of the harvested grains supplied to the grain box; a data sending unit, which sends the field identification information and the measured value of a specific field to the management server via a communication line. The management server is provided with: a receiving unit, which receives the field identification information and the measured value from the combine harvester; a table management unit, which determines a measured value grain component value table for deriving the grain component value from the measured value based on the field identification information; and a grain component value calculation unit, which uses the measured value grain component value table determined by the table management unit to obtain the grain component value from the measured value. The above technical solution does not detect dangerous objects in the harvesting operation area during the harvesting operation of the combine harvester, thereby causing damage to the harvesting operation of the combine harvester. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In order to solve the problem that the existing autonomous driving harvester operation process cannot detect dangerous objects in the harvesting operation area, thereby causing damage to the autonomous driving harvester, and cannot flexibly adjust the harvesting operation path spacing parameters according to the specifications of the autonomous driving harvester, thereby reducing the safety of the autonomous driving harvester and the efficiency of agricultural harvesting operations, the above purpose of accurately establishing a three-dimensional model of the harvesting operation area, intelligently planning the dangerous object detection operation path in the harvesting operation area, accurately identifying the characteristics and locations of dangerous objects in the harvesting operation area, intelligently adjusting the harvesting operation path spacing parameters, and autonomously performing harvesting operations is achieved.

[0006] (II) Technical solution

[0007] The present invention is implemented by the following technical solution: a monitoring and management method for an automatic driving harvester, the method comprising the following steps:

[0008] S1, collecting three-dimensional model data of the harvesting operation area;

[0009] S2, preprocessing the three-dimensional model data of the harvesting operation area to generate the three-dimensional model data of the standard harvesting operation area; performing plane grid division processing on the horizontal plane where the top of the crop is located in the three-dimensional model data of the standard harvesting operation area and establishing a spatial rectangular coordinate system to construct the three-dimensional model space coordinate data of the harvesting operation area;

[0010] S3, analyzing and planning the detection path data of dangerous objects in the harvesting operation area according to the spatial coordinate data of the three-dimensional model of the harvesting operation area and performing the dangerous object detection operation; collecting the object monitoring signal data in the harvesting operation area; using the data recognition algorithm to compare the signal values ​​of the object monitoring signal data in the harvesting operation area with the dangerous object signal data in the harvesting operation area, and generating the dangerous object detection result data in the harvesting operation area according to the signal value comparison result;

[0011] S4. When the detection result data of dangerous objects in the harvesting operation area indicates that dangerous objects exist, collecting spatial position coordinate data of dangerous objects in the harvesting operation area; and performing dangerous object clearing operations according to the spatial position coordinate data of dangerous objects in the harvesting operation area;

[0012] S5. When the detection result data of dangerous objects in the harvesting operation area indicates that no dangerous objects exist or the dangerous object clearing operation is completed, obtaining product feature data of the automatic driving harvester;

[0013] S6. Using a data search algorithm, the autonomous driving harvester product feature data is matched with the harvesting operation path spacing data corresponding to the standard product feature of the autonomous driving harvester to perform autonomous driving harvester product feature keyword matching, and the harvesting operation path spacing data corresponding to the autonomous driving harvester product feature data is searched out and the specific harvesting operation path spacing data of the autonomous driving harvester is identified and generated; based on the specific harvesting operation path spacing data of the autonomous driving harvester, the horizontal plane where the top of the crop is located in the three-dimensional model data of the standard harvesting operation area is plane meshed and a spatial rectangular coordinate system is established to construct the autonomous driving harvester harvesting operation space coordinate data;

[0014] S7. Analyze and plan the harvesting operation path data of the autonomous driving harvester based on the harvesting operation space coordinate data of the autonomous driving harvester and perform the crop harvesting operation.

[0015] Preferably, the steps of collecting the three-dimensional model data of the harvesting operation area are as follows:

[0016] S11. Collect the three-dimensional model of the farmland characteristics and crop characteristics of the harvesting operation area of ​​the autonomous driving harvester through a three-dimensional laser scanner and generate three-dimensional model data of the harvesting operation area .

[0017] The present invention uses a three-dimensional laser scanner to accurately and efficiently establish three-dimensional model parameters of a harvesting operation area, thereby achieving the effect of providing data support for dangerous object detection in the harvesting operation area and automatic crop harvesting operations.

[0018] Preferably, the three-dimensional model data of the harvesting operation area is preprocessed to generate the three-dimensional model data of the standard harvesting operation area; the horizontal plane where the top of the crop is located in the three-dimensional model data of the standard harvesting operation area is plane meshed and a spatial rectangular coordinate system is established to construct the spatial coordinate data of the three-dimensional model of the harvesting operation area as follows:

[0019] S21, using Gaussian filtering algorithm to analyze the three-dimensional model data of the harvesting operation area Perform data noise reduction preprocessing to generate 3D model data of standard harvesting operation area ;

[0020] S22: 3D model data of the standard harvesting operation area The horizontal plane where the top of the crop is located adopts a side length of The square is processed by plane meshing to obtain the standard harvesting operation area 3D model data Establish a spatial rectangular coordinate system with the bottom surface of the three-dimensional model of the harvesting operation area as the base surface, obtain the spatial coordinates of the center point of the square in the plane grid division in the spatial rectangular coordinate system, and construct the spatial coordinate data set of the three-dimensional model of the harvesting operation area , ;in Indicates The spatial coordinate data of the three-dimensional model of the harvesting operation area, Indicates the maximum value of the spatial coordinate data number of the three-dimensional model of the harvesting operation area. ,in Indicates The horizontal coordinate corresponding to the spatial coordinate data of the three-dimensional model of the harvesting operation area, Indicates The vertical coordinate corresponding to the spatial coordinate data of the three-dimensional model of the harvesting operation area, Indicates The vertical coordinate corresponding to the spatial coordinate data of the three-dimensional model of the harvesting operation area.

[0021] The present invention uses a Gaussian filtering algorithm to perform data noise reduction preprocessing on the three-dimensional model parameters of the harvesting operation area, performs plane grid division processing on the three-dimensional model parameters of the harvesting operation area, and establishes a spatial rectangular coordinate system, so as to accurately establish the coordinate parameters of the three-dimensional model of the harvesting operation area, thereby achieving the effect of improving the accuracy of data parameter acquisition of the three-dimensional model of the harvesting operation area and the reliability of the automatic driving harvester monitoring and management system.

[0022] Preferably, the steps of analyzing and planning the detection path data of dangerous objects in the harvesting operation area and performing the dangerous object detection operation are as follows based on the spatial coordinate data of the three-dimensional model of the harvesting operation area; collecting the object monitoring signal data of the harvesting operation area; using the data recognition algorithm to compare the signal values ​​of the object monitoring signal data of the harvesting operation area with the dangerous object signal data of the harvesting operation area, and generating the dangerous object detection result data of the harvesting operation area based on the signal value comparison result are as follows:

[0023] S31, the three-dimensional model space coordinate data set of the harvesting operation area Three-dimensional model spatial coordinate data of the harvesting operation area According to the spatial coordinate data number of the 3D model of the harvesting operation area Orderly combination to generate hazardous object detection path data in the harvesting operation area ;

[0024] S32, the drone is equipped with a metal detector to detect the path data of dangerous objects in the harvesting operation area Perform hazardous material detection operations;

[0025] S33, when performing the dangerous object detection operation, collecting the object monitoring signal data in the harvesting operation area online through the metal detector and generating the object monitoring signal data in the harvesting operation area ;

[0026] S34. Establishing a data set of dangerous signal for harvesting operation area , ;in Indicates Harvesting operation area hazardous material signal data, Indicates the maximum value of the number of hazardous material signal data in the harvesting operation area, wherein the hazardous material signal data in the harvesting operation area includes any one of iron metal hazardous material signal data, copper metal hazardous material signal data, aluminum metal hazardous material signal data, and aluminum alloy hazardous material signal data;

[0027] S35, using a data recognition algorithm to identify the object monitoring signal data in the harvesting operation area A data set of dangerous objects signals in the harvesting operation area Danger signal data in the harvesting operation area Perform signal value comparison. The specific steps of the data recognition algorithm to perform signal value comparison are as follows:

[0028] S351, parameter initialization: update signal recognition bat population size N, maximum number of iterations T, objective function , signal identification of individual bats Location , i=1,2,…,N; signal identification of individual bats Speed , signal identification of individual bats The frequency of the sound wave , signal identification of individual bats The loudness of the sound wave Identify individual bats using signals The frequency of sound waves ;

[0029] S352. Find the best signal to identify the bat position in the current population , that is, the signal recognition bat population in the harvesting operation area of ​​the dangerous signal data set Search the search space to find the monitoring signal data of objects in the harvesting operation area The most matching hazard signal data for the harvesting operation area , the calculation formula for updating the parameters of signal recognition bats is as follows: , , , ,in Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The speed in the search space, Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The speed in the search space, Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The position in the search space, Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The position in the search space, Represents a data set of dangerous signal recognition signals of individual bats in the harvesting operation area The optimal position in the search space, Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The loudness of the sound waves in the search space, Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The loudness of the sound waves in the search space, Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The frequency of sound waves in the search space, Signals to identify individual bats Data collection of hazard signals in harvesting areas The frequency of transmitted sound waves in the search space; represents a function taking values ​​(0, 1), and is a constant and 0< <1, >0;

[0030] S353, generate a random number rand1, rand1 is a random number with a value of [0, 1]. When rand1> , from the data set of hazard signals in the harvesting operation area Select the signal at the optimal position in the search space to identify individual bats and collect dangerous signal data in the harvesting operation area Search the search space to find the monitoring signal data of objects in the harvesting operation area The most matching hazard signal data for the harvesting operation area , the optimal position of the signal to identify individual bats is selected by the formula , , Generate a local and harvesting area object monitoring signal data The most matching hazard signal data for the harvesting operation area The solution of Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The frequency of sound waves in the search space, when rand1≤ When, according to the formula, , Update signal to identify bat location;

[0031] S354, generate another random number rand2, where rand2 is a random number with a value of [0, 1]; when rand2 < , and at this time, the dangerous signal data of the operation area is harvested Corresponding objective function Object monitoring signal data in the harvesting operation area If the fitness value of is better than the new solution in step S353, then accept the position and use the formula , Adjustment and ;in Represents the data set of dangerous object signals in the harvesting operation area after signal recognition bat individual adjustment The position in the search space, Represents the data set of dangerous object signals in the harvesting operation area before signal recognition bat individual adjustment The position in the search space, Indicates The signal recognition of bat individuals in the harvesting operation area in the iteration of the dangerous signal data set The loudness of the sound waves in the search space, is any number between -1 and 1. When, according to the formula, , Update signal to identify bat location;

[0032] S355, sorting the fitness values ​​of all individuals in the signal recognition bat population, that is, searching for all the dangerous signal data of the harvesting operation area Object monitoring signal data in the harvesting operation area Sort the fitness values ​​and find the current best , that is, find the monitoring signal data of objects in the harvesting operation area The most matching hazard signal data for the harvesting operation area ;

[0033] S356: When the maximum number of iterations T is met, output the object monitoring signal data of the harvesting operation area Harvesting area hazard signal data The matching results;

[0034] S36, harvesting the object monitoring signal data in the operation area according to step S356 Harvesting area hazard signal data Perform signal value matching to generate the detection result data of dangerous objects in the harvesting operation area ;

[0035] when and If the signal value matching is successful, it means that there are dangerous objects in the harvesting operation area, and the detection result data of dangerous objects in the harvesting operation area is output. For the presence of dangerous objects;

[0036] when and If the signal value matching is unsuccessful, it means that there are no dangerous objects in the harvesting operation area, and the detection result data of dangerous objects in the harvesting operation area is output. There are no dangerous objects.

[0037] The present invention plans the path parameters for detecting dangerous objects in the harvesting operation area, autonomously performs the dangerous object detection operation in the harvesting operation area, and compares the signal values ​​of the object monitoring signal parameters in the harvesting operation area based on the artificial intelligence bat optimization algorithm with the dangerous object signal parameters in the harvesting operation area to determine whether there are dangerous objects in the harvesting operation area. This achieves intelligent planning of the path for detecting dangerous objects in the harvesting operation area and accurate online identification of dangerous objects in the harvesting operation area, thereby improving the safety of the harvesting operation of the autonomous driving harvester.

[0038] Preferably, when the detection result data of dangerous objects in the harvesting operation area indicates that dangerous objects exist, the spatial position coordinate data of the dangerous objects in the harvesting operation area is collected; and the operation steps of performing the dangerous object clearing operation according to the spatial position coordinate data of the dangerous objects in the harvesting operation area are as follows:

[0039] S41, when the detection result data of dangerous objects in the harvesting operation area When there are dangerous objects, the drone carries a position sensor to collect the spatial coordinates of the dangerous objects in the harvesting operation area in real time and generate the spatial coordinate data of the dangerous objects in the harvesting operation area. ,in Indicates the horizontal coordinate corresponding to the spatial position coordinate data of the dangerous objects in the harvesting operation area, Indicates the vertical coordinate corresponding to the spatial position coordinate data of the dangerous objects in the harvesting operation area, Indicates the vertical coordinate corresponding to the spatial position coordinate data of the dangerous objects in the harvesting operation area;

[0040] S42, the drone carries a gripper based on the spatial position coordinate data of the dangerous objects in the harvesting operation area Arrive at the harvesting operation area where hazardous materials are located and perform hazardous material cleanup operations.

[0041] The present invention uses a position sensor to accurately obtain the spatial position of dangerous objects in the harvesting operation area, and combines the drone-mounted gripper to efficiently clear the dangerous objects, thereby achieving the effect of reducing the safety risks of the harvesting operation of the autonomous harvester.

[0042] Preferably, when the hazardous object detection result data in the harvesting operation area indicates that there is no hazardous object or the hazardous object clearing operation is completed, the operation steps for obtaining the product feature data of the autonomous driving harvester are as follows:

[0043] S51, when the detection result data of dangerous objects in the harvesting operation area When there are no dangerous objects or the dangerous object cleaning operation is completed, the harvester monitoring platform obtains the product characteristic parameters of the autonomous harvester entering the harvesting operation area and generates the product characteristic data of the autonomous harvester The product characteristic data of the autonomous driving harvester includes the model specifications, name, manufacturer name and equipment number of the autonomous driving harvester.

[0044] The present invention collects the product characteristic parameters of the self-driving harvester online through the harvester monitoring platform, so as to provide data support for scientifically adjusting the distance between the harvesting operation paths of the self-driving harvester.

[0045] Preferably, the data search algorithm is used to match the autonomous driving harvester product feature data with the harvesting operation path spacing data corresponding to the standard product feature of the autonomous driving harvester to perform autonomous driving harvester product feature keyword matching, search out the harvesting operation path spacing data corresponding to the autonomous driving harvester product feature data and identify and generate the specific harvesting operation path spacing data of the autonomous driving harvester; based on the specific harvesting operation path spacing data of the autonomous driving harvester, the horizontal plane where the top of the crop is located in the three-dimensional model data of the standard harvesting operation area is plane meshed and a spatial rectangular coordinate system is established, and the operation steps of constructing the autonomous driving harvester harvesting operation space coordinate data are as follows:

[0046] S61. Establish a data set of the distance between harvesting operation paths corresponding to the standard product characteristics of the autonomous driving harvester , ;in Indicates The standard product characteristic data of the self-driving harvester corresponding to the self-driving harvester. Indicates the maximum number of autonomous harvester types, Indicates the standard product characteristic data of the self-driving harvester Corresponding harvesting operation path spacing data; express and The corresponding relationship of correspond , correspond , correspond ; The harvesting operation path spacing data represents the distance between adjacent paths during the harvesting operation performed by the autonomous driving harvester. Different models and specifications of autonomous driving harvesters need to set different adjacent path spacings;

[0047] S62, using the KD tree nearest neighbor search algorithm to obtain the product feature data of the autonomous driving harvester Harvesting operation path spacing data set corresponding to the standard product features of the autonomous harvester Standard product characteristic data of self-driving harvester in Perform keyword matching of self-driving harvester product features to search for the self-driving harvester product feature data The corresponding harvesting operation path spacing data And identify and generate specific harvesting operation path spacing data for the autonomous driving harvester ;

[0048] S63: 3D model data of the standard harvesting operation area The horizontal plane where the top of the crops is located is plane-based on the specific harvesting operation path spacing data of the autonomous harvester The plane grid division process is performed on a square with a side length of , and the spatial coordinates of the center point of the square in the spatial rectangular coordinate system in the plane grid division are obtained by using the spatial rectangular coordinate system established in step S22, and the spatial coordinates of the center point of the square in the spatial rectangular coordinate system are converted into the three-dimensional model data of the harvesting operation area according to the standard The model is arranged in an orderly manner from the edge to the center point to construct the spatial coordinate data set of the harvesting operation of the autonomous driving harvester , ;in Indicates The spatial coordinate data of the harvesting operation of the autonomous harvester, Indicates the maximum value of the spatial coordinate data number of the harvesting operation of the autonomous harvester. ,in Indicates The horizontal coordinate corresponding to the spatial coordinate data of the harvesting operation of the autonomous driving harvester, Indicates The vertical coordinate corresponding to the spatial coordinate data of the harvesting operation of the autonomous harvester, Indicates The vertical coordinate corresponding to the spatial coordinate data of the harvesting operation of the autonomous driving harvester.

[0049] The present invention uses the KD tree nearest neighbor search algorithm to perform keyword matching on the product characteristic parameters of the autonomous driving harvester and the harvesting operation path spacing parameters corresponding to the standard product characteristics of the autonomous driving harvester to obtain the path spacing parameters of the harvesting operation of the autonomous driving harvester, and construct the harvesting operation space coordinate parameters, so as to achieve the effect of flexibly and intelligently adjusting the path spacing parameters of the harvesting operation of the autonomous driving harvester according to the product characteristic parameters of the autonomous driving harvester and scientifically constructing the harvesting operation space coordinate parameters.

[0050] Preferably, the steps of analyzing and planning the harvesting operation path data of the autonomous driving harvester based on the harvesting operation space coordinate data of the autonomous driving harvester and performing the crop harvesting operation are as follows:

[0051] S71: Collecting the harvesting operation space coordinate data of the autonomous driving harvester The spatial coordinate data of the harvesting operation of the autonomous harvester Generate the harvesting operation path data of the autonomous driving harvester according to the ordered combination of the spatial coordinate data numbers of the autonomous driving harvester ;

[0052] S72: The autonomous driving harvester harvests the harvesting operation path data according to the autonomous driving harvester. Perform crop harvesting operations.

[0053] The present invention intelligently plans the harvesting operation path parameters of the automatic driving harvester according to the harvesting operation space coordinate parameters of the automatic driving harvester and autonomously performs the crop harvesting operation, thereby achieving the effect of accurate planning of the harvesting operation path of the automatic driving harvester and efficient and autonomous harvesting operation.

[0054] An artificial intelligence-based self-driving harvester monitoring and management system for implementing the self-driving harvester monitoring and management method, the system comprising a harvesting operation area geographic data processing module, a harvesting operation area dangerous object monitoring execution module, and a self-driving harvester harvesting operation path management module;

[0055] The harvesting operation area geographic data processing module includes a harvesting operation area three-dimensional model acquisition unit, a harvesting operation area three-dimensional model preprocessing unit, and a harvesting operation area three-dimensional model space coordinate generation unit;

[0056] The harvesting operation area 3D model acquisition unit acquires the harvesting operation area 3D model data by a 3D laser scanner; the harvesting operation area 3D model preprocessing unit performs data preprocessing on the harvesting operation area 3D model data to generate standard harvesting operation area 3D model data; the harvesting operation area 3D model space coordinate generation unit performs plane grid division processing on the horizontal plane where the top of the crop is located in the standard harvesting operation area 3D model data and establishes a space rectangular coordinate system to construct the harvesting operation area 3D model space coordinate data;

[0057] The harvesting operation area dangerous object monitoring execution module includes a harvesting operation area dangerous object detection path planning unit, a harvesting operation area dangerous object detection operation execution unit, a harvesting operation area detection signal acquisition unit, a harvesting operation area dangerous object signal storage unit, a harvesting operation area dangerous object detection unit, a harvesting operation area dangerous object position coordinate acquisition unit, and a harvesting operation area dangerous object cleaning execution unit;

[0058] The harvesting operation area hazardous object detection path planning unit plans the harvesting operation area hazardous object detection path data according to the spatial coordinate data of the three-dimensional model of the harvesting operation area; the harvesting operation area hazardous object detection operation execution unit executes the hazardous object detection operation according to the harvesting operation area hazardous object detection path data; the harvesting operation area detection signal acquisition unit acquires the harvesting operation area object monitoring signal data through a metal detector; the harvesting operation area hazardous object signal storage unit is used to store the harvesting operation area hazardous object signal data; the harvesting operation area hazardous object detection unit performs a signal value comparison between the harvesting operation area object monitoring signal data and the harvesting operation area hazardous object signal data based on an intelligent recognition algorithm of artificial intelligence, and generates the harvesting operation area hazardous object detection result data according to the signal value comparison result; the harvesting operation area hazardous object position coordinate acquisition unit acquires the harvesting operation area hazardous object spatial position coordinate data through a position sensor; the harvesting operation area hazardous object cleaning execution unit executes the hazardous object cleaning operation according to the harvesting operation area hazardous object spatial position coordinate data;

[0059] The autonomous driving harvester harvesting operation path management module includes a harvester product feature acquisition unit, a harvester product feature corresponding harvesting operation path spacing data storage unit, a harvester specific harvesting operation path spacing data analysis unit, a harvester harvesting operation path planning unit, and a harvester harvesting operation execution unit;

[0060] The harvester product feature acquisition unit acquires the autonomous driving harvester product feature data through the harvester monitoring platform; the harvester product feature corresponding to the harvesting operation path spacing data storage unit is used to store the autonomous driving harvester standard product feature corresponding to the harvesting operation path spacing data; the harvester specific harvesting operation path spacing data analysis unit uses a data search algorithm to match the autonomous driving harvester product feature data with the autonomous driving harvester standard product feature corresponding to the harvesting operation path spacing data, and searches out the harvesting operation path spacing data corresponding to the autonomous driving harvester product feature data. And identify and generate specific harvesting operation path spacing data of the autonomous driving harvester; the harvesting operation path planning unit of the harvester performs plane grid division processing on the horizontal plane where the top of the crop is located in the three-dimensional model data of the standard harvesting operation area according to the specific harvesting operation path spacing data of the autonomous driving harvester and establishes a spatial rectangular coordinate system to construct the autonomous driving harvester harvesting operation space coordinate data; the autonomous driving harvester harvesting operation path data is analyzed and planned according to the autonomous driving harvester harvesting operation space coordinate data; the harvester harvesting operation execution unit executes the crop harvesting operation according to the autonomous driving harvester harvesting operation path data.

[0061] (III) Beneficial effects

[0062] The present invention provides an artificial intelligence-based self-driving harvester monitoring management system and method, which has the following beneficial effects:

[0063] 1. Through the harvesting operation area 3D model acquisition unit, a 3D laser scanner is used to accurately and efficiently establish the 3D model parameters of the harvesting operation area, providing data support for the detection of dangerous objects in the harvesting operation area and the automatic harvesting of crops; the harvesting operation area 3D model preprocessing unit and the harvesting operation area 3D model space coordinate generation unit cooperate with each other to perform data noise reduction preprocessing on the 3D model parameters of the harvesting operation area, thereby improving the acquisition accuracy of the 3D model data parameters of the harvesting operation area; the 3D model parameters of the harvesting operation area are processed by plane grid division and a spatial rectangular coordinate system is established to accurately establish the coordinate parameters of the 3D model of the harvesting operation area, providing data support for the detection of dangerous objects in the harvesting operation area, and improving the reliability of the monitoring and management system of the autonomous driving harvester.

[0064] 2. Through the cooperation between the harvesting operation area hazardous object detection path planning unit and the harvesting operation area hazardous object detection operation execution unit, the path parameters of the harvesting operation area hazardous object detection are intelligently planned and the harvesting operation area hazardous object detection operation is autonomously executed; the harvesting operation area detection signal acquisition unit and the harvesting operation area hazardous object detection unit cooperate with each other, dynamically collect the harvesting operation area object monitoring signal parameters through the metal detector, and compare the signal values ​​with the harvesting operation area hazardous object signal parameters based on artificial intelligence intelligent recognition algorithm to determine whether there are hazardous objects in the harvesting operation area, so as to achieve accurate online identification of hazardous objects in the harvesting operation area and improve the safety of the harvesting operation of the autonomous driving harvester; the harvesting operation area hazardous object position coordinate acquisition unit and the harvesting operation area hazardous object cleaning execution unit cooperate with each other, use position sensors to accurately obtain the spatial position of hazardous objects in the harvesting operation area and efficiently clean up hazardous objects, reducing the safety risk of the harvesting operation of the autonomous driving harvester.

[0065] 3. Through the harvester product feature acquisition unit, the autonomous driving harvester product feature parameters are collected online through the harvester monitoring platform to provide data support for scientifically adjusting the harvesting operation path spacing of the autonomous driving harvester; the harvester product feature corresponding to the harvesting operation path spacing data storage unit and the harvester specific harvesting operation path spacing data analysis unit cooperate with each other, and use the intelligent search algorithm to match the autonomous driving harvester product feature parameters with the autonomous driving harvester standard product feature corresponding to the harvesting operation path spacing parameters to output the harvesting operation path spacing parameters, so as to realize flexible and intelligent adjustment of the autonomous driving harvester according to the autonomous driving harvester product feature parameters. The path spacing parameters of the autonomous driving harvester's harvesting operation are calculated, thereby improving the efficiency and quality of the autonomous driving harvester's harvesting operation; the harvester's harvesting operation path planning unit and the harvester's harvesting operation execution unit cooperate with each other, and scientifically construct the autonomous driving harvester's harvesting operation space coordinate parameters according to the specific harvesting operation path spacing parameters of the autonomous driving harvester, and intelligently plan the autonomous driving harvester's harvesting operation path parameters according to the autonomous driving harvester's harvesting operation space coordinate parameters and autonomously execute the crop harvesting operation, thereby realizing accurate planning of the autonomous driving harvester's harvesting operation path and efficient autonomous harvesting operation, and improving the applicability of the autonomous driving harvester monitoring and management system. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 A schematic diagram of a module of an artificial intelligence-based self-driving harvester monitoring and management system provided by the present invention;

[0067] Figure 2 A flow chart of a method for monitoring and managing an autonomous driving harvester provided by the present invention. DETAILED DESCRIPTION

[0068] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0069] The embodiment of the monitoring management system and method of the automatic driving harvester based on artificial intelligence is as follows:

[0070] See also Figure 1 - Figure 2 , a monitoring and management method for an autonomous driving harvester, the method comprising the following steps:

[0071] S1, collecting three-dimensional model data of the harvesting operation area;

[0072] S2, preprocessing the 3D model data of the harvesting operation area to generate the 3D model data of the standard harvesting operation area; performing plane grid division processing on the horizontal plane where the top of the crop is located in the 3D model data of the standard harvesting operation area and establishing a spatial rectangular coordinate system to construct the spatial coordinate data of the 3D model of the harvesting operation area;

[0073] S3, plan the detection path data of dangerous objects in the harvesting operation area according to the spatial coordinate data of the three-dimensional model of the harvesting operation area and perform the dangerous object detection operation; collect the object monitoring signal data of the harvesting operation area; use the data recognition algorithm to compare the signal values ​​of the object monitoring signal data of the harvesting operation area with the dangerous object signal data of the harvesting operation area, and generate the dangerous object detection result data of the harvesting operation area according to the signal value comparison result;

[0074] S4. When the detection result data of dangerous objects in the harvesting operation area indicates that dangerous objects exist, the spatial position coordinate data of the dangerous objects in the harvesting operation area is collected; and the dangerous object clearing operation is performed according to the spatial position coordinate data of the dangerous objects in the harvesting operation area;

[0075] S5. When the detection result data of dangerous objects in the harvesting operation area indicates that no dangerous objects exist or the dangerous object clearing operation is completed, obtain the product characteristic data of the automatic driving harvester;

[0076] S6. Use a data search algorithm to match the autonomous driving harvester product feature data with the harvesting operation path spacing data corresponding to the standard product features of the autonomous driving harvester to perform autonomous driving harvester product feature keyword matching, search for the harvesting operation path spacing data corresponding to the autonomous driving harvester product feature data, and identify and generate the specific harvesting operation path spacing data of the autonomous driving harvester; perform plane grid division processing on the horizontal plane where the top of the crop is located in the three-dimensional model data of the standard harvesting operation area according to the specific harvesting operation path spacing data of the autonomous driving harvester, establish a spatial rectangular coordinate system, and construct the autonomous driving harvester harvesting operation space coordinate data;

[0077] S7. Analyze and plan the harvesting operation path data of the autonomous driving harvester based on the harvesting operation space coordinate data and perform the crop harvesting operation.

[0078] For further information, see Figure 1 - Figure 2 The steps for collecting 3D model data of the harvesting operation area are as follows:

[0079] S11. Collect the three-dimensional model of the farmland characteristics and crop characteristics of the harvesting operation area of ​​the autonomous driving harvester through a three-dimensional laser scanner and generate three-dimensional model data of the harvesting operation area .

[0080] The three-dimensional model data of the harvesting operation area is preprocessed to generate the three-dimensional model data of the standard harvesting operation area; the horizontal plane where the top of the crop is located in the three-dimensional model data of the standard harvesting operation area is plane meshed and a spatial rectangular coordinate system is established to construct the spatial coordinate data of the three-dimensional model of the harvesting operation area as follows:

[0081] S21. Use Gaussian filtering algorithm to analyze the 3D model data of the harvesting operation area. Perform data noise reduction preprocessing to generate 3D model data of standard harvesting operation area ;

[0082] S22. 3D model data of standard harvesting operation area The horizontal plane where the top of the crop is located adopts a side length of The plane mesh is processed by the square to obtain the standard harvesting operation area 3D model data Establish a spatial rectangular coordinate system with the bottom surface of the three-dimensional model of the harvesting operation area as the base surface, obtain the spatial coordinates of the center point of the square in the plane grid division in the spatial rectangular coordinate system, and construct the spatial coordinate data set of the three-dimensional model of the harvesting operation area , ;in Indicates The spatial coordinate data of the three-dimensional model of the harvesting operation area, Indicates the maximum value of the spatial coordinate data number of the three-dimensional model of the harvesting operation area. ,in Indicates The horizontal coordinate corresponding to the spatial coordinate data of the three-dimensional model of the harvesting operation area, Indicates The vertical coordinate corresponding to the spatial coordinate data of the three-dimensional model of the harvesting operation area, Indicates The vertical coordinate corresponding to the spatial coordinate data of the three-dimensional model of the harvesting operation area.

[0083] Through the harvesting operation area three-dimensional model acquisition unit, a three-dimensional laser scanner is used to accurately and efficiently establish the three-dimensional model parameters of the harvesting operation area, providing data support for the detection of dangerous objects in the harvesting operation area and the automatic harvesting of crops; the harvesting operation area three-dimensional model preprocessing unit and the harvesting operation area three-dimensional model space coordinate generation unit cooperate with each other to perform data noise reduction preprocessing on the harvesting operation area three-dimensional model parameters, thereby improving the acquisition accuracy of the three-dimensional model data parameters of the harvesting operation area; the three-dimensional model parameters of the harvesting operation area are processed by plane grid division and a spatial rectangular coordinate system is established to accurately establish the coordinate parameters of the three-dimensional model of the harvesting operation area, providing data support for the detection of dangerous objects in the harvesting operation area, and improving the reliability of the automatic driving harvester monitoring and management system.

[0084] For further information, see Figure 1 - Figure 2 , according to the spatial coordinate data of the three-dimensional model of the harvesting operation area, the hazardous object detection path data of the harvesting operation area is planned and the hazardous object detection operation is performed; the object monitoring signal data of the harvesting operation area is collected; the data recognition algorithm is used to compare the signal values ​​of the object monitoring signal data of the harvesting operation area with the hazardous object signal data of the harvesting operation area, and the operation steps of generating the hazardous object detection result data of the harvesting operation area according to the signal value comparison result are as follows:

[0085] S31, collecting the spatial coordinate data of the three-dimensional model of the harvesting operation area Three-dimensional model spatial coordinate data of the harvesting operation area According to the spatial coordinate data number of the 3D model of the harvesting operation area Orderly combination to generate hazardous object detection path data in the harvesting operation area ;

[0086] S32, drones equipped with metal detectors detect dangerous objects in the harvesting operation area based on the path data Perform hazardous material detection operations;

[0087] S33, when performing the dangerous object detection operation, collecting the object monitoring signal data in the harvesting operation area online through the metal detector and generating the object monitoring signal data in the harvesting operation area ;

[0088] S34. Establishing a data set of dangerous signal for harvesting operation area , ;in Indicates Harvesting operation area hazardous material signal data, Indicates the maximum value of the number of hazardous material signal data in the harvesting operation area, where the hazardous material signal data in the harvesting operation area includes any one of iron metal hazardous material signal data, copper metal hazardous material signal data, aluminum metal hazardous material signal data, and aluminum alloy hazardous material signal data;

[0089] S35, using data recognition algorithm to monitor signal data of objects in the harvesting operation area Harvesting area hazard signal data collection Danger signal data in the harvesting operation area Perform signal value comparison. The specific steps of the data recognition algorithm to perform signal value comparison are as follows:

[0090] S351, parameter initialization: update signal recognition bat population size N, maximum number of iterations T, objective function , signal identification of individual bats Location , i=1,2,…,N; signal identification of individual bats Speed , signal identification of individual bats The frequency of the sound wave , signal identification of individual bats The loudness of the sound wave Identify individual bats using signals The frequency of sound waves ;

[0091] S352. Find the best signal to identify the bat position in the current population , that is, the signal recognition bat population in the harvesting operation area of ​​the dangerous signal data set Search the search space to find the monitoring signal data of objects in the harvesting operation area The most matching hazard signal data for the harvesting operation area , the calculation formula for updating the parameters of signal recognition bats is as follows: , , , ,in Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The speed in the search space, Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The speed in the search space, Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The position in the search space, Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The position in the search space, Represents a data set of dangerous signal recognition signals of individual bats in the harvesting operation area The optimal position in the search space, Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The loudness of the sound waves in the search space, Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The loudness of the sound waves in the search space, Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The frequency of sound waves in the search space, Signals to identify individual bats Data collection of hazard signals in harvesting areas The frequency of transmitted sound waves in the search space; represents a function taking values ​​(0, 1), and is a constant and 0< <1, >0;

[0092] S353, generate a random number rand1, rand1 is a random number with a value of [0, 1]. When rand1> , from the data set of hazard signals in the harvesting operation area Select the signal at the optimal position in the search space to identify individual bats and collect dangerous signal data in the harvesting operation area Search the search space to find the monitoring signal data of objects in the harvesting operation area The most matching hazard signal data for the harvesting operation area , the optimal position of the signal to identify the individual bat is selected by the formula , , Generate a local and harvesting area object monitoring signal data The most matching hazard signal data for the harvesting operation area The solution of Indicates The signal identifies the individual bat in the iteration Data collection of hazard signals in harvesting areas The frequency of sound waves in the search space, when rand1≤ When, according to the formula, , Update signal to identify bat location;

[0093] S354, generate another random number rand2, where rand2 is a random number with a value of [0, 1]; when rand2 < , and at this time, the dangerous signal data of the operation area is harvested Corresponding objective function Object monitoring signal data in the harvesting operation area If the fitness value of is better than the new solution in step S353, then accept the position and use the formula , Adjustment and ;in Represents the data set of dangerous object signals in the harvesting operation area after signal recognition bat individual adjustment The position in the search space, Represents the data set of dangerous object signals in the harvesting operation area before signal recognition bat individual adjustment The position in the search space, Indicates The signal recognition of bat individuals in the harvesting operation area in the iteration of the dangerous signal data set The loudness of the sound waves in the search space, is any number between -1 and 1. When, according to the formula, , Update signal to identify bat location;

[0094] S355, sorting the fitness values ​​of all individuals in the signal recognition bat population, that is, searching for all the dangerous signal data of the harvesting operation area Object monitoring signal data in the harvesting operation area Sort the fitness values ​​and find the current best , that is, find the monitoring signal data of objects in the harvesting operation area The most matching hazard signal data for the harvesting operation area ;

[0095] S356: When the maximum number of iterations T is met, output the object monitoring signal data of the harvesting operation area Harvesting area hazard signal data The matching results;

[0096] S36, harvesting the object monitoring signal data in the operation area according to step S356 Harvesting area hazard signal data Perform signal value matching to generate the detection result data of dangerous objects in the harvesting operation area ;

[0097] when and If the signal value matching is successful, it means that there are dangerous objects in the harvesting operation area, and the detection result data of dangerous objects in the harvesting operation area is output. For the presence of dangerous objects;

[0098] when and If the signal value matching is unsuccessful, it means that there are no dangerous objects in the harvesting operation area, and the detection result data of dangerous objects in the harvesting operation area is output. There are no dangerous objects.

[0099] When the detection result data of dangerous objects in the harvesting operation area indicates that dangerous objects exist, the spatial position coordinate data of the dangerous objects in the harvesting operation area is collected; the operation steps for performing the dangerous object cleaning operation according to the spatial position coordinate data of the dangerous objects in the harvesting operation area are as follows:

[0100] S41, the detection result data of dangerous objects in the harvesting operation area When there are dangerous objects, the drone carries a position sensor to collect the spatial coordinates of the dangerous objects in the harvesting operation area in real time and generate the spatial coordinate data of the dangerous objects in the harvesting operation area. ,in Indicates the horizontal coordinate corresponding to the spatial position coordinate data of the dangerous objects in the harvesting operation area, Indicates the vertical coordinate corresponding to the spatial position coordinate data of the dangerous objects in the harvesting operation area, Indicates the vertical coordinate corresponding to the spatial position coordinate data of the hazardous objects in the harvesting operation area;

[0101] S42, the drone carries a gripper based on the spatial position coordinate data of dangerous objects in the harvesting operation area Arrive at the harvesting operation area where hazardous materials are located and perform hazardous material cleanup operations.

[0102] Through the cooperation between the hazardous object detection path planning unit in the harvesting operation area and the hazardous object detection operation execution unit in the harvesting operation area, the path parameters of the hazardous object detection in the harvesting operation area are intelligently planned and the hazardous object detection operation in the harvesting operation area is autonomously executed; the harvesting operation area detection signal acquisition unit and the harvesting operation area hazardous object detection unit cooperate with each other, dynamically collect the object monitoring signal parameters in the harvesting operation area through the metal detector, and compare the signal values ​​with the hazardous object signal parameters in the harvesting operation area based on the intelligent recognition algorithm of artificial intelligence to determine whether there are hazardous objects in the harvesting operation area, thereby realizing accurate online identification of hazardous objects in the harvesting operation area and improving the safety of the harvesting operation of the autonomous driving harvester; the harvesting operation area hazardous object position coordinate acquisition unit and the harvesting operation area hazardous object cleaning execution unit cooperate with each other, use position sensors to accurately obtain the spatial position of hazardous objects in the harvesting operation area and efficiently clean up hazardous objects, thereby reducing the safety risk of the harvesting operation of the autonomous driving harvester.

[0103] For further information, see Figure 1 - Figure 2When the hazardous object detection result data in the harvesting operation area indicates that there is no hazardous object or the hazardous object cleaning operation is completed, the steps for obtaining the product feature data of the autonomous driving harvester are as follows:

[0104] S51, the detection result data of dangerous objects in the harvesting operation area When there are no dangerous objects or the dangerous object cleaning operation is completed, the harvester monitoring platform obtains the product characteristic parameters of the autonomous harvester entering the harvesting operation area and generates the product characteristic data of the autonomous harvester ,The product characteristic data of the autonomous harvester includes the model specifications, name, manufacturer name and equipment number of the autonomous harvester.

[0105] The data search algorithm is used to match the product feature data of the autonomous driving harvester with the harvesting operation path spacing data corresponding to the standard product features of the autonomous driving harvester to perform the product feature keyword matching of the autonomous driving harvester, search out the harvesting operation path spacing data corresponding to the product feature data of the autonomous driving harvester and identify and generate the specific harvesting operation path spacing data of the autonomous driving harvester; according to the specific harvesting operation path spacing data of the autonomous driving harvester, the horizontal plane where the top of the crop is located in the three-dimensional model data of the standard harvesting operation area is plane meshed and a spatial rectangular coordinate system is established, and the operation steps for constructing the spatial coordinate data of the harvesting operation of the autonomous driving harvester are as follows:

[0106] S61. Establish a data set of the distance between harvesting operation paths corresponding to the standard product characteristics of the autonomous driving harvester , ;in Indicates The standard product characteristic data of the self-driving harvester corresponding to the self-driving harvester. Indicates the maximum number of autonomous harvester types, Indicates the standard product characteristic data of the self-driving harvester Corresponding harvesting operation path spacing data; express and The corresponding relationship of correspond , correspond , correspond ; The harvesting operation path spacing data indicates the distance between adjacent paths during the harvesting operation performed by the autonomous harvester. Different models and specifications of autonomous harvesters need to set different adjacent path spacings;

[0107] S62, using the KD tree nearest neighbor search algorithm to automatically drive the harvester product feature data Harvesting operation path spacing data set corresponding to the standard product characteristics of the autonomous harvester Standard product characteristic data of self-driving harvester in Perform keyword matching on the features of self-driving harvesters and search for product feature data of self-driving harvesters Corresponding harvesting operation path spacing data And identify and generate specific harvesting operation path spacing data for the autonomous driving harvester ;

[0108] S63. 3D model data of standard harvesting operation area The horizontal plane where the top of the crops is located is based on the specific harvesting operation path spacing data of the autonomous harvester. The plane grid division process is performed on the square with a side length of 1. The spatial rectangular coordinate system established in step S22 is used to obtain the spatial coordinates of the center point of the square in the plane grid division in the spatial rectangular coordinate system, and the spatial coordinates of the center point of the square in the spatial rectangular coordinate system are converted into the standard harvesting operation area 3D model data. The model is arranged in an orderly manner from the edge to the center point to construct the spatial coordinate data set of the harvesting operation of the autonomous driving harvester , ;in Indicates The spatial coordinate data of the harvesting operation of the autonomous harvester, Indicates the maximum value of the spatial coordinate data number of the harvesting operation of the autonomous harvester. ,in Indicates The horizontal coordinate corresponding to the spatial coordinate data of the harvesting operation of the autonomous driving harvester, Indicates The vertical coordinate corresponding to the spatial coordinate data of the harvesting operation of the autonomous harvester, Indicates The vertical coordinate corresponding to the spatial coordinate data of the harvesting operation of the autonomous driving harvester.

[0109] The steps for planning the harvesting operation path data of the autonomous driving harvester and performing the crop harvesting operation are as follows:

[0110] S71, collect the spatial coordinate data of the harvesting operation of the autonomous driving harvester The spatial coordinate data of the harvesting operation of the autonomous harvester Generate the harvesting operation path data of the autonomous driving harvester according to the ordered combination of the spatial coordinate data numbers of the autonomous driving harvester ;

[0111] S72, the autonomous driving harvester harvests the harvesting operation path data according to the autonomous driving harvester Perform crop harvesting operations.

[0112] Through the harvester product feature acquisition unit, the autonomous driving harvester product feature parameters are collected online through the harvester monitoring platform to provide data support for the scientific adjustment of the autonomous driving harvester harvesting operation path spacing; the harvester product feature corresponding to the harvesting operation path spacing data storage unit and the harvester specific harvesting operation path spacing data analysis unit cooperate with each other, and use the intelligent search algorithm to match the autonomous driving harvester product feature parameters with the autonomous driving harvester standard product feature corresponding to the harvesting operation path spacing parameters to output the harvesting operation path spacing parameters, so as to realize flexible and intelligent adjustment of autonomous driving according to the autonomous driving harvester product feature parameters. The path spacing parameters of the harvesting operation of the harvester improve the efficiency and quality of the harvesting operation of the autonomous driving harvester; the harvester harvesting operation path planning unit and the harvester harvesting operation execution unit cooperate with each other, and scientifically construct the autonomous driving harvester harvesting operation space coordinate parameters according to the specific harvesting operation path spacing parameters of the autonomous driving harvester, and intelligently plan the autonomous driving harvester harvesting operation path parameters according to the autonomous driving harvester harvesting operation space coordinate parameters and autonomously execute crop harvesting operations, thereby realizing accurate planning of the harvesting operation path of the autonomous driving harvester and efficient autonomous harvesting operations, and improving the applicability of the autonomous driving harvester monitoring and management system.

[0113] An automatic driving harvester monitoring and management system based on artificial intelligence is provided to implement an automatic driving harvester monitoring and management method, the system comprising a harvesting operation area geographic data processing module, a harvesting operation area dangerous object monitoring execution module, and an automatic driving harvester harvesting operation path management module;

[0114] The harvesting operation area geographic data processing module includes a harvesting operation area three-dimensional model acquisition unit, a harvesting operation area three-dimensional model preprocessing unit, and a harvesting operation area three-dimensional model space coordinate generation unit;

[0115] The harvesting operation area 3D model acquisition unit collects the 3D model data of the harvesting operation area through a 3D laser scanner; the harvesting operation area 3D model preprocessing unit performs data preprocessing on the 3D model data of the harvesting operation area to generate standard 3D model data of the harvesting operation area; the harvesting operation area 3D model space coordinate generation unit performs plane grid division processing on the horizontal plane where the top of the crop is located in the standard 3D model data of the harvesting operation area and establishes a spatial rectangular coordinate system to construct the 3D model space coordinate data of the harvesting operation area;

[0116] The harvesting operation area hazardous object monitoring execution module includes a harvesting operation area hazardous object detection path planning unit, a harvesting operation area hazardous object detection operation execution unit, a harvesting operation area detection signal acquisition unit, a harvesting operation area hazardous object signal storage unit, a harvesting operation area hazardous object detection unit, a harvesting operation area hazardous object position coordinate acquisition unit, and a harvesting operation area hazardous object cleaning execution unit;

[0117] A harvesting operation area hazardous object detection path planning unit, which plans the harvesting operation area hazardous object detection path data based on the spatial coordinate data of the three-dimensional model of the harvesting operation area; a harvesting operation area hazardous object detection operation execution unit, which executes the hazardous object detection operation based on the harvesting operation area hazardous object detection path data; a harvesting operation area detection signal acquisition unit, which collects the harvesting operation area object monitoring signal data through a metal detector; a harvesting operation area hazardous object signal storage unit, which is used to store the harvesting operation area hazardous object signal data; a harvesting operation area hazardous object detection unit, which compares the harvesting operation area object monitoring signal data with the harvesting operation area hazardous object signal data based on an artificial intelligence intelligent recognition algorithm, and generates the harvesting operation area hazardous object detection result data based on the signal value comparison result; a harvesting operation area hazardous object position coordinate acquisition unit, which collects the harvesting operation area hazardous object spatial position coordinate data through a position sensor; a harvesting operation area hazardous object cleaning execution unit, which executes the hazardous object cleaning operation based on the harvesting operation area hazardous object spatial position coordinate data;

[0118] The harvesting operation path management module of the autonomous driving harvester includes a harvester product feature acquisition unit, a harvester product feature corresponding harvesting operation path spacing data storage unit, a harvester specific harvesting operation path spacing data analysis unit, a harvester harvesting operation path planning unit, and a harvester harvesting operation execution unit;

[0119] A harvester product feature acquisition unit is used to acquire the product feature data of the autonomous driving harvester through the harvester monitoring platform; a harvester product feature corresponding to the harvesting operation path spacing data storage unit is used to store the harvesting operation path spacing data corresponding to the standard product features of the autonomous driving harvester; a harvester specific harvesting operation path spacing data analysis unit is used to match the autonomous driving harvester product feature data with the harvesting operation path spacing data corresponding to the standard product features of the autonomous driving harvester by using a data search algorithm, and search out the harvesting operation path spacing data corresponding to the autonomous driving harvester product feature data. And identify and generate specific harvesting operation path spacing data of the autonomous driving harvester; the harvesting operation path planning unit of the harvester performs plane grid division processing on the horizontal plane where the top of the crop is located in the three-dimensional model data of the standard harvesting operation area according to the specific harvesting operation path spacing data of the autonomous driving harvester and establishes a spatial rectangular coordinate system to construct the harvesting operation space coordinate data of the autonomous driving harvester; the harvesting operation path data of the autonomous driving harvester is planned based on the analysis of the harvesting operation space coordinate data of the autonomous driving harvester; the harvesting operation execution unit of the harvester executes the crop harvesting operation according to the harvesting operation path data of the autonomous driving harvester.

[0120] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A monitoring and management method for an autonomous driving harvester, characterized in that: The method comprises the following steps: S1, collecting the three-dimensional model data of the harvesting operation area, said S1 comprises the following steps: S11. Collect the three-dimensional model of the farmland characteristics and crop characteristics of the harvesting operation area of ​​the autonomous driving harvester through a three-dimensional laser scanner and generate three-dimensional model data of the harvesting operation area ; S2, respectively performing data preprocessing and plane grid division on the three-dimensional model data of the harvesting operation area to establish a coordinate system, generating standard three-dimensional model data of the harvesting operation area and three-dimensional model space coordinate data of the harvesting operation area, wherein S2 comprises the following steps: S21, using Gaussian filtering algorithm to analyze the three-dimensional model data of the harvesting operation area Perform data noise reduction preprocessing to generate 3D model data of standard harvesting operation area ; S22: 3D model data of the standard harvesting operation area The horizontal plane where the top of the crop is located adopts a side length of The square is processed by plane meshing to obtain the standard harvesting operation area 3D model data Establish a spatial rectangular coordinate system with the bottom surface of the three-dimensional model of the harvesting operation area as the base surface, obtain the spatial coordinates of the center point of the square in the plane grid division in the spatial rectangular coordinate system, and construct the spatial coordinate data set of the three-dimensional model of the harvesting operation area ; S3, planning the detection path data of dangerous objects in the harvesting operation area and performing the dangerous object detection operation; collecting the object monitoring signal data in the harvesting operation area and performing the signal value comparison with the dangerous object signal data in the harvesting operation area to generate the dangerous object detection result data in the harvesting operation area, said S3 includes the following steps: S31, the three-dimensional model space coordinate data set of the harvesting operation area The spatial coordinate data of the three-dimensional model of the harvesting operation area is sequentially combined according to the spatial coordinate data number of the three-dimensional model of the harvesting operation area to generate the detection path data of the dangerous objects in the harvesting operation area. ; S32, the drone is equipped with a metal detector to detect the path data of dangerous objects in the harvesting operation area Perform hazardous material detection operations; S33, when performing the dangerous object detection operation, collecting the object monitoring signal data in the harvesting operation area online through the metal detector and generating the object monitoring signal data in the harvesting operation area ; S34. Establishing a data set of dangerous signal for harvesting operation area ; S35: the object monitoring signal data of the harvesting operation area A data set of dangerous objects signals in the harvesting operation area The signal value comparison is performed on the dangerous object signal data in the harvesting operation area to identify the object monitoring signal data in the harvesting operation area Matching results with the dangerous goods signal data in the harvesting operation area; S36: Harvesting the object monitoring signal data in the operation area in step S35 The signal value matching result is performed with the dangerous object signal data of the harvesting operation area to generate the dangerous object detection result data of the harvesting operation area ; when If the signal value is matched successfully with the dangerous object signal data in the harvesting operation area, the detection result data of dangerous objects in the harvesting operation area will be output. For the presence of dangerous objects; when If the signal value matching with the dangerous object signal data of the harvesting operation area is unsuccessful, the detection result data of dangerous objects in the harvesting operation area will be output. There are no dangerous objects; S4, collecting spatial position coordinate data of dangerous objects in the harvesting operation area and performing dangerous object cleaning operations, said S4 includes the following steps: S41, when the detection result data of dangerous objects in the harvesting operation area When there are dangerous objects, the drone carries a position sensor to collect the spatial coordinates of the dangerous objects in the harvesting operation area in real time and generate the spatial coordinate data of the dangerous objects in the harvesting operation area. ; S42, the drone carries a gripper based on the spatial position coordinate data of the dangerous objects in the harvesting operation area Arrive at the harvesting operation area where hazardous materials are located and perform hazardous material cleanup operations; S5, obtaining product feature data of the autonomous driving harvester, wherein S5 comprises the following steps: S51, when the detection result data of dangerous objects in the harvesting operation area When there are no dangerous objects or the dangerous object cleaning operation is completed, the harvester monitoring platform obtains the product characteristic parameters of the autonomous harvester entering the harvesting operation area and generates the product characteristic data of the autonomous harvester , the autonomous driving harvester product characteristic data includes the model specification, name, manufacturer name and equipment number of the autonomous driving harvester; S6, matching the autonomous driving harvester product feature data with the harvesting operation path spacing data corresponding to the autonomous driving harvester standard product feature, generating the specific harvesting operation path spacing data of the autonomous driving harvester, performing plane grid division on the three-dimensional model data of the standard harvesting operation area to establish a coordinate system, and constructing the autonomous driving harvester harvesting operation space coordinate data, wherein S6 includes the following steps: S61. Establish a data set of the distance between harvesting operation paths corresponding to the standard product characteristics of the autonomous driving harvester ; The harvesting operation path spacing data set corresponding to the standard product characteristics of the autonomous driving harvester includes the standard product characteristic data of the autonomous driving harvester and the harvesting operation path spacing data; S62, using the KD tree nearest neighbor search algorithm to obtain the product feature data of the autonomous driving harvester Harvesting operation path spacing data set corresponding to the standard product features of the autonomous harvester The self-driving harvester standard product feature data in the automatic driving harvester is matched with the product feature keywords of the automatic driving harvester to search for the product feature data of the automatic driving harvester The corresponding harvesting operation path spacing data is marked to generate the specific harvesting operation path spacing data of the automatic driving harvester ; S63: 3D model data of the standard harvesting operation area The horizontal plane where the top of the crops is located is plane-based on the specific harvesting operation path spacing data of the autonomous harvester The plane grid division process is performed on a square with a side length of , and the spatial coordinates of the center point of the square in the spatial rectangular coordinate system in the plane grid division are obtained by using the spatial rectangular coordinate system established in step S22, and the spatial coordinates of the center point of the square in the spatial rectangular coordinate system are converted into the three-dimensional model data of the harvesting operation area according to the standard The model is arranged in an orderly manner from the edge to the center point to construct the spatial coordinate data set of the harvesting operation of the autonomous driving harvester ; S7, planning the harvesting operation path data of the automatic driving harvester and performing the crop harvesting operation, wherein S7 includes the following steps: S71: Collecting the harvesting operation space coordinate data of the autonomous driving harvester The spatial coordinate data of the harvesting operation of the autonomous driving harvester is combined in an orderly manner according to the spatial coordinate data numbers of the harvesting operation of the autonomous driving harvester to generate the harvesting operation path data of the autonomous driving harvester ; S72: The autonomous driving harvester harvests the harvesting operation path data according to the autonomous driving harvester. Perform crop harvesting operations.

2. An artificial intelligence-based automatic driving harvester monitoring and management system that implements the automatic driving harvester monitoring and management method as claimed in claim 1, characterized in that: The system includes a harvesting operation area geographic data processing module, a harvesting operation area dangerous object monitoring execution module, and an automatic driving harvester harvesting operation path management module.

Images