Unmanned agricultural machine operation data analysis method and system based on three-dimensional scene reconstruction

Abstract

The invention discloses an unmanned agricultural machinery operation data analysis method and system based on three-dimensional scene reconstruction, and relates to the field of unmanned agricultural machinery. Point cloud data acquisition and data processing are performed through field investigation and grid division by using laser scanning, and a farm three-dimensional scene model is constructed; mapping the track and the pose of the unmanned agricultural machine in a three-dimensional scene model, realizing time sequence association through timestamp marking, performing operation interval division according to historical multi-modal data, determining an operation type in combination with a weighting algorithm, performing operation result evaluation by taking an operation coverage rate and spraying uniformity as indexes, and determining an operation result of the unmanned agricultural machine. And finally, setting an adjustment period, constructing a knowledge graph, and adjusting the operation parameters according to the knowledge graph, thereby achieving the intelligent monitoring of the operation of the unmanned agricultural machine, effectively guaranteeing the operation quality of the unmanned agricultural machine, and improving the operation efficiency of the unmanned agricultural machine.

Description

Technical Field

[0001] The present invention relates to the field of unmanned agricultural machinery, and in particular to an unmanned agricultural machinery operation data analysis method and system based on three-dimensional scene reconstruction. Background Art

[0002] In the field of modern unmanned agricultural machinery operations, unmanned agricultural machinery is a type of machine that uses unmanned aerial vehicles as the main body and integrates navigation, intelligent monitoring, intelligent control and precision operation technologies. It can perform sowing, irrigation and fertilization operations in the air without manual operation. Precision and intelligent management and control are the core requirements for improving operation efficiency and ensuring crop growth quality. Traditional unmanned agricultural machinery farm operations mostly rely on manual surveys of farm boundaries and crop distribution, and manual planning of operation trajectories. This is not only inefficient but also prone to errors due to human factors. At the same time, the lack of accurate judgment of different operation types makes it easy to misjudge behavior and affect the evaluation results. Existing evaluation indicators lack quantitative standards and cannot scientifically measure operation results. When operations do not meet standards, parameter adjustments rely more on experience, resulting in unstable optimization effects and no structured accumulation of operation data. Parameters need to be adjusted repeatedly for each operation, and there is no reuse of historical experience, resulting in low efficiency. These problems make it difficult for unmanned agricultural machinery farms to meet large-scale, high-precision agricultural production needs, so a systematic solution is urgently needed. Summary of the Invention

[0003] The purpose of the present invention is to provide a method and system for analyzing unmanned agricultural machinery operation data based on three-dimensional scene reconstruction to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a method for analyzing unmanned agricultural machinery operation data based on three-dimensional scene reconstruction, the method comprising the following steps: Step S1: Scan the farm using terrestrial laser scanning technology to construct a three-dimensional scene model; Step S1-1, field survey of farm boundaries, crop distribution, and fixed obstacles; Step S1-2: Draw a farm site survey sketch based on the field survey data, marking crops, boundaries and fixed obstacles; Step S1-3: Split the scanning area according to the grid division method, divide the farm into multiple sub-grids according to the "a×a" grid, mark the center point of each grid and set a target at the center point. a is the side length of the sub-grid, which is determined according to the operating range of the unmanned agricultural machinery; Step S1-4: collect high-density point cloud data from each sub-area, use a statistical filtering algorithm to remove discrete noise points from the point cloud data, then use a voxel grid downsampling method to downsample the data, and finally use a kd tree to deduplicate the point cloud; Step S1-5: Based on the center point coordinate target, the pre-processed multiple sub-region point clouds are spliced ​​and coordinate unified using the target matching splicing method; Step S1-6: Import the spliced ​​overall point cloud model into the 3D modeling software to construct a 3D farm scene model.

[0005] Through field surveys, farm boundaries, crop distribution, and fixed obstacles are clearly defined. Grid division and target setting are combined to avoid missing key areas, providing a benchmark for building a three-dimensional scene model. Data processing provides reliable model support for subsequent unmanned agricultural machinery trajectory mapping.

[0006] Step S2: Mapping the posture and operation trajectory of the unmanned agricultural machine into a three-dimensional scene model; Step S2-1: Use the National Geodetic Coordinate System (CGCS2000) to record the coordinate points of the 3D scene model, use the BeiDou Positioning Geographic Coordinate System to record the coordinate points of the UAV, and construct the UAV track data based on the UAV coordinate points; Step S2-2: cleaning and standardizing the coordinate points and attitude angles in the track data of the unmanned agricultural machinery, and time-stamping the coordinate points and attitude angles of the track data; Step S2-3: Map the coordinate points of the unmanned agricultural machine's track data to the three-dimensional model, and correspond the attitude angle data of the unmanned agricultural machine to the coordinate points of the track data through timestamps.

[0007] The operation trajectory of unmanned agricultural machinery is visualized, facilitating real-time monitoring of its operation status. Timestamps are used to mark the operation trajectory, providing basic data with correct timing for subsequent steps. Step S3: Based on the multimodal operation data of the unmanned agricultural machine, the operation data intervals are divided to determine the operation behavior of the unmanned agricultural machine; Step S3-1: Collect multimodal data of the normal operation of historical unmanned agricultural machinery, perform data cleaning and standardization on the multimodal data, use a filtering algorithm to eliminate environmental interference, use linear interpolation to handle missing values, and use a boxplot method to identify outliers; Step S3-2: Use timestamps to unify the data into the same time series; Step S3-3: Divide the typical intervals according to the characteristics of different operation types of the unmanned agricultural machinery, specifically: Sowing: The flight speed range is V[v1, v2], the flight height range is H[h1, h2], and the spray width range is W[w1, w2]; Irrigation: The flight speed range is V[v2, v3], the flight height range is H[h2, h3], and the spray width range is W[w2, w3]; Fertilization: The flight speed range is V[v3, v4], the flight height range is H[h3, h4], and the spray width range is W[w3, w4]; Step S3-4: Compare the flight speed, flight altitude, and spray width at each moment with the typical intervals, and mark the "candidate operation interval" for each parameter; Step S3-5: Set the spray width weight to 0.4, the flight height weight to 0.3, and the flight speed weight to 0.3, and use the weighted algorithm to calculate the total weight score of each operation type, specifically: ; ; Where, is the weight score of a certain job type, T represents the job type, is the flight speed weight, is the flight altitude weight, is the spray width weight, is the judgment function of flight speed, is the flight altitude judgment function, is the spray width judgment function; Step S3-6: If the total score of a certain type of work is greater than 0.7, it is directly determined as that type of work. If all scores are less than 0.7, the process proceeds to "boundary and timing auxiliary judgment"; Step S3-7: When a certain operation type enters the "boundary and timing auxiliary judgment", the flight speed, flight altitude and spray width of the operation type are judged with reference to the interval division of step S3-3. When the flight speed, flight altitude and spray width of the operation type are exactly at the operation interval division threshold boundary v2 or v3, h2 or h3, w2 or w3 of S3-3, it is exactly in the middle of two operation types and the operation type cannot be determined. Then refer to the operation type of the previous five moments. If the previous five operation types are all the same operation type, it is determined that the operation type is consistent with the previous five operation types. If among the first five moments, the operation type at a certain moment in the middle is different from the other four moments, it is determined to be data fluctuation, and the operation type is still determined to be consistent with the four identical operation types.

[0008] Through historical multimodal data, different operation types can be distinguished and the results of operation type determination can be clarified, helping managers to understand the specific operation content and progress of unmanned agricultural machinery and providing a basis for subsequent operation results evaluation.

[0009] Step S4: Analyze the operation results of the unmanned agricultural machine and adjust the posture and operation trajectory of the unmanned agricultural machine according to the analysis results; Step S4-1: Set the operation coverage and spraying uniformity as the core evaluation indicators, specifically: Assignment coverage: ; Where C is the operation coverage rate, The actual operation coverage area of ​​unmanned agricultural machinery, is the target operating area, W is the spray width, L is the trajectory length, k is the trajectory overlap coefficient, a is the spray height coefficient, b is the basic spray width, and H is the flight height; Spraying uniformity: ; ; ; ; In the formula, U is the spraying uniformity, F is the spraying flow rate, V is the flight speed, and Q is the spraying volume per unit area. is the average spraying amount, is the standard deviation, a is the spray height coefficient, b is the basic spray width, i is the sum index variable, represents the actual spraying amount of the i-th point, and n is the total number of sampling points; Step S4-2: Obtain the actual flight speed V of the unmanned agricultural machine through the flight control system act , actual flight altitude H act And the actual spray width W act ; Step S4-3: Set the target operating area before the unmanned agricultural machine is completed, collect the actual coverage area after the unmanned agricultural machine is completed, and use the unmanned agricultural machine flight control system to collect the spraying amount at the sampling point; Step S4-4: Calculate the actual operation coverage C act and actual spraying uniformity U act , set the evaluation threshold C std with U std , when the actual operation coverage C act ≥C std And the actual spraying uniformity U act ≤U std When the work is completed, the standard is met; Step S4-5: When the operation does not meet the standards, adjust the parameters in different dimensions according to different influencing factors: Adjust the flight altitude, specifically: ; Where, is the target altitude value of the unmanned agricultural machinery flight altitude, is the target spray width, a is the spray height coefficient, and b is the basic spray width; Adjust the working track spacing, specifically: If the actual working track distance D act >W act , keep Hact The working track spacing is adjusted as follows: ; Where, is the target trajectory distance, is the actual spray width, is the target trajectory overlap rate; Adjust the flight speed, specifically: ; Where, is the target flight speed, F is the fixed spray flow rate, is the target spraying volume per unit area, is the actual spray width; Stabilize the flight altitude and control the fluctuation amplitude to ≤0.5m.

[0010] By setting evaluation indicators to evaluate the results of the work, the evaluation results are quantified, making the work quality assessment more objective. For substandard work, adjustment plans are provided in different dimensions to improve the quality of subsequent work.

[0011] Step S5: Set the posture and operation trajectory adjustment cycle, build a knowledge graph based on the parameter adjustment changes within the cycle, and adjust the posture and operation trajectory of the unmanned agricultural machinery based on the knowledge graph.

[0012] Step S5-1, recording the target operating area of ​​the unmanned agricultural machine, and recording the parameters adjusted through steps S1-S4 according to the operating area of ​​the unmanned agricultural machine; Step S5-2: Create a target table, an actual table, and a parameter table. The target table includes the target operating area and target spraying uniformity of the unmanned agricultural machine. The actual table includes the actual operating area, actual spraying uniformity, and trajectory length of the unmanned agricultural machine. The parameter table includes the spraying width, flight altitude, and flight speed of the unmanned agricultural machine. Step S5-3: Set a unique ID for each target operation area, and associate the data in the actual table with the data in the parameter table through primary and foreign keys in the target table; Step S5-4: When the unmanned agricultural machine is operating, it is compared with the actual table according to the set target operating area to find the same value. Then, the parameters of the unmanned agricultural machine are set by searching the parameters in the parameter table through the index based on the same value obtained by comparison in the actual table, and finally the unmanned agricultural machine is operated.

[0013] By adjusting the current operating parameters of unmanned agricultural machinery based on historical normal data, the operating efficiency of unmanned agricultural machinery is improved. By adjusting the knowledge graph parameters, the operating parameters are updated, which promotes continuous improvement of operating results.

[0014] To achieve the above objectives, the present invention provides an unmanned agricultural machinery operation data analysis system based on three-dimensional scene reconstruction, the system comprising a three-dimensional scene model construction module, a three-dimensional mapping module, an operation behavior determination module, an operation analysis and parameter adjustment module, and a trajectory optimization and knowledge graph construction module; The 3D scene model building module obtains farm spatial information through field surveys, divides the area and performs point cloud acquisition operations, and generates a 3D farm model after pre-processing the data, providing a spatial reference for unmanned agricultural machinery operation trajectory mapping; The 3D mapping module uses dual coordinates to record 3D scenes and unmanned agricultural machinery data. After cleaning and standardizing the original data of the unmanned agricultural machinery, the operation trajectory is mapped to the 3D model and associated through timestamps to achieve 3D visualization of the unmanned agricultural machinery operation. The operation behavior determination module is based on historical multimodal data and accurately identifies the real-time operation type of unmanned agricultural machinery through preprocessing, typical interval division, weighted score calculation and boundary time series auxiliary determination, providing a behavioral basis for operation evaluation; The parameter adjustment module uses operation coverage and spraying uniformity as core indicators to evaluate operation results, compares thresholds to determine compliance, and adjusts flight altitude, trajectory spacing, and flight speed in different dimensions for operations that do not meet the standards to optimize the posture and trajectory of the unmanned agricultural machine. The trajectory optimization and knowledge graph construction module sets an adjustment cycle, records the operating parameters and adjustment data within the cycle, constructs a "target-actual-parameter" association table, and dynamically optimizes the posture and trajectory of unmanned agricultural machinery based on the historical parameters of the knowledge graph.

[0015] The three-dimensional scene model construction module includes a data acquisition unit and a three-dimensional model construction unit; The data collection unit is responsible for collecting farm space information and drawing a sketch; The three-dimensional model construction unit is responsible for constructing a three-dimensional model of the farm based on the sketch and point cloud data.

[0016] The three-dimensional mapping module includes a coordinate and trajectory data acquisition unit and an unmanned agricultural machine posture and trajectory mapping unit; The trajectory data acquisition unit records the coordinate points of the farm's three-dimensional model using the national geodetic coordinate system and the real-time coordinates of the unmanned agricultural machinery using the Beidou positioning geographic coordinate system. The data is then cleaned and timestamped to ensure temporal consistency. The unmanned agricultural machinery posture and trajectory mapping unit maps the unmanned agricultural machinery track coordinate points with timestamps to the farm three-dimensional model according to the coordinate system correspondence, generates the unmanned agricultural machinery spatial position trajectory, and then associates the posture angle data with the track coordinate points through the timestamp.

[0017] The operation behavior determination module includes a multimodal data processing unit and an operation determination unit; The multimodal data processing unit is responsible for collecting multimodal data of normal operation of historical unmanned agricultural machinery and processing the data; The operation determination unit is responsible for dividing typical operation intervals and determining the operation type of the unmanned agricultural machinery.

[0018] The operation analysis and parameter adjustment module includes an achievement evaluation unit and a parameter adjustment unit; The performance evaluation unit is responsible for determining whether the unmanned agricultural machinery operation meets the standards; The parameter adjustment unit is responsible for adjusting the multimodal data of the unmanned agricultural machinery.

[0019] The trajectory optimization and knowledge graph construction module includes an adjustment parameter recording unit and a trajectory optimization unit; The adjustment parameter recording unit sets an adjustment period, records the target operating area and the adjusted core parameters within the period, establishes a database table, and constructs a knowledge graph basic data structure; The trajectory optimization unit is responsible for adjusting the parameters of the operation type required by the current unmanned agricultural machinery based on historical parameters.

[0020] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention ensures that unmanned agricultural machinery can complete operations accurately and efficiently according to the expected goals by constructing a three-dimensional scene model - model mapping - operation behavior judgment - parameter adjustment.

[0021] 2. The present invention uses quantitative evaluation indicators and thresholds to objectively judge whether the operation meets the standards, thereby avoiding quality risks.

[0022] 3. The present invention realizes parameter reuse through knowledge graph, which reduces the cost and time of manual debugging. The overall process reduces manual intervention and shortens operation time through automatic data processing, improves operation efficiency and the intelligence of unmanned agricultural machinery operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 Schematic diagram of the process of the unmanned agricultural machinery operation data analysis method based on three-dimensional scene reconstruction of the present invention; Figure 2 This is a structural diagram of the unmanned agricultural machinery operation data analysis system based on three-dimensional scene reconstruction of the present invention. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0025] Example 1: Figure 1 As shown, the present invention provides a technical solution, a method for analyzing unmanned agricultural machinery operation data based on three-dimensional scene reconstruction, the method comprising the following steps: Step S1: Scan the farm using terrestrial laser scanning technology to construct a three-dimensional scene model; Step S1-1, field survey of farm boundaries, crop distribution, and fixed obstacles; Step S1-2: Draw a farm site survey sketch based on the field survey data, marking crops, boundaries and fixed obstacles; Step S1-3: Split the scanning area according to the grid division method, divide the farm into multiple sub-grids according to the "a×a" grid, mark the center point of each grid and set a target at the center point. a is the side length of the sub-grid, which is determined according to the operating range of the unmanned agricultural machinery; Step S1-4: collect high-density point cloud data from each sub-area, use a statistical filtering algorithm to remove discrete noise points from the point cloud data, then use a voxel grid downsampling method to downsample the data, and finally use a kd tree to deduplicate the point cloud; Step S1-5: Based on the center point coordinate target, the pre-processed multiple sub-region point clouds are spliced ​​and coordinate unified using the target matching splicing method; Step S1-6: Import the spliced ​​overall point cloud model into the 3D modeling software to construct a 3D farm scene model.

[0026] Through field surveys, farm boundaries, crop distribution, and fixed obstacles are clearly defined. Grid division and target setting are combined to avoid missing key areas, providing a benchmark for building a three-dimensional scene model. Data processing provides reliable model support for subsequent unmanned agricultural machinery trajectory mapping.

[0027] Step S2: Mapping the posture and operation trajectory of the unmanned agricultural machine into a three-dimensional scene model; Step S2-1: Use the National Geodetic Coordinate System (CGCS2000) to record the coordinate points of the 3D scene model, use the BeiDou Positioning Geographic Coordinate System to record the coordinate points of the UAV, and construct the UAV track data based on the UAV coordinate points; Step S2-2: cleaning and standardizing the coordinate points and attitude angles in the track data of the unmanned agricultural machinery, and time-stamping the coordinate points and attitude angles of the track data; Step S2-3: Map the coordinate points of the unmanned agricultural machine's track data to the three-dimensional model, and correspond the attitude angle data of the unmanned agricultural machine to the coordinate points of the track data through timestamps.

[0028] The operation trajectory of unmanned agricultural machinery is visualized, facilitating real-time monitoring of its operation status. Timestamps are used to mark the operation trajectory, providing basic data with correct timing for subsequent steps. Step S3: Based on the multimodal operation data of the unmanned agricultural machine, the operation data intervals are divided to determine the operation behavior of the unmanned agricultural machine; Step S3-1: Collect multimodal data of the normal operation of historical unmanned agricultural machinery, perform data cleaning and standardization on the multimodal data, use a filtering algorithm to eliminate environmental interference, use linear interpolation to handle missing values, and use a boxplot method to identify outliers; Step S3-2: Use timestamps to unify the data into the same time series; Step S3-3: Divide the typical intervals according to the characteristics of different operation types of the unmanned agricultural machinery, specifically: Sowing: The flight speed range is V[v1, v2], the flight height range is H[h1, h2], and the spray width range is W[w1, w2]; Irrigation: The flight speed range is V[v2, v3], the flight height range is H[h2, h3], and the spray width range is W[w2, w3]; Fertilization: The flight speed range is V[v3, v4], the flight height range is H[h3, h4], and the spray width range is W[w3, w4]; Step S3-4: Compare the flight speed, flight altitude, and spray width at each moment with the typical intervals, and mark the "candidate operation interval" for each parameter; Step S3-5: Set the spray width weight to 0.4, the flight height weight to 0.3, and the flight speed weight to 0.3, and use the weighted algorithm to calculate the total weight score of each operation type, specifically: ; ; Where, is the weight score of a certain job type, T represents the job type, is the flight speed weight, is the flight altitude weight, is the spray width weight, is the judgment function of flight speed, is the flight altitude judgment function, is the spray width judgment function; Step S3-6: If the total score of a certain type of work is greater than 0.7, it is directly determined as that type of work. If all scores are less than 0.7, the process proceeds to "boundary and timing auxiliary judgment"; Step S3-7: When a certain operation type enters the "boundary and timing auxiliary judgment", the flight speed, flight altitude and spray width of the operation type are judged with reference to the interval division of step S3-3. When the flight speed, flight altitude and spray width of the operation type are exactly at the operation interval division threshold boundary v2 or v3, h2 or h3, w2 or w3 of S3-3, it is exactly in the middle of two operation types and the operation type cannot be determined. Then refer to the operation type of the previous five moments. If the previous five operation types are all the same operation type, it is determined that the operation type is consistent with the previous five operation types. If among the first five moments, the operation type at a certain moment in the middle is different from the other four moments, it is determined to be data fluctuation, and the operation type is still determined to be consistent with the four identical operation types.

[0029] Through historical multimodal data, different operation types can be distinguished and the results of operation type determination can be clarified, helping managers to understand the specific operation content and progress of unmanned agricultural machinery and providing a basis for subsequent operation results evaluation.

[0030] Step S4: Analyze the operation results of the unmanned agricultural machine and adjust the posture and operation trajectory of the unmanned agricultural machine according to the analysis results; Step S4-1: Set the operation coverage and spraying uniformity as the core evaluation indicators, specifically: Assignment coverage: ; Where C is the operation coverage rate, The actual operation coverage area of ​​unmanned agricultural machinery, is the target operating area, W is the spray width, L is the trajectory length, k is the trajectory overlap coefficient, a is the spray height coefficient, b is the basic spray width, and H is the flight height; Spraying uniformity: ; ; ; ; In the formula, U is the spraying uniformity, F is the spraying flow rate, V is the flight speed, and Q is the spraying volume per unit area. is the average spraying amount, is the standard deviation, a is the spray height coefficient, b is the basic spray width, i is the sum index variable, represents the actual spraying amount of the i-th point, and n is the total number of sampling points; Step S4-2: Obtain the actual flight speed V of the unmanned agricultural machine through the flight control system act , actual flight altitude H actAnd the actual spray width W act ; Step S4-3: Set the target operating area before the unmanned agricultural machine is completed, collect the actual coverage area after the unmanned agricultural machine is completed, and use the unmanned agricultural machine flight control system to collect the spraying amount at the sampling point; Step S4-4: Calculate the actual operation coverage C act and actual spraying uniformity U act , set the evaluation threshold C std with U std , when the actual operation coverage C act ≥C std And the actual spraying uniformity U act ≤U std When the work is completed, the standard is met; Step S4-5: When the operation does not meet the standards, adjust the parameters in different dimensions according to different influencing factors: Adjust the flight altitude, specifically: ; Where, is the target altitude value of the unmanned agricultural machinery flight altitude, is the target spray width, a is the spray height coefficient, and b is the basic spray width; Adjust the working track spacing, specifically: If the actual track spacing D act >W act , keep H act The track spacing remains unchanged, and is adjusted as follows: ; Where, is the target trajectory distance, is the actual spray width, is the target trajectory overlap rate; Adjust the flight speed, specifically: ; Where, is the target flight speed, F is the fixed spray flow rate, is the target spraying volume per unit area, is the actual spray width; Stabilize the flight altitude and control the fluctuation amplitude to ≤0.5m.

[0031] By setting evaluation indicators to evaluate the results of the work, the evaluation results are quantified, making the work quality assessment more objective. For substandard work, adjustment plans are provided in different dimensions to improve the quality of subsequent work.

[0032] Step S5: Set the posture and operation trajectory adjustment cycle, build a knowledge graph based on the parameter adjustment changes within the cycle, and adjust the posture and operation trajectory of the unmanned agricultural machinery based on the knowledge graph.

[0033] Step S5-1, recording the target operating area of ​​the unmanned agricultural machine, and recording the parameters adjusted through steps S1-S4 according to the operating area of ​​the unmanned agricultural machine; Step S5-2: Create a target table, an actual table, and a parameter table. The target table includes the target operating area and target spraying uniformity of the unmanned agricultural machine. The actual table includes the actual operating area, actual spraying uniformity, and trajectory length of the unmanned agricultural machine. The parameter table includes the spraying width, flight altitude, and flight speed of the unmanned agricultural machine. Step S5-3: Set a unique ID for each target operation area, and associate the data in the actual table with the data in the parameter table through primary and foreign keys in the target table; Step S5-4: When the unmanned agricultural machine is operating, it is compared with the actual table according to the set target operating area to find the same value. Then, the parameters of the unmanned agricultural machine are set by searching the parameters in the parameter table through the index based on the same value obtained by comparison in the actual table, and finally the unmanned agricultural machine is operated.

[0034] By adjusting the current operating parameters of unmanned agricultural machinery based on historical normal data, the operating efficiency of unmanned agricultural machinery is improved. By adjusting the knowledge graph parameters, the operating parameters are updated, which promotes continuous improvement of operating results.

[0035] To achieve the above objectives, the present invention provides an unmanned agricultural machinery operation data analysis system based on three-dimensional scene reconstruction, the system comprising a three-dimensional scene model construction module, a three-dimensional mapping module, an operation behavior determination module, an operation analysis and parameter adjustment module, and a trajectory optimization and knowledge graph construction module; The 3D scene model building module obtains farm spatial information through field surveys, divides the area and performs point cloud acquisition operations, and generates a 3D farm model after pre-processing the data, providing a spatial reference for unmanned agricultural machinery operation trajectory mapping; The 3D mapping module uses dual coordinates to record 3D scenes and unmanned agricultural machinery data. After cleaning and standardizing the original data of the unmanned agricultural machinery, the operation trajectory is mapped to the 3D model and associated through timestamps to achieve 3D visualization of the unmanned agricultural machinery operation. The operation behavior determination module is based on historical multimodal data and accurately identifies the real-time operation type of unmanned agricultural machinery through preprocessing, typical interval division, weighted score calculation and boundary time series auxiliary determination, providing a behavioral basis for operation evaluation; The parameter adjustment module uses operation coverage and spraying uniformity as core indicators to evaluate operation results, compares thresholds to determine compliance, and adjusts flight altitude, trajectory spacing, and flight speed in different dimensions for operations that do not meet the standards to optimize the posture and trajectory of the unmanned agricultural machine. The trajectory optimization and knowledge graph construction module sets an adjustment cycle, records the operating parameters and adjustment data within the cycle, constructs a "target-actual-parameter" association table, and dynamically optimizes the posture and trajectory of unmanned agricultural machinery based on the historical parameters of the knowledge graph.

[0036] The three-dimensional scene model construction module includes a data acquisition unit and a three-dimensional model construction unit; The data collection unit is responsible for collecting farm space information and drawing a sketch; The three-dimensional model construction unit is responsible for constructing a three-dimensional model of the farm based on the sketch and point cloud data.

[0037] The three-dimensional mapping module includes a coordinate and trajectory data acquisition unit and an unmanned agricultural machine posture and trajectory mapping unit; The trajectory data acquisition unit records the coordinate points of the farm's three-dimensional model using the national geodetic coordinate system and the real-time coordinates of the unmanned agricultural machinery using the Beidou positioning geographic coordinate system. The data is then cleaned and timestamped to ensure temporal consistency. The unmanned agricultural machinery posture and trajectory mapping unit maps the unmanned agricultural machinery track coordinate points with timestamps to the farm three-dimensional model according to the coordinate system correspondence, generates the unmanned agricultural machinery spatial position trajectory, and then associates the posture angle data with the track coordinate points through the timestamp.

[0038] The operation behavior determination module includes a multimodal data processing unit and an operation determination unit; The multimodal data processing unit is responsible for collecting multimodal data of normal operation of historical unmanned agricultural machinery and processing the data; The operation determination unit is responsible for dividing typical operation intervals and determining the operation type of the unmanned agricultural machinery.

[0039] The operation analysis and parameter adjustment module includes an achievement evaluation unit and a parameter adjustment unit; The performance evaluation unit is responsible for determining whether the unmanned agricultural machinery operation meets the standards; The parameter adjustment unit is responsible for adjusting the multimodal data of the unmanned agricultural machinery.

[0040] The trajectory optimization and knowledge graph construction module includes an adjustment parameter recording unit and a trajectory optimization unit; The adjustment parameter recording unit sets an adjustment period, records the target operating area and the adjusted core parameters within the period, establishes a database table, and constructs a knowledge graph basic data structure; The trajectory optimization unit is responsible for adjusting the parameters of the operation type required by the current unmanned agricultural machinery based on historical parameters.

[0041] Example 2: The scale of the survey farm is 100m long, 50m wide, and the total area is 5000m 2 The middle part is a wheat planting area, surrounded by 1m wide roads, and the fixed obstacles are three irrigation wells; Set the unmanned agricultural machinery spraying fixed flow rate F = 1.2L / min, spray height coefficient a = 0.05m -1 , basic spray width b = 1.5m, complete the fertilization operation in the wheat planting area, the target area is 4000m 2 , target coverage is 95%, target spraying uniformity U=15%, target spraying volume per unit area Q=0.3L / m 2 , target trajectory overlap rate k=10%; The farm boundaries were surveyed on site as (0,0)-(100,0)-(100,50)-(0,50), the wheat planting area boundaries were marked as (5,5)-(95,5)-(5,45) and three irrigation locations, and a survey sketch was drawn; Based on the 10m operating radius of the unmanned agricultural machinery, the subgrid side length a is set to 10m, divided into 50 subgrids, and the center coordinates of each subgrid are (5,5), (15,5)...(95,45), and targets are set at each center point; A laser scanner is used to collect point cloud data of the sub-area, and discrete noise points are proposed through a statistical filtering algorithm. Voxel grid downsampling is used, and then duplicate removal is performed through a kd-tree. Based on the target coordinates, the target matching method was used to stitch the point clouds of 50 sub-regions together, and the coordinates were unified to the CGCS2000 coordinate system with a stitching error of ≤0.05m. Import the spliced ​​point cloud into 3D modeling software to generate a 3D farm scene model; Example 3: Collect multimodal data of historical fertilization operations, process the data and mark it with a timestamp, and divide it into typical operation intervals, fertilization operations V=[1.5, 2.5], W=[2.0, 3.0], H=[2.5, 3.5]; Taking the data at 10:05:00 as an example, V = 2.0m / s, H = 3.0m, and W = 2.5m, mark the three as "fertilization candidate intervals". According to the weighted algorithm, the score = 1.0, and the score > 0.7 is determined to be a fertilization operation; Example 4: After the operation is completed, the actual flight speed V is obtained through the flight control system act =2.2m / s, actual flight altitude H act =3.2m, actual spray width W act =2.6m, actual operation coverage area Sact =3750m 2 , collected the spraying amount of 20 sampling points: 0.28, 0.32, 0.29, 0.31, 0.30, 0.27, 0.33, 0.28, 0.30, 0.31, 0.29, 0.32, 0.28, 0.30, 0.31, 0.29, 0.32, 0.27, 0.33, 0.30; The calculated operation coverage rate was 93.75% < 95%, which was judged as not meeting the standard; Spraying volume per unit area = 3.652L / min, converted to unit area: Q = 0.21L / m 2 , spraying uniformity U=6.67<15%, determined to be substandard; It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. An unmanned agricultural machinery operation data analysis method based on 3D scene reconstruction, characterized by: The method comprises the following steps: Step S1: Scan the farm using terrestrial laser scanning technology to construct a three-dimensional scene model; Step S2: Mapping the posture and operation trajectory of the unmanned agricultural machine into a three-dimensional scene model; Step S3: Based on the multimodal operation data of the unmanned agricultural machine, the operation data intervals are divided to determine the operation behavior of the unmanned agricultural machine; Step S4: Analyze the operation results of the unmanned agricultural machine and adjust the posture and operation trajectory of the unmanned agricultural machine according to the analysis results; Step S5: Set the posture and operation trajectory adjustment cycle, build a knowledge graph based on the parameter adjustment changes within the cycle, and adjust the posture and operation trajectory of the unmanned agricultural machinery based on the knowledge graph.

2. The method for analyzing unmanned agricultural machinery operation data based on three-dimensional scene reconstruction according to claim 1, characterized in that: The specific steps of step S1 are as follows: Step S1-1, field survey of farm boundaries, crop distribution, and fixed obstacles; Step S1-2: Draw a farm site survey sketch based on the field survey data, marking crops, boundaries and fixed obstacles; Step S1-3: Split the scanning area according to the grid division method. Divide the farm into multiple subgrids according to the "a×a" grid. Mark the center point of each grid and set a target at the center point. a is the side length of the subgrid, which is determined according to the operating range of the unmanned agricultural machinery. Step S1-4: collect high-density point cloud data from each sub-area, use a statistical filtering algorithm to remove discrete noise points from the point cloud data, then use a voxel grid downsampling method to downsample the data, and finally use a kd tree to deduplicate the point cloud; Step S1-5: Based on the center point coordinate target, the pre-processed multiple sub-region point clouds are spliced ​​and coordinate unified using the target matching splicing method; Step S1-6: Import the spliced ​​overall point cloud model into the 3D modeling software to construct a 3D farm scene model.

3. The method for analyzing unmanned agricultural machinery operation data based on three-dimensional scene reconstruction according to claim 2, characterized in that: The specific steps of step S2 are as follows: Step S2-1: Use the National Geodetic Coordinate System (CGCS2000) to record the coordinate points of the 3D scene model, use the BeiDou Positioning Geographic Coordinate System to record the coordinate points of the UAV, and construct the UAV track data based on the UAV coordinate points; Step S2-2: cleaning and standardizing the coordinate points and attitude angles in the track data of the unmanned agricultural machinery, and time-stamping the coordinate points and attitude angles of the track data; Step S2-3: Map the coordinate points of the unmanned agricultural machine's track data to the three-dimensional model, and correspond the attitude angle data of the unmanned agricultural machine to the coordinate points of the track data through timestamps.

4. The method for analyzing unmanned agricultural machinery operation data based on three-dimensional scene reconstruction according to claim 3, characterized in that: The specific steps of step S3 are as follows: Step S3-1: Collect multimodal data of the normal operation of historical unmanned agricultural machinery, perform data cleaning and standardization on the multimodal data, use a filtering algorithm to eliminate environmental interference, use linear interpolation to handle missing values, and use a boxplot method to identify outliers; Step S3-2: Use timestamps to unify the data into the same time series; Step S3-3: Divide the typical intervals according to the characteristics of different operation types of the unmanned agricultural machinery, specifically: Sowing: The flight speed range is V[v1, v2], the flight height range is H[h1, h2], and the spray width range is W[w1, w2]; Irrigation: The flight speed range is V[v2, v3], the flight height range is H[h2, h3], and the spray width range is W[w2, w3]; Fertilization: The flight speed range is V[v3, v4], the flight height range is H[h3, h4], and the spray width range is W[w3, w4]; Step S3-4: Compare the flight speed, flight altitude, and spray width at each moment with the typical intervals, and mark the "candidate operation interval" for each parameter; Step S3-5: Set the spray width weight to 0.4, the flight height weight to 0.3, and the flight speed weight to 0.3, and use the weighted algorithm to calculate the total weight score of each operation type, specifically: ； ； Where, is the weight score of a certain job type, T represents the job type, is the flight speed weight, is the flight altitude weight, is the spray width weight, is the judgment function of flight speed, is the flight altitude judgment function, is the spray width judgment function; Step S3-6: If the total score of a certain type of work is greater than 0.7, the work is directly determined to be of that type. If all scores are less than 0.7, the work proceeds to the "boundary and timing auxiliary determination" step. Step S3-7: When a certain operation type enters the "boundary and timing auxiliary determination" phase, the flight speed, flight altitude, and spray width of that operation type are determined with reference to the interval division in step S3-3. If the flight speed, flight altitude, and spray width of that operation type fall exactly within the operation interval division threshold boundaries v2 or v3, h2 or h3, or w2 or w3 in S3-3, it is located between two operation types and the operation type cannot be determined. The operation types at the previous five moments are then referenced. If the previous five operation types are all the same, the operation type is determined to be consistent with the previous five. If, among the previous five moments, the operation type at any intermediate moment is different from that at the remaining four moments, it is determined to be data fluctuation, and the operation type is still determined to be consistent with the four identical operation types.

5. The method for analyzing unmanned agricultural machine operation data based on three-dimensional scene reconstruction according to claim 4 is characterized in that: The specific steps of step S4 are as follows: Step S4-1: Set the operation coverage and spraying uniformity as the core evaluation indicators, specifically: Assignment coverage: ； Where C is the operation coverage rate, The actual operation coverage area of ​​unmanned agricultural machinery, is the target operating area, W is the spray width, L is the trajectory length, k is the trajectory overlap coefficient, a is the spray height coefficient, b is the basic spray width, and H is the flight height; Spraying uniformity: ; ; ; ; In the formula, U is the spraying uniformity, F is the spraying flow rate, V is the flight speed, and Q is the spraying volume per unit area. is the average spraying amount, is the standard deviation, a is the spray height coefficient, b is the basic spray width, i is the sum index variable, represents the actual spraying amount of the i-th point, and n is the total number of sampling points; Step S4-2: Obtain the actual flight speed V of the unmanned agricultural machine through the flight control system act , actual flight altitude H act And the actual spray width W act ; Step S4-3: Set the target operating area before the unmanned agricultural machine is completed, collect the actual coverage area after the unmanned agricultural machine is completed, and use the unmanned agricultural machine flight control system to collect the spraying amount at the sampling point; Step S4-4: Calculate the actual operation coverage C act and actual spraying uniformity U act , set the evaluation threshold C std with U std , when the actual operation coverage C act ≥C std And the actual spraying uniformity U act ≤U std When the work is completed, the standard is met; Step S4-5: When the operation does not meet the standards, adjust the parameters in different dimensions according to different influencing factors: Adjust the flight altitude, specifically: ； Where, is the target altitude value of the unmanned agricultural machinery flight altitude, is the target spray width, a is the spray height coefficient, and b is the basic spray width; Adjust the working track spacing, specifically: If the actual working track distance D act >W act , keep H act The working track spacing is adjusted as follows: ； Where, is the target trajectory distance, is the actual spray width, is the target trajectory overlap rate; Adjust the flight speed, specifically: ； Where, is the target flight speed, F is the fixed spray flow rate, is the target spraying volume per unit area, is the actual spray width; Stabilize the flight altitude and control the fluctuation amplitude to ≤0.5m.

6. The method for analyzing unmanned agricultural machinery operation data based on three-dimensional scene reconstruction according to claim 5, characterized in that: The specific steps of step S5 are as follows: Step S5-1, recording the target operating area of ​​the unmanned agricultural machine, and recording the parameters adjusted through steps S1-S4 according to the operating area of ​​the unmanned agricultural machine; Step S5-2: Create a target table, an actual table, and a parameter table. The target table includes the target operating area and target spraying uniformity of the unmanned agricultural machine. The actual table includes the actual operating area, actual spraying uniformity, and trajectory length of the unmanned agricultural machine. The parameter table includes the spraying width, flight altitude, and flight speed of the unmanned agricultural machine. Step S5-3: Set a unique ID for each target operation area, and associate the data in the actual table with the data in the parameter table through primary and foreign keys in the target table; Step S5-4: When the unmanned agricultural machine is operating, it is compared with the actual table according to the set target operating area to find the same value. Then, the parameters of the unmanned agricultural machine are set by searching the parameters in the parameter table through the index based on the same value obtained by comparison in the actual table, and finally the unmanned agricultural machine is operated.

7. Unmanned agricultural machinery operation data analysis system based on 3D scene reconstruction, characterized by: The system includes a three-dimensional scene model construction module, a three-dimensional mapping module, an operation behavior determination module, an operation analysis and parameter adjustment module, and a trajectory optimization and knowledge graph construction module; The 3D scene model building module obtains farm spatial information through field surveys, divides the area and performs point cloud acquisition operations, and generates a 3D farm model after pre-processing the data, providing a spatial reference for unmanned agricultural machinery operation trajectory mapping; The 3D mapping module uses dual coordinates to record 3D scenes and unmanned agricultural machinery data. After cleaning and standardizing the original data of the unmanned agricultural machinery, the operation trajectory is mapped to the 3D model and associated through timestamps to achieve 3D visualization of the unmanned agricultural machinery operation. The operation behavior determination module is based on historical multimodal data and accurately identifies the real-time operation type of unmanned agricultural machinery through preprocessing, typical interval division, weighted score calculation and boundary time series auxiliary determination, providing a behavioral basis for operation evaluation; The parameter adjustment module uses operation coverage and spraying uniformity as core indicators to evaluate operation results, compares thresholds to determine compliance, and adjusts flight altitude, trajectory spacing, and flight speed in different dimensions for operations that do not meet the standards to optimize the posture and trajectory of the unmanned agricultural machine. The trajectory optimization and knowledge graph construction module sets an adjustment cycle, records the operating parameters and adjustment data within the cycle, constructs a "target-actual-parameter" association table, and dynamically optimizes the posture and trajectory of unmanned agricultural machinery based on the historical parameters of the knowledge graph.

8. The unmanned agricultural machinery operation data analysis system based on 3D scene reconstruction according to claim 7 is characterized in that: The three-dimensional scene model construction module includes a data acquisition unit and a three-dimensional model construction unit; The data collection unit is responsible for collecting farm space information and drawing a sketch; The three-dimensional model construction unit is responsible for constructing a three-dimensional model of the farm based on the sketch and point cloud data; The three-dimensional mapping module includes a coordinate and trajectory data acquisition unit and an unmanned agricultural machine posture and trajectory mapping unit; The trajectory data acquisition unit records the coordinate points of the farm's three-dimensional model using the national geodetic coordinate system and the real-time coordinates of the unmanned agricultural machinery using the Beidou positioning geographic coordinate system. The data is then cleaned and timestamped to ensure temporal consistency. The unmanned agricultural machinery posture and trajectory mapping unit maps the unmanned agricultural machinery track coordinate points with timestamps to the farm three-dimensional model according to the coordinate system correspondence, generates the unmanned agricultural machinery spatial position trajectory, and then associates the posture angle data with the track coordinate points through the timestamp.

9. The unmanned agricultural machinery operation data analysis system based on 3D scene reconstruction according to claim 7, characterized in that: The operation behavior determination module includes a multimodal data processing unit and an operation determination unit; The multimodal data processing unit is responsible for collecting multimodal data of normal operation of historical unmanned agricultural machinery and processing the data; The operation determination unit is responsible for dividing typical operation intervals and determining the operation type of the unmanned agricultural machinery; The operation analysis and parameter adjustment module includes an achievement evaluation unit and a parameter adjustment unit; The performance evaluation unit is responsible for determining whether the unmanned agricultural machinery operation meets the standards; The parameter adjustment unit is responsible for adjusting the multimodal data of the unmanned agricultural machinery.

10. The unmanned agricultural machinery operation data analysis system based on 3D scene reconstruction according to claim 7, characterized in that: The trajectory optimization and knowledge graph construction module includes an adjustment parameter recording unit and a trajectory optimization unit; The adjustment parameter recording unit sets an adjustment period, records the target operating area and the adjusted core parameters within the period, establishes a database table, and constructs a knowledge graph basic data structure; The trajectory optimization unit is responsible for adjusting the parameters of the operation type required by the current unmanned agricultural machinery based on historical parameters.

Images