WO2021068177A1 - Agricultural machinery operation area calculation method based on positioning drift calculation model - Google Patents

Agricultural machinery operation area calculation method based on positioning drift calculation model Download PDF

Info

Publication number
WO2021068177A1
WO2021068177A1 PCT/CN2019/110510 CN2019110510W WO2021068177A1 WO 2021068177 A1 WO2021068177 A1 WO 2021068177A1 CN 2019110510 W CN2019110510 W CN 2019110510W WO 2021068177 A1 WO2021068177 A1 WO 2021068177A1
Authority
WO
WIPO (PCT)
Prior art keywords
area
trajectory
agricultural machinery
positioning
blank
Prior art date
Application number
PCT/CN2019/110510
Other languages
French (fr)
Chinese (zh)
Inventor
黄河
吴晓伟
张炜
史杨
Original Assignee
安徽中科智能感知产业技术研究院有限责任公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 安徽中科智能感知产业技术研究院有限责任公司 filed Critical 安徽中科智能感知产业技术研究院有限责任公司
Priority to CN201980016520.2A priority Critical patent/CN111868566B/en
Priority to PCT/CN2019/110510 priority patent/WO2021068177A1/en
Publication of WO2021068177A1 publication Critical patent/WO2021068177A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
    • G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
    • G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
    • G01S19/53—Determining attitude
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B7/00—Measuring arrangements characterised by the use of electric or magnetic techniques
    • G01B7/32—Measuring arrangements characterised by the use of electric or magnetic techniques for measuring areas
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
    • G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
    • G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
    • G01S19/42—Determining position

Definitions

  • the invention relates to a method for measuring and calculating the working area of agricultural machinery based on a positioning drift measuring and calculating model.
  • Deep loosening of agricultural machinery and land preparation operations are of great significance for realizing soil moisture conservation and improving farmland compaction.
  • the installation of digital technology navigation devices such as geographic information systems and spatial positioning systems on agricultural machinery, combined with the back-end service system mode, can meet the needs of remote monitoring and management of agricultural machinery's deep loosening and land preparation operations, and realize accurate operations of agricultural production, cultivation, and harvesting.
  • New agricultural machinery management and market-oriented service models are becoming more and more mature. Both supply and demand parties require agricultural machinery operation services to provide high-precision, high-reliability, real-time and convenient agricultural machinery operating area calculation results.
  • the purpose of the present invention is to provide a method for measuring and calculating the working area of agricultural machinery based on a positioning drift measurement model to solve the problem of calculating the working area of agricultural machinery when there are multiple types of operations in the process of agricultural machinery farming in the prior art, and overlapping areas and non-operating areas are generated.
  • the problem of low precision is to provide a method for measuring and calculating the working area of agricultural machinery based on a positioning drift measurement model to solve the problem of calculating the working area of agricultural machinery when there are multiple types of operations in the process of agricultural machinery farming in the prior art, and overlapping areas and non-operating areas are generated.
  • the problem of low precision is to provide a method for measuring and calculating the working area of agricultural machinery based on a positioning drift measurement model to solve the problem of calculating the working area of agricultural machinery when there are multiple types of operations in the process of agricultural machinery farming in the prior art, and overlapping areas and non-operating areas are generated.
  • the described method for measuring and calculating the working area of agricultural machinery based on the positioning drift measurement model includes the following steps:
  • Step 1 Collecting and preprocessing the track points of agricultural machinery operation: preprocess the data of the agricultural machinery positioning sensor and the attitude sensor, filter the agricultural machinery operation status information, and construct the track point time series in line with the operation status;
  • Step 2 Construct positioning trajectory and operation trajectory: connect the locating points that conform to the operation status in turn to establish a polyline path of the operation trajectory; at the same time, every two points generate a quadrilateral according to the operation width, and multiple quadrilaterals perform logical operations to create a polygon;
  • Step 3 Calculate the coverage area of the job trajectory: calculate the trajectory area, coverage area and blank area of the job trajectory from different angles;
  • Step 4 Construct a positioning drift measurement model: analyze different trajectory coverage types, extract core features for trajectory coverage area measurement, use the extracted core features to build a positioning drift measurement model, and train samples to learn model parameters;
  • Step 5 Measurement and calculation of agricultural machinery operating area: Input the operating track and operating area to be calculated into the positioning drift measurement model to obtain an accurate agricultural machinery operating area.
  • the step 1 specifically includes the following steps:
  • Step 1.1 Receive data returned every second from GPS sensors and attitude sensors installed on agricultural machinery;
  • Step 1.2 Filter out the positioning points of agricultural machinery that are too close or too far within the interval
  • Step 1.3 Keep the filtered track points that have reached the national deep ploughing requirements as effective track points for agricultural machinery operations and put them into the collection to construct a time sequence of track points.
  • the step 2 specifically includes the following steps:
  • Step 2.1 Connect the positioning points that meet the operating status in step 1 one by one to establish a broken line path of the operating track;
  • Step 2.2 Calculate the azimuth angle between the two adjacent trajectory points P1 and P2 on the agricultural machinery operation trajectory in time series;
  • Step 2.3 Calculate the azimuth angle of the agricultural machinery on the track between the aforementioned two adjacent track points according to the fact that the agricultural machinery is perpendicular to the track;
  • Step 2.4 According to the azimuth angle of the agricultural machinery and the length of the plough R, the four points L1, L2, L3, and L4 of the trajectory points P1 and P2 are obtained, and then constitute the quadrilateral S1 of the coverage area of the trajectory, and so on, calculate All quadrilaterals S1...Sn on the trajectory path;
  • Step 2.5 Perform logical operations on the quadrilaterals S1...Sn obtained in step 2.4 to establish a total work area polygon.
  • the step 3 specifically includes the following steps:
  • Step 3.1 Calculate the theoretical track area of the job: connect the positioning points that meet the job status in turn, calculate the distance of each track separately, and count the area covered by the track according to the width of the job;
  • Step 3.2 Calculate the theoretical coverage area of the job: Calculate the polygons obtained from the logical operation of the small rectangle formed by the two job points based on graphics, and calculate the theoretical coverage area, external contour area, and internal blank area.
  • the step 4 specifically includes the following steps:
  • Step 4.1 Determine the elements that reflect the drift of coordinate positioning, and then extract the core features used to realize the blank area analysis, which are local overlap, global overlap, and global coverage;
  • Step 4.2 Optimize the parameters of the core feature, train the parameters of the core feature according to the sample data, and map the core feature to a specific unified interval [0,1];
  • Step 4.3 Construct a positioning drift measurement model, use the optimized parameters of the core features, and use machine learning methods to establish a positioning drift measurement model.
  • the step 4.1 specifically includes the following steps:
  • Effective work trajectory take a limited area of the envelope is represented as a map of area outside the outer S, in a limited region of the envelope, generating a locus of points forming a quadrangle combined: step 4.1.1, determines a coordinate positioning elements reflect drift
  • the coverage area is expressed as the inner area S inside , there is a difference between S outside and S inside , that is, the internal blank area is expressed as S empty , and the track covered area calculated by combining the length of the track with the width of the plough R is called the track area S Rail
  • Step 4.1.2 Extract the local overlap, global overlap, and global coverage of core features for blank area analysis:
  • the degree of local overlap is defined as the ratio of the track area to the inner area, expressed by ⁇ ,
  • the degree of local overlap ⁇ expresses the degree of local overlap in the trajectory. The denser the trajectory and the more concentrated the overlap area, the greater the degree of local overlap. Ideally, the degree of local overlap is close to 1, which means that the trajectory has no drift and the operation is normal.
  • the global overlap is defined as the ratio of the track area to the outer area, denoted by ⁇
  • the global degree of overlap ⁇ expresses the degree of overall dispersion within the trajectory. The more uniform the distribution of trajectory points and the more scattered the overlap area, the greater the global degree of overlap. Ideally, for a complete operation trajectory, its global overlap is close to 1.
  • the global coverage is defined as the ratio of the inner area to the outer area, denoted by ⁇
  • the global coverage ⁇ expresses the extent to which the actual work area covers the work area. The more standardized the job and the more uniform the trajectory, the greater the global coverage; if the job is abnormal and the overlap is concentrated, the global coverage is smaller.
  • the step 4.2 specifically includes the following steps:
  • Step 4.2.1 Optimize the parameters of the core features, select the sigmod function to map the parameters to the interval [0,1]
  • Step 4.2.2 Train the parameters x and y of the core feature according to the sample data, and determine the specific mapping functions K 1 ( ⁇ ) and K 2 ( ⁇ ).
  • the step 4.3 specifically includes the following steps:
  • the step 4.3 specifically includes the following steps:
  • Step 4.3.1 Construct a positioning drift measurement model, and use the optimized core feature parameters for formal expression
  • the method defines the redundancy rate of the trajectory as the proportion of the redundant operation area in the internal area of the ⁇ redundant operation trajectory:
  • the blank rate ⁇ blank of the trajectory as the ratio of the blank area generated by abnormal operations in the actual trajectory to the blank area after removing the redundant area of the operation, then:
  • Step 4.3.2 Based on the optimized core feature parameters, use machine learning methods to establish a positioning drift measurement model
  • f ( ⁇ , ⁇ ) is a function S S abnormality calculating the proportion of air in which the blank of the known formula ⁇ blank, it is also necessary to establish a blank ⁇ calculation model;
  • the step 5 specifically includes the following steps:
  • Step 5.1 Preprocess the positioning points into a trajectory sequence, calculate the values of different features and parameters, and generate the trajectory to be processed;
  • Step 5.2 Input the trajectory to be processed into the positioning drift measurement model to realize different classification of trajectories, and correct and compensate the drift area to obtain the actual area of agricultural machinery operation.
  • the present invention uses logical operations to construct a work area polygon, analyzes different track coverage types, extracts core features of track coverage area calculations to construct a positioning drift measurement model, and finally obtains accurate agricultural machinery operations area.
  • the measurement model can be continuously revised to improve accuracy.
  • abnormal operation There are two cases of abnormal operation, one is called redundancy, which is the case where the subsequent global overlap is less than 1, and the subsequent filling is performed by other tracks; the other is the blank area that is not covered in the operation process caused by the abnormal operation; the abnormal operation occurs
  • the blank area and the blank area displayed by the positioning drift together constitute the blank area that appears during the measurement.
  • this method provides redundancy rate and blank rate for calculation and analysis, and continuously corrects the blank rate that is not easy to calculate through machine learning, so as to ensure that the error part caused by abnormal operation can be accurately calculated ,
  • Fig. 1 is a flowchart of a method for measuring and calculating the working area of agricultural machinery based on a positioning drift measuring model of the present invention.
  • the present invention provides a method for measuring and calculating the working area of agricultural machinery based on a positioning drift measurement model, which includes the following steps:
  • Step 1 Collecting and preprocessing the track points of agricultural machinery operation: preprocess the data of the agricultural machinery positioning sensor and the attitude sensor, filter the agricultural machinery operation status information, and construct the track point time series in line with the operation status; the specific steps are:
  • Step 1.1 Receive data returned every second from GPS sensors and attitude sensors installed on agricultural machinery;
  • Step 1.2 Filter out the positioning points of agricultural machinery that are too close or too far within the interval
  • Step 1.3 Keep the filtered track points that have reached the national deep ploughing requirements as effective track points for agricultural machinery operations and put them into the collection to construct a time sequence of track points.
  • Step 2 Construct positioning trajectory and operation trajectory: connect the positioning points that conform to the operation status in turn to establish a polyline path of the operation trajectory; at the same time, every two points generate a quadrilateral according to the operation width, and multiple quadrilaterals perform logical operations to create a polygon;
  • the specific steps are:
  • Step 2.1 Connect the positioning points that meet the operating status in step 1 one by one to establish a broken line path of the operating track;
  • Step 2.2 Calculate the azimuth angle between the two adjacent trajectory points P1 and P2 on the agricultural machinery operation trajectory in time series;
  • Step 2.3 Calculate the azimuth angle of the agricultural machinery on the track between the aforementioned two adjacent track points according to the fact that the agricultural machinery is perpendicular to the track;
  • Step 2.4 According to the azimuth angle of the agricultural machinery and the length of the plough R, the four points L1, L2, L3, and L4 of the trajectory points P1 and P2 are obtained, and then constitute the quadrilateral S1 of the coverage area of the trajectory, and so on, calculate All quadrilaterals S1...Sn on the trajectory path;
  • Step 2.5 Perform logical operations on the quadrilaterals S1...Sn obtained in step 2.4 to establish a total work area polygon.
  • the total work area polygon provides the basis for subsequent area calculations.
  • Step 3 Calculate the coverage area of the job trajectory: calculate the trajectory area, coverage area and blank area of the job trajectory from different angles; specifically include the following steps:
  • Step 3.1 Calculate the theoretical track area of the job: connect the positioning points that meet the job status in turn, calculate the distance of each track separately, and count the area covered by the track based on the job width, which includes the overlapped part of each track;
  • Step 3.2 Calculate the theoretical coverage area of the job: perform graphics-based calculations on the polygons obtained by the logical operation of the small rectangles formed by the two job points, and calculate the theoretical coverage area, external contour area, and internal blank area.
  • the outer contour area is the area enclosed by the outer contour of the formed polygon
  • the blank area is the area of the blank part in the outer contour formed by the polygon
  • the theoretical coverage area is the coverage area of the polygon itself.
  • the above-mentioned area is provided to the constructed positioning drift measurement model to obtain the actual operating area.
  • Step 4 Construct a positioning drift measurement model: analyze different trajectory coverage types, extract core features for trajectory coverage area measurement, use the extracted core features to construct a positioning drift measurement model, and train samples to learn model parameters; specifically, it includes the following steps:
  • Step 4.1 Determine the elements that reflect the drift of coordinate positioning, and then extract the core features used to realize the blank area analysis, which are local overlap, global overlap, and global coverage; this step specifically includes the following steps:
  • Effective work trajectory take a limited area of the envelope is represented as a map of area outside the outer S, in a limited region of the envelope, generating a locus of points forming a quadrangle combined: step 4.1.1, determines a coordinate positioning elements reflect drift
  • the coverage area is expressed as the inner area S inside , there is a difference between S outside and S inside , that is, the internal blank area is expressed as S empty , and the track covered area calculated by combining the length of the track with the width of the plough R is called the track area S Rail
  • Step 4.1.2 Extract the local overlap, global overlap, and global coverage of core features for blank area analysis:
  • the degree of local overlap is defined as the ratio of the track area to the inner area, expressed by ⁇ ,
  • the degree of local overlap ⁇ expresses the degree of local overlap in the trajectory. The denser the trajectory and the more concentrated the overlap area, the greater the degree of local overlap. Ideally, the degree of local overlap is close to 1, which means that the trajectory has no drift and the operation is normal.
  • the global overlap is defined as the ratio of the track area to the outer area, denoted by ⁇
  • the global degree of overlap ⁇ expresses the degree of overall dispersion within the trajectory. The more uniform the distribution of trajectory points and the more scattered the overlap area, the greater the global degree of overlap. Ideally, for a complete operation trajectory, its global overlap is close to 1.
  • the global coverage is defined as the ratio of the inner area to the outer area, denoted by ⁇
  • the global coverage ⁇ expresses the extent to which the actual work area covers the work area. The more standardized the job and the more uniform the trajectory, the greater the global coverage; if the job is abnormal and the overlap is concentrated, the global coverage is smaller.
  • the above-mentioned core features can be obtained from the area data calculated in step 3, which are important parameters required to obtain the actual working area, and different trajectory coverage types are determined according to the difference of the above-mentioned core features.
  • Step 4.2 Optimize the parameters of the core feature, train the parameters of the core feature according to the sample data, and map the core feature to a specific uniform interval [0,1]; this step specifically includes the following steps:
  • Step 4.2.1 Optimize the parameters of the core features, select the sigmod function to map the parameters to the interval [0,1]
  • Step 4.2.2 Train the parameters x and y of the core feature according to the sample data, and determine the specific mapping functions K 1 ( ⁇ ) and K 2 ( ⁇ ).
  • the mapping function is very important for subsequent area analysis of abnormal conditions.
  • Step 4.3 Construct a positioning drift measurement model, use the optimized parameters of the core features, and use machine learning methods to establish a positioning drift measurement model. This step specifically includes the following steps:
  • Step 4.3.1 Construct a positioning drift measurement model, and use the optimized core feature parameters for formal expression
  • the method defines the redundancy rate of the trajectory as the proportion of the redundant operation area in the internal area of the ⁇ redundant operation trajectory:
  • the blank rate ⁇ blank of the trajectory as the ratio of the blank area generated by abnormal operations in the actual trajectory to the blank area after removing the redundant area of the operation, then:
  • S anomaly is the uncovered area caused by the abnormal operation.
  • S anomaly is the uncovered area caused by the abnormal operation.
  • Step 4.3.2 Based on the optimized core feature parameters, use machine learning methods to establish a positioning drift measurement model
  • f ( ⁇ , ⁇ ) is a function S S abnormality calculating the proportion of air in which the blank of the known formula ⁇ blank, it is also necessary to establish a blank ⁇ calculation model;
  • mapping functions K 1 ( ⁇ ) and K 2 ( ⁇ ) have been determined in the core characteristics of the sample data training, the S anomaly can also be accurately calculated to ensure the accuracy of the actual work area.
  • Step 5 Measurement and calculation of agricultural machinery operating area: Input the to-be-calculated operating track and operating area into the positioning drift measurement model to obtain an accurate agricultural machinery operating area; specifically including the following steps:
  • Step 5.1 Preprocess the positioning points into a trajectory sequence, calculate the values of different features and parameters, and generate the trajectory to be processed;
  • Step 5.2 Input the trajectory to be processed into the positioning drift measurement model to realize different classification of trajectories, and correct and compensate the drift area to obtain the actual area of agricultural machinery operation.
  • This method removes the blank area caused by abnormal conditions in the results, and the actual operation covers but the S drift that cannot be displayed correctly due to the positioning drift when collecting the data is still included in the calculation result, so that the method can accurately calculate
  • the actual work area is calculated, which overcomes the defect that the calculation accuracy of the agricultural machinery work area is not high when there are multiple types of operations in the agricultural machinery farming process, and the agricultural machinery work area is not high in accuracy when there are overlapping areas and non-operated areas, and the calculation of the actual work area is less subject to positioning Drift phenomenon influence.

Abstract

An agricultural machinery operation area calculation method based on a positioning drift calculation model. The method comprises the following steps: step 1, collecting and preprocessing moving trajectory points of agricultural machinery, involving: preprocessing sensor data, screening out agricultural machinery operation state information, and constructing a trajectory point time sequence conforming to an operation state; step 2, constructing a positioning trajectory and an operation trajectory, and establishing a polygon according to an operation width; step 3, calculating a coverage area of the operation trajectory, involving: calculating a trajectory area, coverage area and blank area of the operation trajectory from different angles; step 4, constructing a positioning drift calculation model by using extracted core features, and a training sample learning model parameters; and step 5, calculating an agricultural machinery operation area, involving: inputting an operation trajectory and operation area to be calculated into the positioning drift calculation model to obtain an accurate agricultural machinery operation area. The problem of a low calculation accuracy of an agricultural machinery operation area caused when there are multiple operation types in an agricultural machinery farming process and overlapping areas and non-operation areas are generated is solved.

Description

一种基于定位漂移测算模型的农机作业面积测算方法A Method for Measuring and Calculating the Working Area of Agricultural Machinery Based on the Positioning Drift Measuring Model 技术领域Technical field
本发明涉及一种基于定位漂移测算模型的农机作业面积测算方法。The invention relates to a method for measuring and calculating the working area of agricultural machinery based on a positioning drift measuring and calculating model.
背景技术Background technique
农机深松、耕整地作业对实现土地保墒，改善农田板结都有重要意义。在农机上安装地理信息系统、空间定位系统等数字技术导航装置，结合后台服务系统的模式，可以满足农机的深松、耕整地作业远程监控、管理需求，实现农业生产耕种收的精准作业。新型的农机管理和市场化服务模式日渐成熟，供求双方都要求农机作业服务提供精度高、可靠性高、实时便捷的农机作业面积计算结果。Deep loosening of agricultural machinery and land preparation operations are of great significance for realizing soil moisture conservation and improving farmland compaction. The installation of digital technology navigation devices such as geographic information systems and spatial positioning systems on agricultural machinery, combined with the back-end service system mode, can meet the needs of remote monitoring and management of agricultural machinery's deep loosening and land preparation operations, and realize accurate operations of agricultural production, cultivation, and harvesting. New agricultural machinery management and market-oriented service models are becoming more and more mature. Both supply and demand parties require agricultural machinery operation services to provide high-precision, high-reliability, real-time and convenient agricultural machinery operating area calculation results.
现有的农机作业面积测算方法主要包括距离法、缓冲区法、栅格法等。这些方法受限于以下两个方面：当低成本的GPS(Global Positioning System，全球定位系统)由于漂移、干扰等原因造成的农机运行轨迹点定位精度不高时，作业面积测算结果会有较大误差；当农机耕作过程中存在多种作业类型，产生重叠区域、未作业区域时，计算精度不高。Existing methods for measuring and calculating agricultural machinery work area mainly include distance method, buffer method, grid method and so on. These methods are limited by the following two aspects: When the low-cost GPS (Global Positioning System, global positioning system) due to drift, interference and other reasons, the positioning accuracy of agricultural machinery running track points is not high, the calculation result of the working area will be larger. Error: When there are multiple types of operations in the agricultural machinery farming process, overlapping areas and non-operating areas are generated, the calculation accuracy is not high.
因此，如何在安装低成本GPS的农机及复杂的作业环境中精确地测算农机作业面积已经成为一个急需解决的技术问题。Therefore, how to accurately measure the operating area of agricultural machinery in a complex operating environment with low-cost GPS installed has become a technical problem that needs to be resolved urgently.
发明内容Summary of the invention
本发明的目的在于提供一种基于定位漂移测算模型的农机作业面积测算方法，以解决现有技术中当农机耕作过程中存在多种作业类型，产生重叠区域、未作业区域时，农机作业面积计算精度不高的问题。The purpose of the present invention is to provide a method for measuring and calculating the working area of agricultural machinery based on a positioning drift measurement model to solve the problem of calculating the working area of agricultural machinery when there are multiple types of operations in the process of agricultural machinery farming in the prior art, and overlapping areas and non-operating areas are generated. The problem of low precision.
所述的一种基于定位漂移测算模型的农机作业面积测算方法，包括下列步骤：The described method for measuring and calculating the working area of agricultural machinery based on the positioning drift measurement model includes the following steps:
步骤1、农机运行轨迹点采集和预处理：将农机定位传感器和姿态传感器的数据进行预处理，过滤农机作业状态信息，构建符合作业状态的轨迹点时间序列；Step 1. Collecting and preprocessing the track points of agricultural machinery operation: preprocess the data of the agricultural machinery positioning sensor and the attitude sensor, filter the agricultural machinery operation status information, and construct the track point time series in line with the operation status;
步骤2、构造定位轨迹和作业轨迹：将符合作业状态的定位点依次连接，建立作业轨迹的折线路径；同时每两个点分别根据作业宽度生成四边形，多个四边 形进行逻辑运算建立一个多边形；Step 2. Construct positioning trajectory and operation trajectory: connect the locating points that conform to the operation status in turn to establish a polyline path of the operation trajectory; at the same time, every two points generate a quadrilateral according to the operation width, and multiple quadrilaterals perform logical operations to create a polygon;
步骤3、计算作业轨迹覆盖面积：从不同的角度对作业轨迹计算轨迹面积、覆盖面积和空白面积；Step 3. Calculate the coverage area of the job trajectory: calculate the trajectory area, coverage area and blank area of the job trajectory from different angles;
步骤4、构造定位漂移测算模型：分析不同轨迹覆盖类型，抽取轨迹覆盖面积测算的核心特征，利用提取的核心特征构建定位漂移测算模型，训练样本对模型参数进行学习；Step 4. Construct a positioning drift measurement model: analyze different trajectory coverage types, extract core features for trajectory coverage area measurement, use the extracted core features to build a positioning drift measurement model, and train samples to learn model parameters;
步骤5、农机作业面积测算：将待计算的作业轨迹和作业面积输入定位漂移测算模型，获得精确的农机作业面积。Step 5. Measurement and calculation of agricultural machinery operating area: Input the operating track and operating area to be calculated into the positioning drift measurement model to obtain an accurate agricultural machinery operating area.
优选的，所述步骤1具体包括下列步骤：Preferably, the step 1 specifically includes the following steps:
步骤1.1、接收安装在农机上的GPS传感器和姿态传感器每秒回传的数据；Step 1.1. Receive data returned every second from GPS sensors and attitude sensors installed on agricultural machinery;
步骤1.2、将间隔时间内农机定位点过近、过远的定位点过滤掉；Step 1.2. Filter out the positioning points of agricultural machinery that are too close or too far within the interval;
步骤1.3、将过滤后的且耕作深度达到国家深耕要求的轨迹点保留下来作为农机作业有效轨迹点放入集合，构建轨迹点时间序列。Step 1.3. Keep the filtered track points that have reached the national deep ploughing requirements as effective track points for agricultural machinery operations and put them into the collection to construct a time sequence of track points.
优选的，所述步骤2具体包括下列步骤：Preferably, the step 2 specifically includes the following steps:
步骤2.1、将步骤1中符合作业状态的定位点依次连接，建立作业轨迹的折线路径；Step 2.1. Connect the positioning points that meet the operating status in step 1 one by one to establish a broken line path of the operating track;
步骤2.2、以农机作业轨迹上时序相邻的2个轨迹点P1、P2的坐标计算他们之间的方位角；Step 2.2: Calculate the azimuth angle between the two adjacent trajectory points P1 and P2 on the agricultural machinery operation trajectory in time series;
步骤2.3、根据农机具与轨迹是垂直的，计算农机具在上述相邻2个轨迹点间轨迹上的方位角；Step 2.3. Calculate the azimuth angle of the agricultural machinery on the track between the aforementioned two adjacent track points according to the fact that the agricultural machinery is perpendicular to the track;
步骤2.4、根据农机具方位角、犁具长度R得出轨迹点P1、P2的延伸四个点L1、L2、L3、L4，进而构成该段轨迹作业覆盖面的四边形S1，以此类推，计算出轨迹路径上的所有四边形S1...Sn；Step 2.4. According to the azimuth angle of the agricultural machinery and the length of the plough R, the four points L1, L2, L3, and L4 of the trajectory points P1 and P2 are obtained, and then constitute the quadrilateral S1 of the coverage area of the trajectory, and so on, calculate All quadrilaterals S1...Sn on the trajectory path;
步骤2.5：将步骤2.4中得到的四边形S1...Sn进行逻辑运算建立一个总作业面积多边形。Step 2.5: Perform logical operations on the quadrilaterals S1...Sn obtained in step 2.4 to establish a total work area polygon.
优选的，所述步骤3具体包括下列步骤：Preferably, the step 3 specifically includes the following steps:
步骤3.1、计算作业的理论轨迹面积：将符合作业状态的定位点依次连接，分别计算各段轨迹的距离，依据作业宽度统计轨迹覆盖的面积；Step 3.1. Calculate the theoretical track area of the job: connect the positioning points that meet the job status in turn, calculate the distance of each track separately, and count the area covered by the track according to the width of the job;
步骤3.2、计算作业的理论覆盖面积：将两两作业点形成的小矩形逻辑运算 获得的多边形进行基于图形学计算，统计理论覆盖的面积、外部轮廓面积、内部空白面积。Step 3.2. Calculate the theoretical coverage area of the job: Calculate the polygons obtained from the logical operation of the small rectangle formed by the two job points based on graphics, and calculate the theoretical coverage area, external contour area, and internal blank area.
优选的，所述步骤4具体包括下列步骤：Preferably, the step 4 specifically includes the following steps:
步骤4.1、确定反映坐标定位漂移的要素，进而提取用于实现空白面积分析的核心特征，分别是局部重叠度、全局重叠度、全局覆盖度；Step 4.1. Determine the elements that reflect the drift of coordinate positioning, and then extract the core features used to realize the blank area analysis, which are local overlap, global overlap, and global coverage;
步骤4.2、优化核心特征的参数，根据样本数据训练核心特征的参数，将核心特征映射到特定的统一区间[0,1]；Step 4.2: Optimize the parameters of the core feature, train the parameters of the core feature according to the sample data, and map the core feature to a specific unified interval [0,1];
步骤4.3、构造定位漂移测算模型，利用优化后的核心特征的参数，运用机器学习方法建立定位漂移测算模型。Step 4.3: Construct a positioning drift measurement model, use the optimized parameters of the core features, and use machine learning methods to establish a positioning drift measurement model.
优选的，所述步骤4.1具体包括下列步骤：Preferably, the step 4.1 specifically includes the following steps:
步骤4.1.1、确定反映坐标定位漂移的要素：取作业轨迹在地图上的有限的包络区域面积表示为外面积S 外，在有限的包络区域内，轨迹点生成的四边形合并形成的有效覆盖面积表示为内面积S 内，S 外和S 内之间存在一个差值，即内部空白面积表示为S 空，轨迹长度结合犁具宽度R计算形成的轨迹覆盖的面积称之为轨迹面积S 轨； Effective work trajectory take a limited area of the envelope is represented as a map of area outside the outer S, in a limited region of the envelope, generating a locus of points forming a quadrangle combined: step 4.1.1, determines a coordinate positioning elements reflect drift The coverage area is expressed as the inner area S inside , there is a difference between S outside and S inside , that is, the internal blank area is expressed as S empty , and the track covered area calculated by combining the length of the track with the width of the plough R is called the track area S Rail
步骤4.1.2：抽取实现空白面积分析的核心特征局部重叠度、全局重叠度、全局覆盖度：Step 4.1.2: Extract the local overlap, global overlap, and global coverage of core features for blank area analysis:
局部重叠度定义为轨迹面积和内面积的比值，用表示用α表示，The degree of local overlap is defined as the ratio of the track area to the inner area, expressed by α,
Figure PCTCN2019110510-appb-000001
Figure PCTCN2019110510-appb-000001
局部重叠度α表达了轨迹内局部重叠的程度，轨迹越密集、重叠区域越集中，则局部重叠度越大。理想情况下，局部重叠度趋近于1，表示轨迹无漂移，作业正常。The degree of local overlap α expresses the degree of local overlap in the trajectory. The denser the trajectory and the more concentrated the overlap area, the greater the degree of local overlap. Ideally, the degree of local overlap is close to 1, which means that the trajectory has no drift and the operation is normal.
全局重叠度定义为轨迹面积和外面积的比值，用β表示The global overlap is defined as the ratio of the track area to the outer area, denoted by β
Figure PCTCN2019110510-appb-000002
Figure PCTCN2019110510-appb-000002
全局重叠度β表达了轨迹内部整体分散的程度，轨迹点分布越均匀，重叠区域越分散，则全局重叠度越大。理想情况下，完整作业轨迹，其全局重叠度趋近 于1。The global degree of overlap β expresses the degree of overall dispersion within the trajectory. The more uniform the distribution of trajectory points and the more scattered the overlap area, the greater the global degree of overlap. Ideally, for a complete operation trajectory, its global overlap is close to 1.
全局覆盖度定义为内面积和外面积的比值，用γ表示The global coverage is defined as the ratio of the inner area to the outer area, denoted by γ
Figure PCTCN2019110510-appb-000003
Figure PCTCN2019110510-appb-000003
全局覆盖度γ表达了实际作业面积覆盖作业区域的程度。作业越规范，轨迹越均匀，则全局覆盖度越大；作业异常，重叠集中，则全局覆盖度越小。The global coverage γ expresses the extent to which the actual work area covers the work area. The more standardized the job and the more uniform the trajectory, the greater the global coverage; if the job is abnormal and the overlap is concentrated, the global coverage is smaller.
优选的，所述步骤4.2具体包括下列步骤：Preferably, the step 4.2 specifically includes the following steps:
步骤4.2.1：优化核心特征的参数，选择sigmod函数将参数映射到[0,1]区间Step 4.2.1: Optimize the parameters of the core features, select the sigmod function to map the parameters to the interval [0,1]
Figure PCTCN2019110510-appb-000004
Figure PCTCN2019110510-appb-000004
步骤4.2.2：根据样本数据训练核心特征的参数x和y，确定具体映射函数K 1(α)和K 2(β)。 Step 4.2.2: Train the parameters x and y of the core feature according to the sample data, and determine the specific mapping functions K 1 (α) and K 2 (β).
优选的，所述步骤4.3具体包括下列步骤：Preferably, the step 4.3 specifically includes the following steps:
所述步骤4.3具体包括下列步骤：The step 4.3 specifically includes the following steps:
步骤4.3.1：构造定位漂移测算模型，利用优化后的核心特征的参数进行形式化表示；Step 4.3.1: Construct a positioning drift measurement model, and use the optimized core feature parameters for formal expression;
特别地，对S 空空的结构进行分析，S 空＝S 漂移+S 异常中S 异常可以具体分为两种类型，一种是冗余部分，特别当全局重叠度β小于1的情况下，后续有其他的轨迹进行填充；另一种是未作业空白部分，由于作业过程中未覆盖产生空白区域，面积表示为S 未作业； In particular, the empty space S for analysis of the structure, space S = S + S drift anomalies S abnormalities may particularly be divided into two types, one is redundant portion, especially in the case where the overall degree of overlap less than 1 β, There will be other tracks to fill in the follow-up; the other is the unworked blank part, because the blank area is not covered during the operation, the area is expressed as S unworked ;
方法定义轨迹的冗余率为θ 冗余作业轨迹内部面积中冗余的作业面积占据的比例： The method defines the redundancy rate of the trajectory as the proportion of the redundant operation area in the internal area of the θ redundant operation trajectory:
Figure PCTCN2019110510-appb-000005
Figure PCTCN2019110510-appb-000005
以S 冗余表示对应的作业冗余面积，β为全局重叠度，则有： S redundancy represents the corresponding job redundancy area, β is the global overlap degree, then:
S 冗余＝θ 冗余×S 空 S redundancy = θ redundancy × S empty
同时定义轨迹的空白率θ 空白为实际轨迹内异常作业产生空白区域的面积占据除去作业冗余面积后的空白面积的比例，则有： At the same time, define the blank rate θ blank of the trajectory as the ratio of the blank area generated by abnormal operations in the actual trajectory to the blank area after removing the redundant area of the operation, then:
S 未作业＝θ 空白×(S 空-S 冗余) S not working = θ blank × (S blank- S redundancy )
S 异常是异常作业产生的作业未覆盖面积，综上分析可知，应表示如下： S anomaly is the uncovered area caused by the abnormal operation. To sum up, it can be seen from the above analysis that it should be expressed as follows:
S 异常＝S 冗余+S 未作业 S abnormal = S redundancy + S not working
步骤4.3.2：基于优化后的核心特征的参数，运用机器学习方法建立定位漂移测算模型；Step 4.3.2: Based on the optimized core feature parameters, use machine learning methods to establish a positioning drift measurement model;
基于试验样本的轨迹，我们将实际作业面积用S 正表示，存在关系： Based on the trajectory of the test sample, we denote the actual work area as positive S, and there is a relationship:
S 正＝S 外-S 空×f(α，β) S positive = S outside- S empty × f(α, β)
其中f(α，β)为S 异常在S 空中占比的计算函数，其中空白率θ 空白的算式已知，因此还需要建立θ 空白的计算模型； Wherein f (α, β) is a function S S abnormality calculating the proportion of air in which the blank of the known formula θ blank, it is also necessary to establish a blank θ calculation model;
经过试验发现，当β保持不变，α越大，θ 空白的值越大；当α保持不变，β越大值，θ 空白越小；依据上述结论，通过sigmod函数构建空白率θ 空白的计算模型： After testing found that when beta] remain constant, [alpha], the greater the blank value [theta]; [alpha] remains unchanged when, the larger the value beta], [theta] is smaller blank; According to the above conclusion, by constructing sigmod function of [theta] Blank Blank Calculation model:
Figure PCTCN2019110510-appb-000006
Figure PCTCN2019110510-appb-000006
运用机器学习方法，结合具体的样本轨迹建立定位漂移测算模型的明确表示；基于试验样本的轨迹，我们将实际作业面积用S 正表示，存在关系： Using machine learning methods, combined with the specific sample trajectory to establish a clear representation of the positioning drift measurement model; based on the trajectory of the test sample, we use the actual working area as positive S, and there is a relationship:
S 正＝S 外-S 异常 S positive = S outside- S abnormal
即S 正＝S 外-θ 冗余×S 空-θ 空白×(1-θ 冗余)×S 空 That is, S positive = S outside- θ redundancy × S empty- θ blank × (1-θ redundancy ) × S empty
由此得到定位漂移测算模型。Thus, a positioning drift measurement model is obtained.
优选的，所述步骤5具体包括下列步骤：Preferably, the step 5 specifically includes the following steps:
步骤5.1：将定位点预处理成轨迹序列，计算不同特征和参数的取值，生成待处理的轨迹；Step 5.1: Preprocess the positioning points into a trajectory sequence, calculate the values of different features and parameters, and generate the trajectory to be processed;
步骤5.2：将待处理的轨迹输入定位漂移测算模型，实现轨迹的不同分类，并对漂移区域进行纠正和补偿，得到农机作业的实际面积。Step 5.2: Input the trajectory to be processed into the positioning drift measurement model to realize different classification of trajectories, and correct and compensate the drift area to obtain the actual area of agricultural machinery operation.
本发明具有如下优点：The present invention has the following advantages:
本发明将采集的农机运行轨迹点经过预处理后，利用逻辑运算构造作业面积多边形，通过分析不同轨迹覆盖类型，抽取轨迹覆盖面积测算的核心特征来构建定位漂移测算模型，最终获得精确的农机作业面积。利用机器学习方法，能将测算模型不断修正提高精确性。作业异常的两种情况，一种称为冗余，为后续全局重叠度小于1，后续由其他轨迹进行填充的情况，另一种为作业异常造成作业过程中未覆盖的空白区域；作业异常产生的空白区域和定位漂移显示出的空白区域共同组成了测算时出现的空白区域。针对作业异常的两种情况，本方法都提供了用于计算分析的冗余率和空白率，并通过机器学习来不断修正不易计算的空白率，从而保证对作业异常产生的误差部分能精确计算，从而解决了由于低成本GPS定位漂移及农机操作偏移等导致的农机作业面积测算误差问题，提高了农机作业面积的测量精度，为农机精准作业及农机手作业补贴提供了依据。After preprocessing the collected agricultural machinery running track points, the present invention uses logical operations to construct a work area polygon, analyzes different track coverage types, extracts core features of track coverage area calculations to construct a positioning drift measurement model, and finally obtains accurate agricultural machinery operations area. Using machine learning methods, the measurement model can be continuously revised to improve accuracy. There are two cases of abnormal operation, one is called redundancy, which is the case where the subsequent global overlap is less than 1, and the subsequent filling is performed by other tracks; the other is the blank area that is not covered in the operation process caused by the abnormal operation; the abnormal operation occurs The blank area and the blank area displayed by the positioning drift together constitute the blank area that appears during the measurement. In view of the two situations of abnormal operation, this method provides redundancy rate and blank rate for calculation and analysis, and continuously corrects the blank rate that is not easy to calculate through machine learning, so as to ensure that the error part caused by abnormal operation can be accurately calculated , Thereby solving the problem of measurement error of agricultural machinery working area caused by low-cost GPS positioning drift and agricultural machinery operation offset, etc., improving the measurement accuracy of agricultural machinery working area, and providing a basis for agricultural machinery precision operation and agricultural machinery manual operation subsidies.
附图说明Description of the drawings
图1为本发明一种基于定位漂移测算模型的农机作业面积测算方法的流程图。Fig. 1 is a flowchart of a method for measuring and calculating the working area of agricultural machinery based on a positioning drift measuring model of the present invention.
具体实施方式Detailed ways
下面对照附图，通过对实施例的描述，对本发明具体实施方式作进一步详细的说明，以帮助本领域的技术人员对本发明的发明构思、技术方案有更完整、准确和深入的理解。With reference to the accompanying drawings, the specific implementation of the present invention will be further described in detail through the description of the embodiments to help those skilled in the art have a more complete, accurate and in-depth understanding of the inventive concept and technical solution of the present invention.
如图1所示，本发明提供了一种基于定位漂移测算模型的农机作业面积测算方法，包括下列步骤：As shown in Fig. 1, the present invention provides a method for measuring and calculating the working area of agricultural machinery based on a positioning drift measurement model, which includes the following steps:
步骤1、农机运行轨迹点采集和预处理：将农机定位传感器和姿态传感器的数据进行预处理，过滤农机作业状态信息，构建符合作业状态的轨迹点时间序列；其具体步骤为：Step 1. Collecting and preprocessing the track points of agricultural machinery operation: preprocess the data of the agricultural machinery positioning sensor and the attitude sensor, filter the agricultural machinery operation status information, and construct the track point time series in line with the operation status; the specific steps are:
步骤1.1、接收安装在农机上的GPS传感器和姿态传感器每秒回传的数据；Step 1.1. Receive data returned every second from GPS sensors and attitude sensors installed on agricultural machinery;
步骤1.2、将间隔时间内农机定位点过近、过远的定位点过滤掉；Step 1.2. Filter out the positioning points of agricultural machinery that are too close or too far within the interval;
步骤1.3、将过滤后的且耕作深度达到国家深耕要求的轨迹点保留下来作为农机作业有效轨迹点放入集合，构建轨迹点时间序列。Step 1.3. Keep the filtered track points that have reached the national deep ploughing requirements as effective track points for agricultural machinery operations and put them into the collection to construct a time sequence of track points.
步骤2、构造定位轨迹和作业轨迹：将符合作业状态的定位点依次连接，建立作业轨迹的折线路径；同时每两个点分别根据作业宽度生成四边形，多个四边形进行逻辑运算建立一个多边形；其具体步骤为：Step 2. Construct positioning trajectory and operation trajectory: connect the positioning points that conform to the operation status in turn to establish a polyline path of the operation trajectory; at the same time, every two points generate a quadrilateral according to the operation width, and multiple quadrilaterals perform logical operations to create a polygon; The specific steps are:
步骤2.1、将步骤1中符合作业状态的定位点依次连接，建立作业轨迹的折线路径；Step 2.1. Connect the positioning points that meet the operating status in step 1 one by one to establish a broken line path of the operating track;
步骤2.2、以农机作业轨迹上时序相邻的2个轨迹点P1、P2的坐标计算他们之间的方位角；Step 2.2: Calculate the azimuth angle between the two adjacent trajectory points P1 and P2 on the agricultural machinery operation trajectory in time series;
步骤2.3、根据农机具与轨迹是垂直的，计算农机具在上述相邻2个轨迹点间轨迹上的方位角；Step 2.3. Calculate the azimuth angle of the agricultural machinery on the track between the aforementioned two adjacent track points according to the fact that the agricultural machinery is perpendicular to the track;
步骤2.4、根据农机具方位角、犁具长度R得出轨迹点P1、P2的延伸四个点L1、L2、L3、L4，进而构成该段轨迹作业覆盖面的四边形S1，以此类推，计算出轨迹路径上的所有四边形S1...Sn；Step 2.4. According to the azimuth angle of the agricultural machinery and the length of the plough R, the four points L1, L2, L3, and L4 of the trajectory points P1 and P2 are obtained, and then constitute the quadrilateral S1 of the coverage area of the trajectory, and so on, calculate All quadrilaterals S1...Sn on the trajectory path;
步骤2.5：将步骤2.4中得到的四边形S1...Sn进行逻辑运算建立一个总作业面积多边形。总作业面积多边形为后续面积计算提供依据。Step 2.5: Perform logical operations on the quadrilaterals S1...Sn obtained in step 2.4 to establish a total work area polygon. The total work area polygon provides the basis for subsequent area calculations.
步骤3、计算作业轨迹覆盖面积：从不同的角度对作业轨迹计算轨迹面积、覆盖面积和空白面积；具体包括下列步骤：Step 3. Calculate the coverage area of the job trajectory: calculate the trajectory area, coverage area and blank area of the job trajectory from different angles; specifically include the following steps:
步骤3.1、计算作业的理论轨迹面积：将符合作业状态的定位点依次连接，分别计算各段轨迹的距离，依据作业宽度统计轨迹覆盖的面积，该面积包括各段轨迹之间重叠的部分；Step 3.1. Calculate the theoretical track area of the job: connect the positioning points that meet the job status in turn, calculate the distance of each track separately, and count the area covered by the track based on the job width, which includes the overlapped part of each track;
步骤3.2、计算作业的理论覆盖面积：将两两作业点形成的小矩形逻辑运算获得的多边形进行基于图形学计算，统计理论覆盖的面积、外部轮廓面积、内部空白面积。外部轮廓面积为组成的多边形的外部轮廓围出的面积，空白面积为多边形形成的位于外部轮廓内的空白部分的面积，理论覆盖面积为多边形自身的覆盖面积。上述面积提供给构建的定位漂移测算模型以求得实际作业面积。Step 3.2. Calculate the theoretical coverage area of the job: perform graphics-based calculations on the polygons obtained by the logical operation of the small rectangles formed by the two job points, and calculate the theoretical coverage area, external contour area, and internal blank area. The outer contour area is the area enclosed by the outer contour of the formed polygon, the blank area is the area of the blank part in the outer contour formed by the polygon, and the theoretical coverage area is the coverage area of the polygon itself. The above-mentioned area is provided to the constructed positioning drift measurement model to obtain the actual operating area.
步骤4、构造定位漂移测算模型：分析不同轨迹覆盖类型，抽取轨迹覆盖面积测算的核心特征，利用提取的核心特征构建定位漂移测算模型，训练样本对模 型参数进行学习；具体包括下列步骤：Step 4. Construct a positioning drift measurement model: analyze different trajectory coverage types, extract core features for trajectory coverage area measurement, use the extracted core features to construct a positioning drift measurement model, and train samples to learn model parameters; specifically, it includes the following steps:
步骤4.1、确定反映坐标定位漂移的要素，进而提取用于实现空白面积分析的核心特征，分别是局部重叠度、全局重叠度、全局覆盖度；该步骤又具体包括下列步骤：Step 4.1. Determine the elements that reflect the drift of coordinate positioning, and then extract the core features used to realize the blank area analysis, which are local overlap, global overlap, and global coverage; this step specifically includes the following steps:
步骤4.1.1、确定反映坐标定位漂移的要素：取作业轨迹在地图上的有限的包络区域面积表示为外面积S 外，在有限的包络区域内，轨迹点生成的四边形合并形成的有效覆盖面积表示为内面积S 内，S 外和S 内之间存在一个差值，即内部空白面积表示为S 空，轨迹长度结合犁具宽度R计算形成的轨迹覆盖的面积称之为轨迹面积S 轨； Effective work trajectory take a limited area of the envelope is represented as a map of area outside the outer S, in a limited region of the envelope, generating a locus of points forming a quadrangle combined: step 4.1.1, determines a coordinate positioning elements reflect drift The coverage area is expressed as the inner area S inside , there is a difference between S outside and S inside , that is, the internal blank area is expressed as S empty , and the track covered area calculated by combining the length of the track with the width of the plough R is called the track area S Rail
步骤4.1.2：抽取实现空白面积分析的核心特征局部重叠度、全局重叠度、全局覆盖度：Step 4.1.2: Extract the local overlap, global overlap, and global coverage of core features for blank area analysis:
局部重叠度定义为轨迹面积和内面积的比值，用表示用α表示，The degree of local overlap is defined as the ratio of the track area to the inner area, expressed by α,
Figure PCTCN2019110510-appb-000007
Figure PCTCN2019110510-appb-000007
局部重叠度α表达了轨迹内局部重叠的程度，轨迹越密集、重叠区域越集中，则局部重叠度越大。理想情况下，局部重叠度趋近于1，表示轨迹无漂移，作业正常。The degree of local overlap α expresses the degree of local overlap in the trajectory. The denser the trajectory and the more concentrated the overlap area, the greater the degree of local overlap. Ideally, the degree of local overlap is close to 1, which means that the trajectory has no drift and the operation is normal.
全局重叠度定义为轨迹面积和外面积的比值，用β表示The global overlap is defined as the ratio of the track area to the outer area, denoted by β
Figure PCTCN2019110510-appb-000008
Figure PCTCN2019110510-appb-000008
全局重叠度β表达了轨迹内部整体分散的程度，轨迹点分布越均匀，重叠区域越分散，则全局重叠度越大。理想情况下，完整作业轨迹，其全局重叠度趋近于1。The global degree of overlap β expresses the degree of overall dispersion within the trajectory. The more uniform the distribution of trajectory points and the more scattered the overlap area, the greater the global degree of overlap. Ideally, for a complete operation trajectory, its global overlap is close to 1.
全局覆盖度定义为内面积和外面积的比值，用γ表示The global coverage is defined as the ratio of the inner area to the outer area, denoted by γ
Figure PCTCN2019110510-appb-000009
Figure PCTCN2019110510-appb-000009
全局覆盖度γ表达了实际作业面积覆盖作业区域的程度。作业越规范，轨迹 越均匀，则全局覆盖度越大；作业异常，重叠集中，则全局覆盖度越小。上述核心特征均可由步骤3计算出的面积数据获得，是求得实际作业面积所需的重要参数，根据上述核心特征的差异决定了不同轨迹覆盖类型。The global coverage γ expresses the extent to which the actual work area covers the work area. The more standardized the job and the more uniform the trajectory, the greater the global coverage; if the job is abnormal and the overlap is concentrated, the global coverage is smaller. The above-mentioned core features can be obtained from the area data calculated in step 3, which are important parameters required to obtain the actual working area, and different trajectory coverage types are determined according to the difference of the above-mentioned core features.
步骤4.2、优化核心特征的参数，根据样本数据训练核心特征的参数，将核心特征映射到特定的统一区间[0,1]；该步骤又具体包括下列步骤：Step 4.2: Optimize the parameters of the core feature, train the parameters of the core feature according to the sample data, and map the core feature to a specific uniform interval [0,1]; this step specifically includes the following steps:
步骤4.2.1：优化核心特征的参数，选择sigmod函数将参数映射到[0,1]区间Step 4.2.1: Optimize the parameters of the core features, select the sigmod function to map the parameters to the interval [0,1]
Figure PCTCN2019110510-appb-000010
Figure PCTCN2019110510-appb-000010
步骤4.2.2：根据样本数据训练核心特征的参数x和y，确定具体映射函数K 1(α)和K 2(β)。映射函数对后续异常情况的面积分析非常重要。 Step 4.2.2: Train the parameters x and y of the core feature according to the sample data, and determine the specific mapping functions K 1 (α) and K 2 (β). The mapping function is very important for subsequent area analysis of abnormal conditions.
步骤4.3、构造定位漂移测算模型，利用优化后的核心特征的参数，运用机器学习方法建立定位漂移测算模型，该步骤又具体包括下列步骤：Step 4.3. Construct a positioning drift measurement model, use the optimized parameters of the core features, and use machine learning methods to establish a positioning drift measurement model. This step specifically includes the following steps:
步骤4.3.1：构造定位漂移测算模型，利用优化后的核心特征的参数进行形式化表示；Step 4.3.1: Construct a positioning drift measurement model, and use the optimized core feature parameters for formal expression;
特别地，对S 空空的结构进行分析，S 空＝S 漂移+S 异常中S 异常可以具体分为两种类型，一种是冗余部分，特别当全局重叠度β小于1的情况下，后续有其他的轨迹进行填充；另一种是未作业空白部分，由于作业过程中未覆盖产生空白区域，面积表示为S 未作业； In particular, the empty space S for analysis of the structure, space S = S + S drift anomalies S abnormalities may particularly be divided into two types, one is redundant portion, especially in the case where the overall degree of overlap less than 1 β, There will be other tracks to fill in the follow-up; the other is the unworked blank part, because the blank area is not covered during the operation, the area is expressed as S unworked ;
方法定义轨迹的冗余率为θ 冗余作业轨迹内部面积中冗余的作业面积占据的比例： The method defines the redundancy rate of the trajectory as the proportion of the redundant operation area in the internal area of the θ redundant operation trajectory:
Figure PCTCN2019110510-appb-000011
Figure PCTCN2019110510-appb-000011
以S 冗余表示对应的作业冗余面积，β为全局重叠度，则有： S redundancy represents the corresponding job redundancy area, β is the global overlap degree, then:
S 冗余＝θ 冗余×S 空 S redundancy = θ redundancy × S empty
同时定义轨迹的空白率θ 空白为实际轨迹内异常作业产生空白区域的面积占 据除去作业冗余面积后的空白面积的比例，则有： At the same time, define the blank rate θ blank of the trajectory as the ratio of the blank area generated by abnormal operations in the actual trajectory to the blank area after removing the redundant area of the operation, then:
S 未作业＝θ 空白×(S 空-S 冗余) S not working = θ blank × (S blank- S redundancy )
S 异常是异常作业产生的作业未覆盖面积，综上分析可知，应表示如下： S anomaly is the uncovered area caused by the abnormal operation. In summary, it can be seen that it should be expressed as follows:
S 异常＝S 冗余+S 未作业 S abnormal = S redundancy + S not working
步骤4.3.2：基于优化后的核心特征的参数，运用机器学习方法建立定位漂移测算模型；Step 4.3.2: Based on the optimized core feature parameters, use machine learning methods to establish a positioning drift measurement model;
基于试验样本的轨迹，我们将实际作业面积用S 正表示，存在关系： Based on the trajectory of the test sample, we denote the actual work area as positive S, and there is a relationship:
S 正＝S 外-S 空×f(α，β) S positive = S outside- S empty × f(α, β)
其中f(α，β)为S 异常在S 空中占比的计算函数，其中空白率θ 空白的算式已知，因此还需要建立θ 空白的计算模型； Wherein f (α, β) is a function S S abnormality calculating the proportion of air in which the blank of the known formula θ blank, it is also necessary to establish a blank θ calculation model;
经过试验发现，当β保持不变，α越大，θ 空白的值越大；当α保持不变，β越大值，θ 空白越小；依据上述结论，通过sigmod函数构建空白率θ 空白的计算模型： After testing found that when beta] remain constant, [alpha], the greater the blank value [theta]; [alpha] remains unchanged when, the larger the value beta], [theta] is smaller blank; According to the above conclusion, by constructing sigmod function of [theta] Blank Blank Calculation model:
Figure PCTCN2019110510-appb-000012
Figure PCTCN2019110510-appb-000012
运用机器学习方法，结合具体的样本轨迹建立定位漂移测算模型的明确表示；基于试验样本的轨迹，我们将实际作业面积用S 正表示，存在关系： Using machine learning methods, combined with the specific sample trajectory to establish a clear representation of the positioning drift measurement model; based on the trajectory of the test sample, we use the actual working area as positive S, and there is a relationship:
S 正＝S 外-S 异常 S positive = S outside- S abnormal
即S 正＝S 外-θ 冗余×S 空-θ 空白×(1-θ 冗余)×S 空 That is, S positive = S outside- θ redundancy × S empty- θ blank × (1-θ redundancy ) × S empty
由此得到定位漂移测算模型。由于映射函数K 1(α)和K 2(β)已经在样本数据训练核心特征得以确定，因此S 异常也能准确计算出来，保证实际作业面积的准确性。 Thus, a positioning drift measurement model is obtained. Since the mapping functions K 1 (α) and K 2 (β) have been determined in the core characteristics of the sample data training, the S anomaly can also be accurately calculated to ensure the accuracy of the actual work area.
步骤5、农机作业面积测算：将待计算的作业轨迹和作业面积输入定位漂移测算模型，获得精确的农机作业面积；具体包括下列步骤：Step 5. Measurement and calculation of agricultural machinery operating area: Input the to-be-calculated operating track and operating area into the positioning drift measurement model to obtain an accurate agricultural machinery operating area; specifically including the following steps:
步骤5.1：将定位点预处理成轨迹序列，计算不同特征和参数的取值，生成待处理的轨迹；Step 5.1: Preprocess the positioning points into a trajectory sequence, calculate the values of different features and parameters, and generate the trajectory to be processed;
步骤5.2：将待处理的轨迹输入定位漂移测算模型，实现轨迹的不同分类，并对漂移区域进行纠正和补偿，得到农机作业的实际面积。Step 5.2: Input the trajectory to be processed into the positioning drift measurement model to realize different classification of trajectories, and correct and compensate the drift area to obtain the actual area of agricultural machinery operation.
本方法在结果中去除了异常情况造成的空白部分面积，而实际作业覆盖到但采集数据时因定位漂移造成未能正确显示的S 漂移则仍然包括在计算结果内，从而令本方法能精确计算出实际作业面积，克服了现有技术因农机耕作过程中存在多种作业类型，产生重叠区域、未作业区域时，农机作业面积计算精度不高的缺陷，并且实际作业面积的计算较少受定位漂移现象影响。 This method removes the blank area caused by abnormal conditions in the results, and the actual operation covers but the S drift that cannot be displayed correctly due to the positioning drift when collecting the data is still included in the calculation result, so that the method can accurately calculate The actual work area is calculated, which overcomes the defect that the calculation accuracy of the agricultural machinery work area is not high when there are multiple types of operations in the agricultural machinery farming process, and the agricultural machinery work area is not high in accuracy when there are overlapping areas and non-operated areas, and the calculation of the actual work area is less subject to positioning Drift phenomenon influence.
上面结合附图对本发明进行了示例性描述，显然本发明具体实现并不受上述方式的限制，只要采用了本发明的发明构思和技术方案进行的各种非实质性的改进，或未经改进将本发明构思和技术方案直接应用于其它场合的，均在本发明保护范围之内。The present invention is exemplarily described above with reference to the accompanying drawings. It is obvious that the specific implementation of the present invention is not limited by the above-mentioned manners, as long as various insubstantial improvements made by the inventive concept and technical solutions of the present invention are adopted, or no improvements are made. Application of the concept and technical solution of the present invention to other occasions directly falls within the protection scope of the present invention.

Claims (9)

  1. 一种基于定位漂移测算模型的农机作业面积测算方法，其特征在于：包括下列步骤：A method for measuring and calculating the working area of agricultural machinery based on a positioning drift measuring model, which is characterized in that it includes the following steps:
    步骤1、农机运行轨迹点采集和预处理：将农机定位传感器和姿态传感器的数据进行预处理，过滤农机作业状态信息，构建符合作业状态的轨迹点时间序列；Step 1. Collecting and preprocessing the track points of agricultural machinery operation: preprocess the data of the agricultural machinery positioning sensor and the attitude sensor, filter the agricultural machinery operation status information, and construct the track point time series in line with the operation status;
    步骤2、构造定位轨迹和作业轨迹：将符合作业状态的定位点依次连接，建立作业轨迹的折线路径；同时每两个点分别根据作业宽度生成四边形，多个四边形进行逻辑运算建立一个多边形；Step 2. Construct positioning trajectory and operation trajectory: connect the positioning points that conform to the operation status in turn to establish a polyline path of the operation trajectory; at the same time, every two points generate a quadrilateral according to the operation width, and multiple quadrilaterals perform logical operations to create a polygon;
    步骤3、计算作业轨迹覆盖面积：从不同的角度对作业轨迹计算轨迹面积、覆盖面积和空白面积；Step 3. Calculate the coverage area of the job trajectory: calculate the trajectory area, coverage area and blank area of the job trajectory from different angles;
    步骤4、构造定位漂移测算模型：分析不同轨迹覆盖类型，抽取轨迹覆盖面积测算的核心特征，利用提取的核心特征构建定位漂移测算模型，训练样本对模型参数进行学习；Step 4. Construct a positioning drift measurement model: analyze different trajectory coverage types, extract core features for trajectory coverage area measurement, use the extracted core features to build a positioning drift measurement model, and train samples to learn model parameters;
    步骤5、农机作业面积测算：将待计算的作业轨迹和作业面积输入定位漂移测算模型，获得精确的农机作业面积。Step 5. Measurement and calculation of agricultural machinery operating area: Input the operating track and operating area to be calculated into the positioning drift measurement model to obtain an accurate agricultural machinery operating area.
  2. 根据权利要求1所述的一种基于定位漂移测算模型的农机作业面积测算方法，其特征在于：所述步骤1具体包括下列步骤：The method for measuring and calculating the working area of agricultural machinery based on the positioning drift measurement model according to claim 1, wherein said step 1 specifically includes the following steps:
    步骤1.1、接收安装在农机上的GPS传感器和姿态传感器每秒回传的数据；Step 1.1. Receive data returned every second from GPS sensors and attitude sensors installed on agricultural machinery;
    步骤1.2、将间隔时间内农机定位点过近、过远的定位点过滤掉；Step 1.2. Filter out the positioning points of agricultural machinery that are too close or too far within the interval;
    步骤1.3、将过滤后的且耕作深度达到国家深耕要求的轨迹点保留下来作为农机作业有效轨迹点放入集合，构建轨迹点时间序列。Step 1.3. Keep the filtered track points that have reached the national deep ploughing requirements as effective track points for agricultural machinery operations and put them into the collection to construct a time sequence of track points.
  3. 根据权利要求1所述的一种基于定位漂移测算模型的农机作业面积测算方法，其特征在于：所述步骤2具体包括下列步骤：The method for measuring and calculating the working area of agricultural machinery based on the positioning drift measuring model according to claim 1, wherein said step 2 specifically includes the following steps:
    步骤2.1、将步骤1中符合作业状态的定位点依次连接，建立作业轨迹的折线路径；Step 2.1. Connect the positioning points that meet the operating status in step 1 one by one to establish a broken line path of the operating track;
    步骤2.2、以农机作业轨迹上时序相邻的2个轨迹点P1、P2的坐标计算他们之间的方位角；Step 2.2: Calculate the azimuth angle between the two adjacent trajectory points P1 and P2 on the agricultural machinery operation trajectory in time series;
    步骤2.3、根据农机具与轨迹是垂直的，计算农机具在上述相邻2个轨迹点间轨迹上的方位角；Step 2.3. Calculate the azimuth angle of the agricultural machinery on the track between the aforementioned two adjacent track points according to the fact that the agricultural machinery is perpendicular to the track;
    步骤2.4、根据农机具方位角、犁具长度R得出轨迹点P1、P2的延伸四个 点L1、L2、L3、L4，进而构成该段轨迹作业覆盖面的四边形S1，以此类推，计算出轨迹路径上的所有四边形S1...Sn；Step 2.4. According to the azimuth angle of the agricultural machinery and the length of the plough R, the four points L1, L2, L3, and L4 of the trajectory points P1 and P2 are obtained, and then constitute the quadrilateral S1 of the coverage area of the trajectory, and so on, calculate All quadrilaterals S1...Sn on the trajectory path;
    步骤2.5：将步骤2.4中得到的四边形S1...Sn进行逻辑运算建立一个总作业面积多边形。Step 2.5: Perform logical operations on the quadrilaterals S1...Sn obtained in step 2.4 to establish a total work area polygon.
  4. 根据权利要求1所述的一种基于定位漂移测算模型的农机作业面积测算方法，其特征在于：所述步骤3具体包括下列步骤：The method for measuring and calculating the working area of agricultural machinery based on the positioning drift measurement model according to claim 1, wherein said step 3 specifically includes the following steps:
    步骤3.1、计算作业的理论轨迹面积：将符合作业状态的定位点依次连接，分别计算各段轨迹的距离，依据作业宽度统计轨迹覆盖的面积；Step 3.1. Calculate the theoretical track area of the job: connect the positioning points that meet the job status in turn, calculate the distance of each track separately, and count the area covered by the track according to the width of the job;
    步骤3.2、计算作业的理论覆盖面积：将两两作业点形成的小矩形逻辑运算获得的多边形进行基于图形学计算，统计理论覆盖的面积、外部轮廓面积、内部空白面积。Step 3.2. Calculate the theoretical coverage area of the job: perform graphics-based calculations on the polygons obtained by the logical operation of the small rectangles formed by the two job points, and calculate the theoretical coverage area, external contour area, and internal blank area.
  5. 根据权利要求1所述的一种基于定位漂移测算模型的农机作业面积测算方法，其特征在于：所述步骤4具体包括下列步骤：The method for measuring and calculating the working area of agricultural machinery based on the positioning drift measurement model according to claim 1, wherein said step 4 specifically includes the following steps:
    步骤4.1、确定反映坐标定位漂移的要素，进而提取用于实现空白面积分析的核心特征，分别是局部重叠度、全局重叠度、全局覆盖度；Step 4.1. Determine the elements that reflect the drift of coordinate positioning, and then extract the core features used to realize the blank area analysis, which are local overlap, global overlap, and global coverage;
    步骤4.2、优化核心特征的参数，根据样本数据训练核心特征的参数，将核心特征映射到特定的统一区间[0,1]；Step 4.2: Optimize the parameters of the core feature, train the parameters of the core feature according to the sample data, and map the core feature to a specific unified interval [0,1];
    步骤4.3、构造定位漂移测算模型，利用优化后的核心特征的参数，运用机器学习方法建立定位漂移测算模型。Step 4.3: Construct a positioning drift measurement model, use the optimized parameters of the core features, and use machine learning methods to establish a positioning drift measurement model.
  6. 根据权利要求5所述的一种基于定位漂移测算模型的农机作业面积测算方法，其特征在于：所述步骤4.1具体包括下列步骤：The method for measuring and calculating the working area of agricultural machinery based on the positioning drift measurement model according to claim 5, wherein said step 4.1 specifically includes the following steps:
    步骤4.1.1、确定反映坐标定位漂移的要素：取作业轨迹在地图上的有限的包络区域面积表示为外面积S 外，在有限的包络区域内，轨迹点生成的四边形合并形成的有效覆盖面积表示为内面积S 内，S 外和S 内之间存在一个差值，即内部空白面积表示为S 空，轨迹长度结合犁具宽度R计算形成的轨迹覆盖的面积称之为轨迹面积S 轨； Effective work trajectory take a limited area of the envelope is represented as a map of area outside the outer S, in a limited region of the envelope, generating a locus of points forming a quadrangle combined: step 4.1.1, determines a coordinate positioning elements reflect drift The coverage area is expressed as the inner area S inside , there is a difference between S outside and S inside , that is, the internal blank area is expressed as S empty , and the track covered area calculated by combining the length of the track with the width of the plough R is called the track area S Rail
    步骤4.1.2：抽取实现空白面积分析的核心特征局部重叠度、全局重叠度、全局覆盖度：Step 4.1.2: Extract the local overlap, global overlap, and global coverage of core features for blank area analysis:
    局部重叠度定义为轨迹面积和内面积的比值，用表示用α表示，The degree of local overlap is defined as the ratio of the track area to the inner area, expressed by α,
    Figure PCTCN2019110510-appb-100001
    Figure PCTCN2019110510-appb-100001
    局部重叠度α表达了轨迹内局部重叠的程度，轨迹越密集、重叠区域越集中，则局部重叠度越大；理想情况下，局部重叠度趋近于1，表示轨迹无漂移，作业正常；The degree of local overlap α expresses the degree of local overlap in the trajectory. The denser the trajectory and the more concentrated the overlap area, the greater the degree of local overlap; ideally, the local overlap degree approaches 1, indicating that the trajectory has no drift and the operation is normal;
    全局重叠度定义为轨迹面积和外面积的比值，用β表示The global overlap is defined as the ratio of the track area to the outer area, denoted by β
    Figure PCTCN2019110510-appb-100002
    Figure PCTCN2019110510-appb-100002
    全局重叠度β表达了轨迹内部整体分散的程度，轨迹点分布越均匀，重叠区域越分散，则全局重叠度越大；理想情况下，完整作业轨迹，其全局重叠度趋近于1；The global overlap degree β expresses the overall degree of dispersion within the trajectory. The more uniform the distribution of trajectory points and the more scattered the overlap area, the greater the global overlap degree; ideally, the global overlap degree of the complete operation trajectory is close to 1;
    全局覆盖度定义为内面积和外面积的比值，用γ表示The global coverage is defined as the ratio of the inner area to the outer area, denoted by γ
    Figure PCTCN2019110510-appb-100003
    Figure PCTCN2019110510-appb-100003
    全局覆盖度γ表达了实际作业面积覆盖作业区域的程度；作业越规范，轨迹越均匀，则全局覆盖度越大；作业异常，重叠集中，则全局覆盖度越小。The global coverage γ expresses the extent to which the actual work area covers the work area; the more standardized the work and the more uniform the trajectory, the greater the global coverage; the greater the work is abnormal and the overlap is concentrated, the smaller the global coverage.
  7. 根据权利要求6所述的一种基于定位漂移测算模型的农机作业面积测算方法，其特征在于：所述步骤4.2具体包括下列步骤：A method for measuring and calculating the working area of agricultural machinery based on a positioning drift measurement model according to claim 6, wherein said step 4.2 specifically includes the following steps:
    步骤4.2.1：优化核心特征的参数，选择sigmod函数将参数映射到[0,1]区间Step 4.2.1: Optimize the parameters of the core features, select the sigmod function to map the parameters to the interval [0,1]
    Figure PCTCN2019110510-appb-100004
    Figure PCTCN2019110510-appb-100004
    步骤4.2.2：根据样本数据训练核心特征的参数x和y，确定具体映射函数K 1(α)和K 2(β)。 Step 4.2.2: Train the parameters x and y of the core feature according to the sample data, and determine the specific mapping functions K 1 (α) and K 2 (β).
  8. 根据权利要求7所述的一种基于定位漂移测算模型的农机作业面积测算方法，其特征在于：The method for measuring and calculating the working area of agricultural machinery based on the positioning drift measurement model according to claim 7, characterized in that:
    所述步骤4.3具体包括下列步骤：The step 4.3 specifically includes the following steps:
    步骤4.3.1：构造定位漂移测算模型，利用优化后的核心特征的参数进行形式化表示；Step 4.3.1: Construct a positioning drift measurement model, and use the optimized core feature parameters for formal expression;
    特别地，对S 空空的结构进行分析，S 空＝S 漂移+S 异常中S 异常可以具体分为两种类型，一种是冗余部分，特别当全局重叠度β小于1的情况下，后续有其他的轨迹进行填充；另一种是未作业空白部分，由于作业过程中未覆盖产生空白区域，面积表示为S 未作业； In particular, the empty space S for analysis of the structure, space S = S + S drift anomalies S abnormalities may particularly be divided into two types, one is redundant portion, especially in the case where the overall degree of overlap less than 1 β, There will be other tracks to fill in the follow-up; the other is the unworked blank part, because the blank area is not covered during the operation, the area is expressed as S unworked ;
    方法定义轨迹的冗余率为θ 冗余作业轨迹内部面积中冗余的作业面积占据的比例： The method defines the redundancy rate of the trajectory as the proportion of the redundant operation area in the internal area of the θ redundant operation trajectory:
    Figure PCTCN2019110510-appb-100005
    Figure PCTCN2019110510-appb-100005
    以S 冗余表示对应的作业冗余面积，β为全局重叠度，则有： S redundancy represents the corresponding job redundancy area, β is the global overlap degree, then:
    S 冗余＝θ 冗余×S 空 S redundancy = θ redundancy × S empty
    同时定义轨迹的空白率θ 空白为实际轨迹内异常作业产生空白区域的面积占据除去作业冗余面积后的空白面积的比例，则有： At the same time, define the blank rate θ blank of the trajectory as the ratio of the blank area generated by abnormal operations in the actual trajectory to the blank area after removing the redundant area of the operation, then:
    S 未作业＝θ 空白×(S 空-S 冗余) S not working = θ blank × (S blank- S redundancy )
    S 异常是异常作业产生的作业未覆盖面积，综上分析可知，应表示如下： S anomaly is the uncovered area caused by the abnormal operation. In summary, it can be seen that it should be expressed as follows:
    S 异常＝S 冗余+S 未作业 S abnormal = S redundancy + S not working
    步骤4.3.2：基于优化后的核心特征的参数，运用机器学习方法建立定位漂移测算模型；Step 4.3.2: Based on the optimized core feature parameters, use machine learning methods to establish a positioning drift measurement model;
    基于试验样本的轨迹，我们将实际作业面积用S 正表示，存在关系： Based on the trajectory of the test sample, we denote the actual work area as positive S, and there is a relationship:
    S 正＝S 外-S 空×f(α，β) S positive = S outside- S empty × f(α, β)
    其中f(α，β)为S 异常在S 空中占比的计算函数，其中空白率θ 空白的算式已知，因此还需要建立θ 空白的计算模型； Wherein f (α, β) is a function S S abnormality calculating the proportion of air in which the blank of the known formula θ blank, it is also necessary to establish a blank θ calculation model;
    经过试验发现，当β保持不变，α越大，θ 空白的值越大；当α保持不变，β越大值，θ 空白越小；依据上述结论，通过sigmod函数构建空白率θ 空白的计算 模型： After testing found that when beta] remain constant, [alpha], the greater the blank value [theta]; [alpha] remains unchanged when, the larger the value beta], [theta] is smaller blank; According to the above conclusion, by constructing sigmod function of [theta] Blank Blank Calculation model:
    Figure PCTCN2019110510-appb-100006
    Figure PCTCN2019110510-appb-100006
    运用机器学习方法，结合具体的样本轨迹建立定位漂移测算模型的明确表示；基于试验样本的轨迹，我们将实际作业面积用S 正表示，存在关系： Using machine learning methods, combined with the specific sample trajectory to establish a clear representation of the positioning drift measurement model; based on the trajectory of the test sample, we use the actual working area as positive S, and there is a relationship:
    S 正＝S 外-S 异常 S positive = S outside- S abnormal
    即S 正＝S 外-θ 冗余×S 空-θ 空白×(1-θ 冗余）×S 空 That is, S positive = S outside- θ redundancy × S empty- θ blank × (1-θ redundancy ) × S empty
    由此得到定位漂移测算模型。Thus, a positioning drift measurement model is obtained.
  9. 根据权利要求1所述的一种基于定位漂移测算模型的农机作业面积测算方法，其特征在于：The method for calculating the working area of agricultural machinery based on the positioning drift measurement model according to claim 1, characterized in that:
    所述步骤5具体包括下列步骤：The step 5 specifically includes the following steps:
    步骤5.1：将定位点预处理成轨迹序列，计算不同特征和参数的取值，生成待处理的轨迹；Step 5.1: Preprocess the positioning points into a trajectory sequence, calculate the values of different features and parameters, and generate the trajectory to be processed;
    步骤5.2：将待处理的轨迹输入定位漂移测算模型，实现轨迹的不同分类，并对漂移区域进行纠正和补偿，得到农机作业的实际面积。Step 5.2: Input the trajectory to be processed into the positioning drift measurement model to realize different classification of trajectories, and correct and compensate the drift area to obtain the actual area of agricultural machinery operation.
PCT/CN2019/110510 2019-10-11 2019-10-11 Agricultural machinery operation area calculation method based on positioning drift calculation model WO2021068177A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201980016520.2A CN111868566B (en) 2019-10-11 2019-10-11 Agricultural machinery operation area measuring and calculating method based on positioning drift measuring and calculating model
PCT/CN2019/110510 WO2021068177A1 (en) 2019-10-11 2019-10-11 Agricultural machinery operation area calculation method based on positioning drift calculation model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/110510 WO2021068177A1 (en) 2019-10-11 2019-10-11 Agricultural machinery operation area calculation method based on positioning drift calculation model

Publications (1)

Publication Number Publication Date
WO2021068177A1 true WO2021068177A1 (en) 2021-04-15

Family

ID=72970932

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/110510 WO2021068177A1 (en) 2019-10-11 2019-10-11 Agricultural machinery operation area calculation method based on positioning drift calculation model

Country Status (2)

Country Link
CN (1) CN111868566B (en)
WO (1) WO2021068177A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114526669A (en) * 2022-02-22 2022-05-24 上海联适导航技术股份有限公司 Method, device and equipment for measuring and calculating real-time leveling area of agricultural machinery operation
CN115436973A (en) * 2022-09-02 2022-12-06 湖北地信科技集团股份有限公司 Beidou agricultural machine operation track rapid filtering and grouping method

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113405519B (en) * 2021-01-29 2023-07-18 丰疆智能(深圳)有限公司 Agricultural machinery operation acre calculation method and agricultural machinery operation acre calculation system
CN113436248B (en) * 2021-06-18 2023-05-23 黑龙江惠达科技发展有限公司 Method and device for calculating working area of agricultural machine
CN113761730B (en) * 2021-08-27 2023-09-29 浙江理工大学 Tea garden four-bar intertillage mechanism multi-target point track comprehensive method based on improved difference
CN114662621B (en) * 2022-05-24 2022-09-06 灵枭科技(武汉)有限公司 Agricultural machinery working area calculation method and system based on machine learning

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102890706A (en) * 2012-08-27 2013-01-23 首都师范大学 Data processing method and device
CN103673937A (en) * 2013-11-11 2014-03-26 安徽赛为信息技术有限责任公司 Dual-algorithm mode farm machinery field operation area calculating system and method
CN104992072A (en) * 2015-07-21 2015-10-21 江苏北斗卫星应用产业研究院有限公司 Operation land parcel automatic identification and area statistics method based on spatial mesh division
WO2015193822A1 (en) * 2014-06-17 2015-12-23 Casella Macchine Agricole S.R.L. Method and device for measuring vegetation cover on farmland
CN106017400A (en) * 2016-07-13 2016-10-12 哈尔滨工业大学 Farm machinery operation area measurement method based on farming trajectory equivalent rectangle accumulation
CN107036572A (en) * 2017-04-12 2017-08-11 中国农业大学 A kind of agricultural machinery working area acquisition methods and device
CN108332652A (en) * 2018-01-15 2018-07-27 中国农业大学 A kind of farmland working area metering method and device
CN109813273A (en) * 2019-03-19 2019-05-28 中电科卫星导航运营服务有限公司 A kind of agricultural machinery repetition working area determination method based on spatial analysis

Family Cites Families (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2168870B (en) * 1984-05-10 1987-09-03 Secr Defence Imaging system
WO1992000500A1 (en) * 1989-07-14 1992-01-09 Von Schrader Company Apparatus for determining an area coverage rate
US5902343A (en) * 1996-11-22 1999-05-11 Case Corporation Automatic scaling of GPS field maps
US6087984A (en) * 1998-05-04 2000-07-11 Trimble Navigation Limited GPS guidance system for use with circular cultivated agricultural fields
US6711501B2 (en) * 2000-12-08 2004-03-23 Satloc, Llc Vehicle navigation system and method for swathing applications
US6539303B2 (en) * 2000-12-08 2003-03-25 Mcclure John A. GPS derived swathing guidance system
JP5996468B2 (en) * 2013-03-28 2016-09-21 株式会社クボタ Agricultural management support system
US9282688B2 (en) * 2014-04-25 2016-03-15 Deere & Company Residue monitoring and residue-based control
CN204666091U (en) * 2015-06-04 2015-09-23 江苏北斗卫星应用产业研究院有限公司 Based on the agricultural machinery terminal of satnav working area statistics
CN105539851B (en) * 2015-12-09 2017-06-20 华南农业大学 Unmanned plane agricultural chemicals based on wireless sense network precisely sprays operating system and method
US9891629B2 (en) * 2016-02-04 2018-02-13 Deere & Company Autonomous robotic agricultural machine and system thereof
CN106247926B (en) * 2016-07-13 2019-05-14 哈尔滨工业大学 Farm machinery tillage Area computing method based on cell scanning and GPS track interpolation
CN108036717A (en) * 2017-11-30 2018-05-15 北京博创联动科技有限公司 A kind of agricultural machine working area measuring method and device
CN108917590A (en) * 2018-05-11 2018-11-30 农业部南京农业机械化研究所 Plant protection drone operation quality detection device and its detection method
CN109917430B (en) * 2019-04-03 2021-01-22 安徽中科智能感知产业技术研究院有限责任公司 Satellite positioning track drift correction method based on track smoothing algorithm

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102890706A (en) * 2012-08-27 2013-01-23 首都师范大学 Data processing method and device
CN103673937A (en) * 2013-11-11 2014-03-26 安徽赛为信息技术有限责任公司 Dual-algorithm mode farm machinery field operation area calculating system and method
WO2015193822A1 (en) * 2014-06-17 2015-12-23 Casella Macchine Agricole S.R.L. Method and device for measuring vegetation cover on farmland
CN104992072A (en) * 2015-07-21 2015-10-21 江苏北斗卫星应用产业研究院有限公司 Operation land parcel automatic identification and area statistics method based on spatial mesh division
CN106017400A (en) * 2016-07-13 2016-10-12 哈尔滨工业大学 Farm machinery operation area measurement method based on farming trajectory equivalent rectangle accumulation
CN107036572A (en) * 2017-04-12 2017-08-11 中国农业大学 A kind of agricultural machinery working area acquisition methods and device
CN108332652A (en) * 2018-01-15 2018-07-27 中国农业大学 A kind of farmland working area metering method and device
CN109813273A (en) * 2019-03-19 2019-05-28 中电科卫星导航运营服务有限公司 A kind of agricultural machinery repetition working area determination method based on spatial analysis

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
HUI LIU, ZHIJUN MENG, WEIQIANG FU: "Overlap and skip evaluation for agricultural machinery operation based on GPS track logs", TRANSACTIONS OF THE CHINESE SOCIETY OF AGRICULTURAL ENGINEERING, CHINESE SOCIETY OF AGRICULTURAL ENGINEERING, ZHONGGUO NONGYE GONGCHENG XUEHUI, CN, vol. 28, no. 18, 1 September 2012 (2012-09-01), CN, pages 149 - 154, XP055800896, ISSN: 1002-6819, DOI: 10.3969/j.issn.10026819.2012.18.022 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114526669A (en) * 2022-02-22 2022-05-24 上海联适导航技术股份有限公司 Method, device and equipment for measuring and calculating real-time leveling area of agricultural machinery operation
CN115436973A (en) * 2022-09-02 2022-12-06 湖北地信科技集团股份有限公司 Beidou agricultural machine operation track rapid filtering and grouping method

Also Published As

Publication number Publication date
CN111868566A (en) 2020-10-30
CN111868566B (en) 2023-10-03

Similar Documents

Publication Publication Date Title
WO2021068177A1 (en) Agricultural machinery operation area calculation method based on positioning drift calculation model
CN105842679B (en) A kind of in-orbit geometric calibration method and system of domestic laser satellite altimeter
CN107462208A (en) A kind of agricultural machinery and agricultural machinery working area measuring device and measuring method
CN107944018A (en) A kind of automatic quality detecting method of map vector positional precision based on laser point cloud data
CN102750413B (en) Data processing and mapping method of topographic surveying of electric transmission line tower positions
CN110030972A (en) Tunnel Overbreak & Underbreak detection method based on ExcelVBA
CN105303063A (en) Leaf area index inversion method and system of merged phenological data and remote sensing data
CN101285686A (en) Agricultural machines navigation hierarchical positioning process and system
CN109063553A (en) Field-crop growth defect area's remote sensing fast diagnosis method after a kind of land control
CN111105496A (en) High-precision DEM construction method based on airborne laser radar point cloud data
CN108919304A (en) POS error compensating method in a kind of traverse measurement system based on reference planes
CN111256730A (en) Earth mass balance correction calculation method for low-altitude oblique photogrammetry technology
CN110456319A (en) A kind of radar intervisibility calculation method based on SRTM
CN115855067A (en) Path planning method for curved farmland boundary
CN101846512B (en) Method for implementing land patrol
CN107985400A (en) Operating area paths planning method and device
CN205121343U (en) Vegetation data collection system and unmanned vehicles based on unmanned vehicles
CN109282850A (en) A kind of farming parameter measuring apparatus and method, agricultural machinery and implement operation tilling depth measurement method
CN105717517B (en) A kind of vehicle-mounted big dipper multi-mode GNSS high accuracy roads basis collecting method
CN114184179B (en) Outdoor geographical survey and drawing management and control system based on internet
LU500256B1 (en) Agricultural Machinery Operation Area Calculation Method Based on Positioning Drift Calculation Model
CN115962755A (en) Earth and stone calculation method based on unmanned aerial vehicle oblique photography technology
KR101436045B1 (en) Measurement Method of Road Slope Using Vehicle's Travel Data
CN110804912B (en) Method for extracting deformation information of railway line and area along railway line
CN115346128A (en) Optical three-dimensional satellite DEM elevation correction and fusion method

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19948284

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19948284

Country of ref document: EP

Kind code of ref document: A1