Steering control method and system for crawler bulldozer based on agricultural big data

Abstract

The present invention provides a crawler bulldozer steering control method and system based on agricultural big data, which belongs to the technical field of big data processing. The method comprises: step 1: extracting the operation attribute characteristics of the target crawler bulldozer; step 2: acquiring agricultural big data suitable for training according to the operation attribute characteristics; step 3: establishing a simulation model according to the agricultural big data, setting the control parameters of the target crawler bulldozer in the simulation model, and determining the target simulation operation process set; step 4: determining the actual control parameters and performing steering control according to the target simulation operation process set and the actual operation process. The crawler bulldozer steering control method and system based on agricultural big data of the present invention acquires agricultural big data suitable for training according to the operation attribute characteristics to establish a simulation model, determines the actual control parameters according to the target simulation operation process set and the actual operation process to perform steering control, and improves the suitability of the steering control of the target crawler bulldozer and the stability of the target crawler bulldozer.

Description

Technical Field

[0001] The present invention relates to the technical field of big data processing, and in particular to a steering control method and system for a crawler bulldozer based on agricultural big data. Background Art

[0002] As a typical low-pressure agricultural machinery, crawler bulldozers are gaining more and more attention for their maneuverability. At present, most of the research on crawler bulldozers focuses on their straight-line driving stability. However, in actual agricultural operations, crawler bulldozers often encounter situations where they need to turn. By performing precise steering control on them, they can better adapt to complex environments.

[0003] The invention patent with application number: CN202010619826.1 discloses a bulldozer steering control method and system, wherein the method includes: obtaining the steering mode of the bulldozer; obtaining the steering control instruction and engine speed of the bulldozer; obtaining the target steering speed of the bulldozer according to the steering mode, steering control instruction and engine speed; obtaining the load pressure of the bulldozer and the actual steering speed under the load pressure; obtaining the steering speed difference of the bulldozer according to the target steering speed and the actual steering speed; adjusting the engine speed and the speed of the left and right travel motors in real time according to the steering speed difference to stably control the steering speed of the bulldozer. The above invention can automatically adjust the steering speed of the bulldozer under load, thereby avoiding the situation of too fast or too slow steering speed, and can improve the working efficiency and driving comfort of the bulldozer. In addition, it can also effectively reduce the difficulty of the bulldozer steering operation and reduce the operational impact caused by the bulldozer steering.

[0004] However, the above-mentioned prior art can only set a fixed steering mode for steering control, while the actual bulldozer operation scene is complex and changeable. When the steering mode is set unreasonably, the steering control is also unsuitable, further reducing the stability of the crawler bulldozer.

[0005] In view of this, there is an urgent need for a steering control method and system for a crawler bulldozer based on agricultural big data to at least solve the above-mentioned deficiencies. Summary of the invention

[0006] One of the purposes of the present invention is to provide a steering control method and system for a crawler bulldozer based on agricultural big data, extract the operating attribute characteristics of the target crawler bulldozer, obtain agricultural big data based on the operating attribute characteristics to establish a simulation model, simulate the bulldozing scene suitable for the operating attributes of the target crawler bulldozer, determine the simulation result as a target simulation operation process set for successful simulated steering according to the set control parameters, match each target simulation operation process in the target simulation operation process set with the actual operation process, determine the actual control parameters corresponding to the control parameters of the process-matched target simulation operation process for steering control, thereby improving the suitability of the steering control and further improving the stability of the crawler bulldozer.

[0007] The crawler bulldozer steering control method based on agricultural big data provided by the embodiment of the present invention includes:

[0008] Step 1: Obtain the operating attributes of the target crawler bulldozer and extract the operating attribute features;

[0009] Step 2: Obtain agricultural big data suitable for training based on the operation attribute characteristics;

[0010] Step 3: Establish a simulation model based on agricultural big data, set the control parameters of the target crawler bulldozer in the simulation model, and determine the target simulation operation process set;

[0011] Step 4: According to the target simulated operation process set and the actual operation process, determine the actual control parameters of the target crawler bulldozer and perform steering control.

[0012] Preferably, step 1: obtaining the operating attributes of the target crawler bulldozer and extracting operating attribute features, includes:

[0013] According to the job attributes, determine the job attribute feature extraction standard;

[0014] The operation attribute features are extracted according to the operation attribute feature extraction standard; the operation attribute features include: steering torque, track gauge, steering angular velocity and speed on both sides of the track.

[0015] Preferably, step 2: obtaining agricultural big data suitable for training according to operation attribute characteristics, including:

[0016] Determine a first agricultural operation scene of the target crawler bulldozer; the first agricultural operation scene is: a historical agricultural operation scene of the target crawler bulldozer;

[0017] Determine first target data according to the first agricultural operation scenario;

[0018] Determine the template according to the operation attribute characteristics and the operation scene, and determine the second agricultural operation scene;

[0019] Based on the agricultural big data platform, the second target data is determined according to the second agricultural operation scenario;

[0020] The first target data and the second target data are taken together as agricultural big data suitable for training.

[0021] Preferably, determining a template according to the operation attribute characteristics and the operation scene to determine the second agricultural operation scene includes:

[0022] Traverse each job attribute feature in turn, and use the job attribute feature being traversed as the first target job attribute feature;

[0023] Matching the first target operation attribute feature with the second target operation attribute feature of the same operation attribute feature type as the first target operation attribute feature in the operation scene determination template, and if the match is satisfactory, determining a third agricultural operation scene associated with the second target operation attribute feature that matches the match;

[0024] According to the matched operation attribute feature type and the third agricultural operation scene, a weight of the operation attribute feature type is determined, and associated with the third agricultural operation scene;

[0025] When the operation attribute feature traversal is completed, the operation attribute feature type weights associated with the third agricultural operation scene are accumulated to obtain the target weight sum;

[0026] If the sum of the target weights is greater than or equal to the preset target threshold corresponding to the third agricultural operation scene, the corresponding third agricultural operation scene will be used as the second agricultural operation scene.

[0027] Preferably, based on the agricultural big data platform, according to the second agricultural operation scenario, determining the second target data includes:

[0028] Based on the agricultural big data platform, according to the second agricultural operation scenario, the third target data is determined; the third target data is: the verified data of the agricultural big data platform;

[0029] Obtaining an operating posture set of a historical operating crawler bulldozer in a second agricultural operation scene;

[0030] According to the operation posture set, the duration ratio of the steering posture duration to the non-steering posture duration of each historical operation crawler bulldozer is calculated;

[0031] If the duration ratio is greater than or equal to a preset duration ratio threshold, the second target data is determined according to the bulldozing trajectory of the crawler bulldozer of the corresponding historical operation and the third target data.

[0032] Preferably, setting the control parameters of the target crawler bulldozer in the simulation model includes:

[0033] Obtaining the working information of the target crawler bulldozer;

[0034] Determine the complexity of the job based on the work information;

[0035] If the operation complexity is less than or equal to the preset operation complexity threshold, the control parameters are randomly set according to the work information to determine the simulation operation process;

[0036] If the operation complexity is greater than a preset operation complexity threshold, a manual bulldozing operation is determined according to the work information and a preset manual bulldozing operation model;

[0037] Determining a second operation feature of the target crawler bulldozer according to a first operation feature of a manual bulldozer operation and a preset feature conversion template;

[0038] The control parameters are determined according to the second operation characteristic and a preset control parameter input template.

[0039] The crawler bulldozer steering control method based on agricultural big data provided by the embodiment of the present invention also includes:

[0040] When the steering control of the target crawler bulldozer is performed, the tipping detection of the target crawler bulldozer is performed, and the balance control is performed according to the tipping detection result.

[0041] Preferably, when performing steering control on the target crawler bulldozer, a dumping detection of the target crawler bulldozer is performed, and balance control is performed according to the dumping detection result, including:

[0042] Obtaining a first gravity acceleration vector of the straight-ahead posture of the target crawler bulldozer at a target time before performing steering control of the target crawler bulldozer, and marking it in a three-dimensional model diagram of the target crawler bulldozer;

[0043] Obtaining a second gravity acceleration vector of the target crawler bulldozer in a steering posture, and marking it in a three-dimensional model diagram of the target crawler bulldozer;

[0044] According to the three-dimensional model diagram, the cosine of the angle between the first gravity acceleration vector and the second gravity acceleration vector is calculated;

[0045] According to the cosine of the included angle, the moving direction and moving distance of the balancing slider preset in the target crawler bulldozer are determined;

[0046] Control the balance slider to move in the moving direction and distance;

[0047] When the moving distance reaches the maximum distance of the balance slider in the moving direction, the change of the cosine of the angle within the preset time period is determined. If the cosine of the angle still decreases, the radar data around the target crawler bulldozer is obtained;

[0048] Based on the radar data, the on-site reminder personnel are determined and voice reminders are given.

[0049] Preferably, obtaining radar data around the target crawler bulldozer includes:

[0050] Determine the force condition in the tipping direction according to the change of the cosine of the angle within a preset time period;

[0051] According to the force conditions, the body parameters of the target crawler bulldozer in the dumping direction and the current driving ground conditions, the dumping of the target crawler bulldozer is simulated to obtain the simulated dumping result;

[0052] Determine the scanning range of the laser radar based on the simulated dumping results;

[0053] Control the laser radar to obtain radar data within the scanning range.

[0054] The crawler bulldozer steering control system based on agricultural big data provided by the embodiment of the present invention includes:

[0055] The operation attribute feature extraction subsystem is used to obtain the operation attributes of the target crawler bulldozer and extract the operation attribute features;

[0056] Agricultural big data acquisition subsystem, used to acquire agricultural big data suitable for training according to operation attribute characteristics;

[0057] The target simulation operation process set determination subsystem is used to establish a simulation model based on agricultural big data, set the control parameters of the target crawler bulldozer in the simulation model, and determine the target simulation operation process set;

[0058] The steering control subsystem is used to determine the actual control parameters of the target crawler bulldozer according to the target simulated operation process set and the actual operation process, and perform steering control.

[0059] Preferably, the job attribute feature extraction subsystem includes:

[0060] A job attribute feature extraction standard determination module is used to determine the job attribute feature extraction standard according to the job attributes;

[0061] The feature extraction module is used to extract the operation attribute features according to the operation attribute feature extraction standard; the operation attribute features include: steering torque, track gauge, steering angular velocity and speed on both sides of the track.

[0062] Preferably, the agricultural big data acquisition subsystem includes:

[0063] A first agricultural operation scene determination module is used to determine a first agricultural operation scene of a target crawler bulldozer; the first agricultural operation scene is: a historical agricultural operation scene of the target crawler bulldozer;

[0064] A first target data determination module, used to determine first target data according to a first agricultural operation scenario;

[0065] A second agricultural operation scene determination module, used to determine a template according to operation attribute characteristics and the operation scene, and determine a second agricultural operation scene;

[0066] A second target data determination module, used to determine second target data based on the agricultural big data platform and according to a second agricultural operation scenario;

[0067] The agricultural big data determination module is used to use the first target data and the second target data together as agricultural big data suitable for training.

[0068] The beneficial effects of the present invention are:

[0069] The present invention extracts the operating attribute characteristics of the target crawler bulldozer, acquires agricultural big data based on the operating attribute characteristics to establish a simulation model, simulates a bulldozer scene suitable for the operating attributes of the target crawler bulldozer, determines the simulation result as a target simulation operating process set for successful simulated steering according to the set control parameters, matches each target simulation operating process in the target simulation operating process set with the actual operating process, determines the actual control parameters corresponding to the control parameters of the process-matched target simulation operating process, performs steering control, improves the suitability of the steering control, and further improves the stability of the crawler bulldozer.

[0070] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in the present application documents.

[0071] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0073] Figure 1 It is a schematic diagram of a crawler bulldozer steering control method based on agricultural big data in an embodiment of the present invention;

[0074] Figure 2 It is a schematic diagram of a steering control system of a crawler bulldozer based on agricultural big data in an embodiment of the present invention. DETAILED DESCRIPTION

[0075] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0076] The embodiment of the present invention provides a steering control method for a crawler bulldozer based on agricultural big data, such as Figure 1 As shown, including:

[0077] Step 1: Obtain the operating attributes of the target crawler bulldozer and extract the operating attribute features; wherein the target crawler bulldozer is: a crawler bulldozer that needs to be turned; the operating attributes are: operating information of the target crawler bulldozer, such as: bulldozing capacity, speed and operation mode; the operating attribute features are: a characteristic representation of the information in the operating attributes that affects the steering performance of the target crawler bulldozer, such as: steering torque, track gauge, steering angular velocity and speed on both sides of the track. When extracting, just refer to the manually set feature extraction template for extraction;

[0078] Step 2: Obtain agricultural big data suitable for training according to the operation attribute characteristics; the agricultural big data suitable for training is: agricultural related data suitable for the operation attributes of the target crawler bulldozer, such as: soil type, soil flatness and soil moisture content;

[0079] Step 3: Establish a simulation model based on agricultural big data, set the control parameters of the target crawler bulldozer in the simulation model, and determine the target simulation operation process set; wherein, the simulation model is: a bulldozer scene model simulated based on agricultural big data; the control parameters are: virtual control parameters of the three-dimensional model of the target crawler bulldozer in the simulated bulldozer scene; the target simulation operation process set is: a set of process sequences whose simulation results are successful in simulated steering and contain the sequence of simulated operation processes;

[0080] Step 4: According to the target simulated operation process set and the actual operation process, determine the actual control parameters of the target crawler bulldozer and perform steering control. When determining the actual control parameters of the target crawler bulldozer according to the target simulated operation process set and the actual operation process, each target simulated operation process in the target simulated operation process set is matched with the actual operation process to determine the process matching value, which indicates the similarity between the target simulated operation process and the actual operation process; if the maximum process matching value is greater than or equal to the preset process matching value threshold, the actual control parameters of the crawler bulldozer are determined according to the control parameters of the corresponding target simulated operation process.

[0081] The working principle and beneficial effects of the above technical solution are:

[0082] The present application extracts the operating attribute characteristics of the target crawler bulldozer, obtains agricultural big data based on the operating attribute characteristics to establish a simulation model, simulates a bulldozer scene suitable for the operating attributes of the target crawler bulldozer, and determines the simulation result as a target simulation operating process set for successful simulated steering according to the set control parameters. Each target simulation operating process in the target simulation operating process set is matched with the actual operating process, and the actual control parameters corresponding to the control parameters of the process-matched target simulation operating process are determined for steering control, thereby improving the suitability of the steering control and further improving the stability of the crawler bulldozer.

[0083] In one embodiment, step 1: obtaining the operation attributes of the target crawler bulldozer and extracting the operation attribute features includes:

[0084] According to the job attributes, the job attribute feature extraction standard is determined; wherein the job attribute feature extraction standard is: the job attribute feature extraction specification;

[0085] The operation attribute features are extracted according to the operation attribute feature extraction standard; the operation attribute features include: steering torque, track gauge, steering angular velocity and speed on both sides of the track.

[0086] The working principle and beneficial effects of the above technical solution are:

[0087] This application introduces a job attribute feature extraction standard to extract job attribute features, thereby improving the accuracy of job attribute features.

[0088] In one embodiment, step 2: obtaining agricultural big data suitable for training according to operation attribute characteristics, including:

[0089] Determine a first agricultural operation scene of the target crawler bulldozer; the first agricultural operation scene is: a historical agricultural operation scene of the target crawler bulldozer; wherein the historical agricultural operation scene is: a bulldozer operation scene encountered by the target crawler bulldozer in past agricultural operations;

[0090] Determine first target data according to the first agricultural operation scene; wherein the first target data is: soil conditions in the first agricultural operation scene;

[0091] Determine a template according to the operation attribute characteristics and the operation scene, and determine a second agricultural operation scene; wherein the operation scene determination template is: a template for determining a suitable agricultural operation scene according to the operation attribute characteristics; the second agricultural operation scene is: an operation scene suitable for the working performance of the target crawler bulldozer;

[0092] Based on the agricultural big data platform, according to the second agricultural operation scene, second target data is determined; the second target data is: soil conditions in the second agricultural operation scene;

[0093] The first target data and the second target data are taken together as agricultural big data suitable for training.

[0094] The working principle and beneficial effects of the above technical solution are:

[0095] When simulating the turning of the target crawler bulldozer, not all soil conditions are suitable for simulation. For example, the tidal flat will not be encountered during the actual operation of the target crawler bulldozer, so the data does not need to be simulated. Therefore, this application extracts the first target data in the historical agricultural operation scene of the target crawler bulldozer. In addition, an operation scene determination template is introduced for determining the second agricultural operation scene by comparing the operation attribute characteristics, and then the second target data is obtained from the agricultural big data platform. The first target data and the second target data are used together as agricultural big data suitable for training, which improves the rationality and comprehensiveness of agricultural big data acquisition.

[0096] In one embodiment, determining a template according to the operation attribute characteristics and the operation scene to determine the second agricultural operation scene includes:

[0097] Traverse each job attribute feature in turn, and use the job attribute feature being traversed as the first target job attribute feature;

[0098] Matching the first target operation attribute feature with the second target operation attribute feature of the same operation attribute feature type as the first target operation attribute feature in the operation scene determination template, and if the match is satisfactory, determining the third agricultural operation scene associated with the second target operation attribute feature that matches the match; wherein the operation attribute feature type is: feature type, such as: steering torque;

[0099] According to the matched operation attribute feature type and the third agricultural operation scene, the operation attribute feature type weight is determined and associated with the third agricultural operation scene; wherein the operation attribute feature type weight represents the importance of the operation feature of the operation attribute feature type in the third agricultural operation scene, and the higher the operation attribute feature type weight is, the more important it is. For example, for the third agricultural operation scene requiring steering operation, the operation attribute feature type weight of the steering torque is 0.5, and for the third agricultural operation scene not requiring steering operation, the operation attribute feature type weight of the steering torque is 0.1;

[0100] When the operation attribute feature traversal is completed, the operation attribute feature type weights associated with the third agricultural operation scene are accumulated to obtain the target weight sum;

[0101] If the sum of the target weights is greater than or equal to the preset target threshold corresponding to the third agricultural operation scene, the corresponding third agricultural operation scene is used as the second agricultural operation scene. The preset target threshold is manually preset.

[0102] The working principle and beneficial effects of the above technical solution are:

[0103] The present application matches the first target operation attribute feature being traversed with the second target operation attribute feature of the same operation attribute feature type as the first target operation attribute feature in the operation scene determination template, and determines the third agricultural operation scene associated with the matched second target operation attribute feature. According to the matched operation attribute feature type and the third agricultural operation scene, the operation attribute feature type weight is determined, and the operation attribute feature type weight corresponding to the third agricultural operation scene is summed to obtain the target weight sum, and the third agricultural operation scene whose target weight sum is greater than or equal to the preset target threshold corresponding to the third agricultural operation scene is used as the second agricultural operation scene, so as to improve the suitability of the second agricultural operation scene screening.

[0104] In one embodiment, based on the agricultural big data platform, according to the second agricultural operation scenario, determining the second target data includes:

[0105] Based on the agricultural big data platform, according to the second agricultural operation scenario, the third target data is determined; the third target data is: the verified data of the agricultural big data platform;

[0106] Obtaining the operating posture set of the historical operating crawler bulldozer in the second agricultural operation scene; wherein the historical operating crawler bulldozer is: the crawler bulldozer that worked historically in the second agricultural operation scene; the operating posture set is: a set of operating actions of the historical operating crawler bulldozer, and the operating action is, for example: straight-moving bulldozer, turning bulldozer;

[0107] According to the working posture set, the duration ratio of the steering posture duration and the non-steering posture duration of each historical working crawler bulldozer is calculated; wherein the steering posture duration is: the length of time the steering posture is maintained, such as: 10 minutes; the non-steering posture duration is: the duration of other postures except the steering posture when the historical working crawler bulldozer is working, such as: 40 minutes; the duration ratio is: the result obtained by dividing the steering posture duration by the non-steering posture duration;

[0108] If the duration ratio is greater than or equal to the preset duration ratio threshold, the second target data is determined based on the bulldozer's bulldozer trajectory of the corresponding historical operation and the third target data; wherein the preset duration ratio threshold is manually pre-set; the specific process of determining the second target data is: first determine the soil area corresponding to the bulldozer trajectory of the historical operation crawler bulldozer whose duration ratio is greater than or equal to the preset duration ratio threshold, and then determine the third target data corresponding to the soil area as the second target data.

[0109] The working principle and beneficial effects of the above technical solution are:

[0110] When extracting the third target data through the second agricultural operation scene, it is necessary to extract valuable data, such as: the proportion of the path of the steering control should be appropriate, so as to obtain more steering control learning information. Therefore, this application introduces the operating posture set of the historical operating crawler bulldozer in the second agricultural operation scene, and calculates the duration ratio of the steering posture duration and the non-steering posture duration of each historical operating crawler bulldozer. The duration ratio represents the proportion of the steering posture to all postures. The larger the proportion, the more steering information can be learned. Therefore, if the duration ratio is greater than or equal to the preset duration ratio threshold, the soil area corresponding to the bulldozer trajectory of the corresponding historical operating crawler bulldozer is determined, and then the third target data corresponding to the soil area is determined as the second target data. The second target data can be learned with higher value.

[0111] In one embodiment, setting control parameters of the target crawler bulldozer in the simulation model includes:

[0112] Obtaining working information of a target crawler bulldozer; wherein the working information includes: working task information;

[0113] Determine the job complexity based on the job information; job complexity is: a quantitative representation of the complexity of the job task;

[0114] If the operation complexity is less than or equal to a preset operation complexity threshold, the control parameters are randomly set according to the operation information to determine the simulated operation process; wherein the preset operation complexity threshold is manually set in advance; the control parameters are randomly set according to the operation information to determine the simulated operation process refers to: randomly setting the control parameters to determine the simulated operation process corresponding to the control parameters for achieving the work task;

[0115] If the operation complexity is greater than a preset operation complexity threshold, a manual bulldozing operation is determined according to the operation information and a preset manual bulldozing operation model; wherein the manual bulldozing operation model is: an AI model that automatically determines the manual bulldozing operation according to the operation information;

[0116] According to the first operation feature of the manual bulldozer operation and the preset feature conversion template, the second operation feature of the target crawler bulldozer is determined; wherein the first operation feature is: the operation feature representation of the manual bulldozer operation, for example: first push point A, then push point B; the preset feature conversion template is: a template for generating machine operation features by reference to the manual operation feature, and the second operation feature is: the operation feature representation of the machine bulldozer operation, for example: first control the target crawler bulldozer to reach point A to perform the bulldozer operation, and then plan the shortest path from point A to point B to perform the bulldozer operation at point B;

[0117] The control parameter is determined according to the second operation characteristic and a preset control parameter input template, wherein the preset control parameter input template is a conversion template of the machine operation characteristic and the simulated control parameter.

[0118] The working principle and beneficial effects of the above technical solution are:

[0119] The present application introduces the working information of the target crawler bulldozer to determine the operation complexity of the work task. When the operation complexity is relatively small, the control parameters are randomly set directly according to the work information, and the control parameters are randomly set to determine the simulated operation process corresponding to the control parameters for achieving the work task. When the operation complexity is greater than the preset operation complexity threshold, a manual bulldozer operation model is introduced to determine the manual bulldozer operation. According to the first operation feature of the manual bulldozer operation and the preset feature conversion template, the second operation feature of the target crawler bulldozer is determined, and then the control parameter input template is introduced to convert the second operation feature control parameter, thereby improving the efficiency of determining the control parameters.

[0120] The embodiment of the present invention provides a steering control method for a crawler bulldozer based on agricultural big data, further comprising:

[0121] When the steering control of the target crawler bulldozer is performed, the tipping detection of the target crawler bulldozer is performed, and the balance control is performed according to the tipping detection result. The tipping detection is to detect whether the target crawler bulldozer has a tipping risk; and the balance control refers to measures to adjust the posture of the bulldozer.

[0122] The working principle and beneficial effects of the above technical solution are:

[0123] The present application detects whether a target crawler bulldozer has a risk of tipping over when the target crawler bulldozer is turning, and performs balance control when there is a risk of tipping over, thereby improving safety.

[0124] In one embodiment, when performing steering control on a target crawler bulldozer, performing a dumping detection on the target crawler bulldozer, and performing a balance control according to the dumping detection result, including:

[0125] The first gravity acceleration vector of the straight-ahead posture of the target crawler bulldozer at the target time before the steering control of the target crawler bulldozer is performed is obtained, and is marked in the three-dimensional model diagram of the target crawler bulldozer; wherein the target time is: 1 second before the steering control of the target crawler bulldozer is performed; the first gravity acceleration vector is: a vector with the center of gravity of the target crawler bulldozer as the vector starting point, the gravity acceleration direction of the target crawler bulldozer at the target time as the vector direction, and the gravity acceleration magnitude of the target crawler bulldozer at the target time as the vector modulus; the three-dimensional model diagram is: a three-dimensional schematic diagram of the target crawler bulldozer; when marking the first gravity acceleration vector, the first gravity acceleration vector is marked correspondingly in the three-dimensional schematic diagram of the target crawler bulldozer according to the relative position of the first gravity acceleration vector and the target crawler bulldozer;

[0126] The second gravity acceleration vector of the target crawler bulldozer in the steering posture is obtained and marked in the three-dimensional model diagram of the target crawler bulldozer; wherein the second gravity acceleration vector is: a vector having the center of gravity of the target crawler bulldozer as the vector starting point, the direction of the gravity acceleration of the target crawler bulldozer at the steering moment as the vector direction, and the magnitude of the gravity acceleration of the target crawler bulldozer at the steering moment as the vector modulus; when marking the second gravity acceleration vector, the second gravity acceleration vector is marked correspondingly in the three-dimensional schematic diagram of the target crawler bulldozer according to the relative position of the second gravity acceleration vector and the target crawler bulldozer;

[0127] According to the three-dimensional model diagram, the cosine of the angle between the first gravity acceleration vector and the second gravity acceleration vector is calculated; wherein the cosine of the angle is: the cosine value of the vector angle between the first gravity acceleration vector and the second gravity acceleration vector;

[0128] According to the cosine of the included angle, the moving direction and moving distance of the balancing slider preset in the target crawler bulldozer are determined; wherein, when the target crawler bulldozer is stationary on a horizontal ground, the center of gravity of the balancing slider preset in the target crawler bulldozer coincides with the center of gravity of the target crawler bulldozer, and the guide rail where the balancing slider is located is perpendicular to the bisector that bisects the target crawler bulldozer. When determining the moving direction and moving distance, the moving direction is opposite to the turning direction, the larger the cosine value, the smaller the moving distance, and the specific corresponding relationship between the cosine value and the moving distance is manually preset;

[0129] Control the balance slider to move in the moving direction and distance;

[0130] When the moving distance reaches the maximum distance of the balance slider in the moving direction, the change of the cosine of the angle within the preset time period is determined. If the cosine of the angle still decreases, the radar data around the target crawler bulldozer is obtained; wherein the preset time period is: 3 seconds after the moving distance reaches the maximum distance of the balance slider in the moving direction; the change is: whether the cosine of the angle increases, remains unchanged or decreases; the radar data is obtained according to the laser radar preset on the target crawler bulldozer;

[0131] According to the radar data, the on-site reminder personnel are determined and voice reminders are given. When the on-site reminder personnel are determined according to the radar data, if a three-dimensional human body model is identified, the on-site personnel corresponding to the three-dimensional human body model is determined as the reminder personnel, and the reminder is given based on a preset voice playback device.

[0132] The working principle and beneficial effects of the above technical solution are:

[0133] The present application marks the first gravity acceleration vector of the straight-ahead posture of the target crawler bulldozer at the target moment before the steering control of the target crawler bulldozer and the second gravity acceleration vector at the turning moment in the same three-dimensional model diagram, and calculates the cosine of the angle between the first gravity acceleration vector and the second gravity acceleration vector. The smaller the cosine of the angle, the more serious the roll of the target crawler bulldozer. In particular, when the cosine of the angle is 1 and the angle is 0°, it means that the target crawler bulldozer has not rolled.

[0134] A balancing slider is preset in the target crawler bulldozer, which can be a high-density metal block. The balancing slider is on a guide rail perpendicular to the bisector of the target crawler bulldozer, and when the target crawler bulldozer is stationary, the center of gravity of the balancing slider coincides with the center of gravity of the target crawler bulldozer. When the target crawler bulldozer tilts, the cosine of the angle begins to decrease, and the moving distance corresponding to the cosine of the angle is determined and the movement of the balancing slider is controlled based on the Internet of Things technology.

[0135] In particular, when the balance slider moves to the end of the guide rail, if the cosine of the angle continues to decrease, there is a possibility of rollover. Therefore, the change of the cosine of the angle within the preset time period is judged. If the cosine of the angle still decreases, the radar data around the target crawler bulldozer is obtained to determine whether there are other people around. If there are other people, they will be reminded, which improves the accuracy and suitability of balance control and is also safer.

[0136] In one embodiment, obtaining radar data around a target crawler bulldozer includes:

[0137] According to the change of the cosine of the included angle within the preset time period, the force condition in the tipping direction is determined; wherein the force condition is: the force in the tipping direction predicted according to the change of the cosine of the included angle;

[0138] According to the force conditions, the body parameters of the target crawler bulldozer in the dumping direction and the current driving ground conditions, the dumping of the target crawler bulldozer is simulated to obtain the simulated dumping result; wherein the body parameters are: the body parameters of the target crawler bulldozer, such as: vehicle length, vehicle width and vehicle height; the simulated dumping result is: the passing area of ​​the simulated target crawler bulldozer when dumping in the simulation model;

[0139] According to the simulated dumping result, the scanning range of the laser radar is determined; wherein the scanning range is: the passing area of ​​the target crawler bulldozer simulated in the simulated dumping model when dumping corresponds to the actual working area of ​​the target crawler bulldozer;

[0140] Control the laser radar to obtain radar data within the scanning range.

[0141] The working principle and beneficial effects of the above technical solution are:

[0142] When the target crawler bulldozer rolls over, the possible rollover scenarios may be different depending on the situation during the rollover, such as direct rollover or rolling over for several rounds and then stopping. This results in different areas passed by the target crawler bulldozer when rolling over. Directly scanning a large area around the target crawler bulldozer and then issuing an early warning is inefficient. Therefore, according to the change of the cosine of the angle within a preset time period, the force condition in the dumping direction is determined. According to the force condition, the body parameters of the target crawler bulldozer in the dumping direction and the current driving ground conditions, the dumping of the target crawler bulldozer is simulated. The area passed by the target crawler bulldozer when dumping simulated in the simulated dumping model corresponds to the actual working area of ​​the target crawler bulldozer and is used as the scanning range. The laser radar is controlled to obtain radar data within the scanning range for early warning, thereby improving the reminder efficiency.

[0143] The embodiment of the present invention provides a crawler bulldozer steering control system based on agricultural big data, such as Figure 2 As shown, including:

[0144] The operation attribute feature extraction subsystem 1 is used to obtain the operation attributes of the target crawler bulldozer and extract the operation attribute features;

[0145] Agricultural big data acquisition subsystem 2, used to acquire agricultural big data suitable for training according to operation attribute characteristics;

[0146] The target simulation operation process set determination subsystem 3 is used to establish a simulation model based on agricultural big data, set control parameters of the target crawler bulldozer in the simulation model, and determine the target simulation operation process set;

[0147] The steering control subsystem 4 is used to determine the actual control parameters of the target crawler bulldozer according to the target simulated operation process set and the actual operation process, and perform steering control.

[0148] In one embodiment, the job attribute feature extraction subsystem includes:

[0149] A job attribute feature extraction standard determination module is used to determine the job attribute feature extraction standard according to the job attributes;

[0150] The feature extraction module is used to extract the operation attribute features according to the operation attribute feature extraction standard; the operation attribute features include: steering torque, track gauge, steering angular velocity and speed on both sides of the track.

[0151] In one embodiment, the agricultural big data acquisition subsystem includes:

[0152] A first agricultural operation scene determination module is used to determine a first agricultural operation scene of a target crawler bulldozer; the first agricultural operation scene is: a historical agricultural operation scene of the target crawler bulldozer;

[0153] A first target data determination module, used to determine first target data according to a first agricultural operation scenario;

[0154] A second agricultural operation scene determination module, used to determine a template according to operation attribute characteristics and the operation scene, and determine a second agricultural operation scene;

[0155] A second target data determination module, used to determine second target data based on the agricultural big data platform and according to a second agricultural operation scenario;

[0156] The agricultural big data determination module is used to use the first target data and the second target data together as agricultural big data suitable for training.

[0157] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A crawler bulldozer steering control method based on agricultural big data, characterized in that: include: Step 1: Obtain the operating attributes of the target crawler bulldozer and extract the operating attribute features; Step 2: Obtain agricultural big data suitable for training based on the operation attribute characteristics; Step 2: Obtain agricultural big data suitable for training based on the operation attribute characteristics, including: Determine a first agricultural operation scene of the target crawler bulldozer; the first agricultural operation scene is: a historical agricultural operation scene of the target crawler bulldozer; Determine first target data according to the first agricultural operation scenario; Determine the template according to the operation attribute characteristics and the operation scene, and determine the second agricultural operation scene; Based on the agricultural big data platform, the second target data is determined according to the second agricultural operation scenario; The first target data and the second target data are used together as agricultural big data suitable for training; Based on the agricultural big data platform, the second target data is determined according to the second agricultural operation scenario, including: Based on the agricultural big data platform, according to the second agricultural operation scenario, the third target data is determined; the third target data is: the verified data of the agricultural big data platform; Obtaining an operating posture set of a historical operating crawler bulldozer in a second agricultural operation scene; According to the operation posture set, the duration ratio of the steering posture duration to the non-steering posture duration of each historical operation crawler bulldozer is calculated; If the duration ratio is greater than or equal to a preset duration ratio threshold, determining the second target data according to the bulldozing trajectory of the crawler bulldozer of the corresponding historical operation and the third target data; Step 3: Establish a simulation model based on agricultural big data, set the control parameters of the target crawler bulldozer in the simulation model, and determine the target simulation operation process set; Step 4: According to the target simulated operation process set and the actual operation process, determine the actual control parameters of the target crawler bulldozer and perform steering control.

2. The crawler bulldozer steering control method based on agricultural big data according to claim 1, characterized in that: Step 1: Obtain the operating attributes of the target crawler bulldozer and extract the operating attribute features, including: According to the job attributes, determine the job attribute feature extraction standard; The operation attribute features are extracted according to the operation attribute feature extraction standard; the operation attribute features include: steering torque, track gauge, steering angular velocity and speed on both sides of the track.

3. The crawler bulldozer steering control method based on agricultural big data according to claim 1, characterized in that: Also includes: When the steering control of the target crawler bulldozer is performed, the tipping detection of the target crawler bulldozer is performed, and the balance control is performed according to the tipping detection result.

4. The crawler bulldozer steering control method based on agricultural big data as claimed in claim 3, characterized in that: When the steering control of the target crawler bulldozer is performed, the dumping detection of the target crawler bulldozer is performed, and the balance control is performed according to the dumping detection result, including: Obtaining a first gravity acceleration vector of the straight-ahead posture of the target crawler bulldozer at a target time before performing steering control of the target crawler bulldozer, and marking it in a three-dimensional model diagram of the target crawler bulldozer; Obtaining a second gravity acceleration vector of the target crawler bulldozer in a steering posture, and marking it in a three-dimensional model diagram of the target crawler bulldozer; According to the three-dimensional model diagram, the cosine of the angle between the first gravity acceleration vector and the second gravity acceleration vector is calculated; According to the cosine of the included angle, the moving direction and moving distance of the balancing slider preset in the target crawler bulldozer are determined; Control the balance slider to move in the moving direction and distance; When the moving distance reaches the maximum distance of the balance slider in the moving direction, the change of the cosine of the angle within the preset time period is determined. If the cosine of the angle still decreases, the radar data around the target crawler bulldozer is obtained; Based on the radar data, the on-site reminder personnel are determined and voice reminders are given.

5. The crawler bulldozer steering control method based on agricultural big data according to claim 4, characterized in that: Acquire radar data around the target crawler bulldozer, including: Determine the force condition in the tipping direction according to the change of the cosine of the angle within a preset time period; According to the force conditions, the body parameters of the target crawler bulldozer in the dumping direction and the current driving ground conditions, the dumping of the target crawler bulldozer is simulated to obtain the simulated dumping result; Determine the scanning range of the laser radar based on the simulated dumping results; Control the laser radar to obtain radar data within the scanning range.

6. The crawler bulldozer steering control system based on agricultural big data is characterized by: include: The operation attribute feature extraction subsystem is used to obtain the operation attributes of the target crawler bulldozer and extract the operation attribute features; The agricultural big data acquisition subsystem is used to obtain agricultural big data suitable for training based on the operation attribute characteristics and perform the following operations: Determine a first agricultural operation scene of the target crawler bulldozer; the first agricultural operation scene is: a historical agricultural operation scene of the target crawler bulldozer; Determine first target data according to the first agricultural operation scenario; Determine the template according to the operation attribute characteristics and the operation scene, and determine the second agricultural operation scene; Based on the agricultural big data platform, the second target data is determined according to the second agricultural operation scenario; The first target data and the second target data are used together as agricultural big data suitable for training; Based on the agricultural big data platform, the second target data is determined according to the second agricultural operation scenario, including: Based on the agricultural big data platform, according to the second agricultural operation scenario, the third target data is determined; the third target data is: the verified data of the agricultural big data platform; Obtaining an operating posture set of a historical operating crawler bulldozer in a second agricultural operation scene; According to the operation posture set, the duration ratio of the steering posture duration to the non-steering posture duration of each historical operation crawler bulldozer is calculated; If the duration ratio is greater than or equal to a preset duration ratio threshold, determining the second target data according to the bulldozing trajectory of the crawler bulldozer of the corresponding historical operation and the third target data; The target simulation operation process set determination subsystem is used to establish a simulation model based on agricultural big data, set the control parameters of the target crawler bulldozer in the simulation model, and determine the target simulation operation process set; The steering control subsystem is used to determine the actual control parameters of the target crawler bulldozer according to the target simulated operation process set and the actual operation process, and perform steering control.

7. The crawler bulldozer steering control system based on agricultural big data according to claim 6, characterized in that: The job attribute feature extraction subsystem includes: A job attribute feature extraction standard determination module is used to determine the job attribute feature extraction standard according to the job attributes; The feature extraction module is used to extract the operation attribute features according to the operation attribute feature extraction standard; the operation attribute features include: steering torque, track gauge, steering angular velocity and speed on both sides of the track.

Images