Trapped vehicle prediction method and device for agricultural machinery and storage medium

Abstract

The embodiment of the invention provides a trapped vehicle prediction method and device for agricultural machinery and a storage medium. The method comprises the following steps: acquiring a historical operation data set of the agricultural machinery under different working conditions, wherein the historical operation data set comprises at least one of historical attitude data, historical positioning data, historical vehicle body control data and historical earth surface monitoring data; determining historical operation fluctuation data of the agricultural machinery under each operation condition according to the historical vehicle body control data under each operation condition; according to the historical operation data sets and the historical operation fluctuation data under the different operation conditions and the historical advancing state parameters of the agricultural machinery under the different operation conditions, an agricultural machinery sinking prediction model is constructed, and the agricultural machinery sinking prediction model is used for predicting whether the agricultural machinery sinks in the corresponding driving front area or not; the agricultural machinery trapping prediction model with higher prediction precision is established, the efficiency and accuracy of subsequent agricultural machinery trapping prediction are improved, and the operation of the agricultural machinery is safer.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of agricultural machinery, in particular to a vehicle sinking prediction method and device for agricultural machinery, a storage medium and agricultural machinery. BACKGROUND

[0002] When the agricultural machinery is working in the field, it is easy to sink in the mud and hard ground with ups and downs, which greatly affects the working efficiency of the agricultural machinery and also greatly reduces the driving safety of the agricultural machinery. In the prior art, a single vehicle body posture information or a front wheel speed difference is used to predict whether the agricultural machinery sinks, and the efficiency and accuracy of the prediction are low by using this method to determine whether the agricultural machinery sinks, which affects the working safety of the agricultural machinery. SUMMARY

[0003] The purpose of the embodiments of the present application is to provide a vehicle sinking prediction method and device for agricultural machinery, a storage medium and agricultural machinery, so as to solve the problem of slow and inaccurate prediction of agricultural machinery sinking in the prior art.

[0004] In order to achieve the above-mentioned purpose, the first aspect of the present application provides a vehicle sinking prediction method for agricultural machinery, comprising:

[0005] Obtaining a historical running data set of the agricultural machinery under different working conditions, the historical running data set comprising at least one of the following: historical posture data, historical positioning data, historical vehicle body control data and historical ground surface monitoring data;

[0006] Determining historical running fluctuation data of the agricultural machinery under each working condition according to the historical vehicle body control data under each working condition;

[0007] Constructing a tractor sinking prediction model according to the historical running data set, the historical running fluctuation data and the historical running state parameters of the agricultural machinery under different working conditions, wherein the tractor sinking prediction model is used to predict whether the agricultural machinery sinks in the corresponding driving front area.

[0008] In the embodiments of the present application, constructing the tractor sinking prediction model according to the historical running data set, the historical running fluctuation data and the historical running state parameters of the agricultural machinery under different working conditions comprises: determining the covariance coefficients between the historical running data set and the historical running fluctuation data and the historical running state parameters under the corresponding working conditions; based on the covariance coefficients, screening a target training data set from the historical running data set and the historical running fluctuation data; inputting the target training data set into a preset neural network, so that the preset neural network outputs the corresponding historical running state parameters, so as to construct the tractor sinking prediction model.

[0009] In the embodiment of the present application, the target training data set is selected from the historical running data set and the historical running fluctuation data based on the covariance coefficients, including: taking all historical data corresponding to the first N covariance coefficients with the largest values in all covariance coefficients as the qualified historical data, wherein N is a positive integer; and constructing the target training data set according to the qualified historical data.

[0010] In the embodiment of the present application, the historical vehicle body control data includes the vehicle body speed, the motor current and the motor torque of the agricultural machine during driving, and the historical running fluctuation data includes the speed fluctuation, the current fluctuation and the torque fluctuation of the agricultural machine during driving; and the historical running fluctuation data of the agricultural machine under each working condition is determined according to the historical vehicle body control data under each working condition, including: determining the speed fluctuation of the agricultural machine under each working condition according to the vehicle body speed of the agricultural machine; determining the current fluctuation of the agricultural machine under each working condition according to the motor current of the agricultural machine; and determining the torque fluctuation of the agricultural machine under each working condition according to the motor torque of the agricultural machine.

[0011] In the embodiment of the present application, the vehicle sinking prediction method further includes: obtaining a current running data set of the agricultural machine under the current working condition; inputting the current running data set into the agricultural machine vehicle sinking prediction model to obtain a travel state prediction parameter of the agricultural machine in the corresponding driving front area; and predicting whether the agricultural machine sinks in the driving front area according to the travel state prediction parameter.

[0012] In the embodiment of the present application, predicting whether the agricultural machine sinks in the driving front area according to the travel state prediction parameter includes: in the case that the travel state prediction parameter is a first value, determining that the agricultural machine will not sink in the driving front area; and in the case that the travel state prediction parameter is a second value or a third value, determining that the agricultural machine will sink in the driving front area.

[0013] In the embodiment of the present application, the vehicle sinking prediction method further includes: in the case that it is determined that the agricultural machine does not sink in the driving front area, controlling the agricultural machine to enter a high-speed working mode; in the case that it is determined that the agricultural machine sinks in the driving front area, determining a sinking level of the agricultural machine, the sinking level including a first level and a second level, and the sinking degree corresponding to the second level being higher than the sinking degree corresponding to the first level; in the case that the sinking level is the first level, controlling the agricultural machine to enter an anti-sinking gear shifting mode to enable the agricultural machine to pass through the corresponding sinking area; and in the case that the sinking level is the second level, controlling the agricultural machine to enter a shutdown mode to control the agricultural machine to stop working.

[0014] In the embodiment of the present application, the stuck vehicle prediction method further comprises: in the case that it is determined that the agricultural machine is not stuck in the area in front of the driving area, not sending a warning prompt and an alarm prompt; in the case that it is determined that the agricultural machine is stuck in the area in front of the driving area and the stuck vehicle level is the first level, sending an alarm prompt corresponding to the first level; in the case that it is determined that the agricultural machine is stuck in the area in front of the driving area and the stuck vehicle level is the second level, sending an alarm prompt corresponding to the second level.

[0015] The second aspect of the present application provides a stuck vehicle prediction device for an agricultural machine, comprising:

[0016] a memory configured to store instructions; and

[0017] a processor configured to call the instructions from the memory and capable of realizing the above-mentioned stuck vehicle prediction method for an agricultural machine when executing the instructions.

[0018] The third aspect of the present application provides a machine-readable storage medium, which stores instructions thereon, the instructions, when executed by a processor, cause the processor to be configured to execute the above-mentioned stuck vehicle prediction method for an agricultural machine.

[0019] Through the above technical solution, the attitude data, positioning data, vehicle body control data and ground surface monitoring data of the agricultural machine are multi-source fused, and the influence of the running fluctuation data of the agricultural machine in the running process on the running state parameters is considered, so that a stuck vehicle prediction model with higher prediction accuracy is established, the efficiency and accuracy of subsequent stuck vehicle prediction of the agricultural machine are improved, and the operation of the agricultural machine is safer.

[0020] Other features and advantages of the embodiments of the present application will be described in detail in the following specific implementation part. BRIEF DESCRIPTION OF DRAWINGS

[0021] The accompanying drawings are included to provide a further understanding of the embodiments of the present application, and constitute a part of the specification, and are used together with the following specific implementation to explain the embodiments of the present application, but do not constitute a limitation on the embodiments of the present application. In the drawings:

[0022] Figure 1 The flowchart of the stuck vehicle prediction method for an agricultural machine according to the embodiments of the present application is schematically shown;

[0023] Figure 2 The structural block diagram of the stuck vehicle prediction device for an agricultural machine according to the embodiments of the present application is schematically shown;

[0024] Figure 3 The internal structure diagram of the computer device according to the embodiments of the present application is schematically shown. DETAILED DESCRIPTION

[0025] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. It should be understood that the specific implementation described herein is only used to explain and illustrate the embodiments of the present application and should not be used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0026] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative positional relationship, movement condition, etc. between components in a certain posture (as shown in the drawings), and if the certain posture changes, the directional indications also change accordingly.

[0027] In addition, if the embodiments of the present application involve descriptions such as “first”, “second”, etc., the descriptions of “first”, “second”, etc. are only for description purposes and should not be understood as indicating or implying the relative importance of the indicated technical features or implicitly indicating the number of the indicated technical features. Therefore, the features limited by “first”, “second” can explicitly or implicitly include at least one of the features. In addition, the technical solutions of each embodiment can be combined with each other, but it must be based on the fact that a person of ordinary skill in the art can realize it, and when the combination of technical solutions contradicts each other or cannot be realized, it should be considered that the combination of technical solutions does not exist and is not within the scope of protection claimed by the present application.

[0028] Figure 1 A flowchart of a vehicle sinking prediction method for an agricultural machine according to an embodiment of the present application is schematically shown. As shown in Figure 1 In an embodiment of the present application, a vehicle sinking prediction method for an agricultural machine is provided, comprising the following steps:

[0029] Step 101: Obtain a historical running data set of the agricultural machine under different working conditions, and the historical running data set comprises at least one of the following: historical attitude data, historical positioning data, historical vehicle body control data and historical ground monitoring data.

[0030] An agricultural machine refers to a machine used in the process of performing agricultural work. The agricultural machine can include a farm power machine, a farmland construction machine, a soil tillage machine, a planting and fertilizing machine, a plant protection machine, a farmland irrigation and drainage machine, a crop harvesting machine, an agricultural product processing machine, a livestock industry machine and an agricultural transportation machine. The agricultural machine can sink during the process of performing agricultural work. Therefore, the agricultural machine can be subjected to vehicle sinking prediction.

[0031] In the agricultural machine trap prediction, historical operation data sets of the agricultural machine under different working conditions can be obtained, and the historical operation data sets include at least one of historical attitude data, historical positioning data, historical vehicle body control data, and historical ground monitoring data. The working conditions include working conditions and control modes. The working conditions can include muddy ground and hard ground, etc. The control modes can include anti-trap mode, working mode, and neutral mode.

[0032] The different working conditions can correspond to different historical attitude data, historical positioning data, historical vehicle body control data, and historical ground monitoring data. The different historical attitude data, historical positioning data, historical vehicle body control data, and historical ground monitoring data can correspond to different working conditions. The historical attitude data can reflect the attitude of the agricultural machine under the corresponding working condition, the historical positioning data can reflect the position of the agricultural machine under the corresponding working condition, the historical vehicle body control data can reflect the control state of the agricultural machine under the corresponding working condition, and the historical ground monitoring data can reflect the ground information of the area in front of the agricultural machine under the corresponding working condition, for example, whether the area in front of the agricultural machine is concave or non-concave.

[0033] In the embodiments of the present application, the agricultural machine includes a vehicle body attitude prediction device, and the vehicle body attitude prediction device includes an inertial measurement unit (IMU). The trap prediction method further includes: obtaining historical attitude data by the inertial measurement unit (IMU), wherein the historical attitude data includes at least one of a vehicle body position, a vehicle body roll angle, a vehicle body yaw angle, and a vehicle body pitch angle of the agricultural machine.

[0034] The agricultural machine can include a vehicle body attitude prediction device. The vehicle body attitude prediction device can include an inertial measurement unit (IMU). The inertial measurement unit (IMU) is used to predict the historical attitude information of the agricultural machine during the agricultural operation or driving of the agricultural machine. The historical attitude data includes at least one of the vehicle body position, the vehicle body roll angle, the vehicle body yaw angle, and the vehicle body pitch angle of the agricultural machine. The vehicle body position can be defined by the X direction / Y direction / Z direction coordinates of the vehicle body.

[0035] The installation position of the vehicle body attitude prediction device can be customized according to the actual situation. For example, the vehicle body attitude prediction device can be installed at a position convenient for predicting the vehicle body roll angle, the vehicle body yaw angle, and the vehicle body pitch angle. During the agricultural operation or driving of the agricultural machine, the vehicle body attitude prediction device can predict at least one of the vehicle body position, the vehicle body roll angle, the vehicle body yaw angle, and the vehicle body pitch angle of the agricultural machine, and can send the predicted attitude data to the processor, so that the processor can obtain the attitude data sent by the vehicle body attitude prediction device.

[0036] In the embodiment of the present application, the agricultural machine comprises a vehicle body positioning prediction device, and the vehicle body positioning prediction device comprises an RTK base station. The method further comprises: acquiring historical positioning data of the agricultural machine by the RTK base station, wherein the historical positioning data comprises at least one of a latitude value, a longitude value, an elevation value, a travel speed, an elevation difference and a position deviation of the agricultural machine.

[0037] The agricultural machine can comprise a vehicle body positioning prediction device. The vehicle body positioning prediction device can comprise an RTK (Real-time kinematic) base station, which can be used to predict historical positioning data of the agricultural machine during the execution of agricultural work or travel of the agricultural machine. The historical positioning data comprises at least one of a latitude value, a longitude value, an elevation value, a travel speed, an elevation difference and a position deviation of the agricultural machine.

[0038] The installation position of the vehicle body positioning prediction device can be customized according to actual conditions. During the execution of agricultural work or travel of the agricultural machine, the vehicle body positioning prediction device can predict at least one of a latitude value, a longitude value, an elevation value, a travel speed, an elevation difference and a position deviation of the agricultural machine, and can send the predicted positioning data to a processor, so that the processor can acquire the positioning data sent by the vehicle body positioning prediction device.

[0039] In the embodiment of the present application, the agricultural machine comprises a vehicle body control prediction sensor, and the vehicle body control prediction sensor comprises a rotation speed sensor, a current sensor and a torque sensor. The method further comprises: acquiring a vehicle body rotation speed of the agricultural machine included in historical vehicle body control data by the rotation speed sensor; acquiring a motor current of the agricultural machine included in the historical vehicle body control data by the current sensor; and acquiring a motor torque of the agricultural machine included in the historical vehicle body control data by the torque sensor.

[0040] The agricultural machine can comprise a vehicle body control prediction sensor, and the vehicle body control prediction sensor can comprise a rotation speed sensor, a current sensor and a torque sensor. The rotation speed sensor can be used to predict a vehicle body rotation speed of the agricultural machine during the execution of agricultural work or travel of the agricultural machine, the current sensor can be used to predict a motor current of the agricultural machine during the execution of agricultural work or travel of the agricultural machine, and the torque sensor can be used to predict a motor torque of the agricultural machine during the execution of agricultural work or travel of the agricultural machine.

[0041] The installation positions of the rotation speed sensor, the current sensor, and the torque sensor can be customized according to actual conditions. During the agricultural machine performing the agricultural work or traveling, the rotation speed sensor, the current sensor, and the torque sensor can respectively predict the vehicle body rotation speed, the motor current, and the motor torque of the agricultural machine, and can respectively send the predicted vehicle body rotation speed, motor current, and motor torque to the processor. The processor can respectively acquire the vehicle body rotation speed, motor current, and motor torque sent by the rotation speed sensor, the current sensor, and the torque sensor.

[0042] In the embodiment of the present application, the agricultural machine comprises a ground surface monitoring sensor, and the ground surface monitoring sensor comprises a sonar sensor. The method further comprises: acquiring historical ground surface monitoring data by the sonar sensor, wherein the historical ground surface monitoring data comprises the ground flatness of the area in front of the agricultural machine.

[0043] The agricultural machine can comprise a ground surface monitoring sensor, and the ground surface monitoring sensor can comprise a sonar sensor. The sonar sensor can be used to predict the historical ground surface monitoring data of the agricultural machine during the agricultural machine performing the agricultural work or traveling. The historical ground surface monitoring data comprises the ground flatness of the area in front of the agricultural machine.

[0044] The installation positions of the sonar sensors can be customized according to actual conditions. During the agricultural machine performing the agricultural work or traveling, the sonar sensor can predict the ground flatness of the area in front of the agricultural machine, and can send the predicted ground flatness of the area in front of the agricultural machine to the processor. The processor can acquire the ground flatness of the area in front of the agricultural machine sent by the sonar sensor.

[0045] Step 102: determining the historical running fluctuation data of the agricultural machine in each work condition according to the historical vehicle body control data in each work condition.

[0046] The processor can determine the historical running fluctuation data of the agricultural machine in each work condition according to the historical vehicle body control data in each work condition. The historical vehicle body control data can reflect the control state of the agricultural machine in the corresponding work condition, and the historical running fluctuation data can reflect the fluctuation change of the control parameter of the agricultural machine during traveling.

[0047] In the embodiments of the present application, the historical vehicle body control data includes the vehicle body speed, the motor current and the motor torque of the agricultural machine during driving, and the historical running fluctuation data includes the speed fluctuation, the current fluctuation and the torque fluctuation of the agricultural machine during driving; the historical running fluctuation data of the agricultural machine under each working condition is determined according to the historical vehicle body control data under each working condition, including: the speed fluctuation of the agricultural machine under each working condition is determined according to the vehicle body speed of the agricultural machine; the current fluctuation of the agricultural machine under each working condition is determined according to the motor current of the agricultural machine; and the torque fluctuation of the agricultural machine under each working condition is determined according to the motor torque of the agricultural machine.

[0048] The historical vehicle body control data can include the vehicle body speed, the motor current and the motor torque of the agricultural machine. The historical running fluctuation data can reflect the fluctuation changes of the control parameters of the agricultural machine during driving. The historical running fluctuation data can include the speed fluctuation, the current fluctuation and the torque fluctuation of the agricultural machine during driving. Specifically, the processor can determine the speed fluctuation, the current fluctuation and the torque fluctuation of the agricultural machine during driving according to the vehicle body speed, the motor current and the motor torque of the agricultural machine, respectively.

[0049] Step 103: constructing a tractor stuck prediction model according to the historical running data set under different working conditions, the historical running fluctuation data and the historical running state parameters of the agricultural machine under different working conditions, wherein the tractor stuck prediction model is used to predict whether the agricultural machine is stuck in the corresponding driving front area.

[0050] The processor can construct a tractor stuck prediction model according to the historical attitude data, the historical positioning data, the historical running fluctuation data, the historical ground monitoring data and the historical running state parameters of the agricultural machine under different working conditions. The historical running state of the agricultural machine includes normal state, slight stuck state and serious stuck state. When the agricultural machine is in the slight stuck state, it means that the agricultural machine may be slightly stuck when driving to the driving front area, but the degree of stuck does not reach the preset degree. When the agricultural machine is in the serious stuck state, it means that the agricultural machine may be seriously stuck when driving to the driving front area, and the degree of stuck reaches the preset degree. The preset degree can be customized according to actual conditions. The historical running state parameters include parameters corresponding to the normal state, parameters corresponding to the slight stuck state and parameters corresponding to the serious stuck state, which can be customized according to requirements. For example, the parameters corresponding to the normal state can be set to 0, the parameters corresponding to the slight stuck state can be set to 1, and the parameters corresponding to the serious stuck state can be set to 2.

[0051] In the embodiment of the present application, the construction of the agricultural machinery stuck prediction model according to the historical running data set, the historical running fluctuation data and the historical running state parameters of the agricultural machinery under different working conditions comprises: determining the covariance coefficients between the historical running data set and the historical running fluctuation data and the historical running state parameters corresponding to the working conditions; screening a target training data set from the historical running data set and the historical running fluctuation data based on the covariance coefficients; inputting the target training data set into a preset neural network to make the preset neural network output corresponding historical running state parameters, so as to construct the agricultural machinery stuck prediction model.

[0052] The processor can determine the covariance coefficients between the historical running data set and the historical running fluctuation data and the historical running state parameters corresponding to the working conditions. If the historical running data set comprises historical attitude data, historical positioning data, historical vehicle body control data and historical ground monitoring data, the covariance coefficients between all the historical attitude data, all the historical positioning data, all the historical vehicle body control data, all the historical running fluctuation data and all the historical ground monitoring data and corresponding historical running state parameters can be determined. That is, a group of historical attitude data, historical positioning data, historical vehicle body control data, historical running fluctuation data and historical ground monitoring data can correspond to a historical running state parameter. The covariance coefficient for the historical attitude data can be determined according to all the historical attitude data and all the historical running state parameters, the covariance coefficient for the historical positioning data can be determined according to all the historical positioning data and all the historical running state parameters, the covariance coefficient for the historical vehicle body control data can be determined according to all the historical vehicle body control data and all the historical running state parameters, the covariance coefficient for the historical running fluctuation data can be determined according to all the historical running fluctuation data and all the historical running state parameters, and the covariance coefficient for the historical ground monitoring data can be determined according to all the historical ground monitoring data and all the historical running state parameters. The processor can screen a target training data set from the historical running data set and the historical running fluctuation data based on all the covariance coefficients.

[0053] In the embodiment of the present application, screening a target training data set from the historical running data set and the historical running fluctuation data based on the covariance coefficients comprises: taking all the historical data corresponding to the first N covariance coefficients with the largest values in all the covariance coefficients as qualified historical data, wherein N is a positive integer; and constructing a target training data set according to the qualified historical data.

[0054] The greater the covariance coefficient is, the stronger the correlation between the corresponding historical data and the historical travel state parameter is, and vice versa. The processor can take all the historical data corresponding to the first N covariance coefficients with the largest values in all the covariance coefficients as the historical data meeting the condition, where N is a positive integer. N can be self-defined according to actual conditions, for example, N can be 8.

[0055] In one embodiment, the processor can arrange all the covariance coefficients in descending order to obtain the arrangement order of all the covariance coefficients. The processor can take all the historical data corresponding to the covariance coefficients arranged before a preset value in the arrangement order as the historical data meeting the condition. The preset value can be set according to requirements. For example, the preset value can be set to 8. The processor can construct the target training data set according to the historical data meeting the condition.

[0056] For example, the historical attitude data includes a vehicle body roll angle, a vehicle body yaw angle and a vehicle body pitch angle, the historical positioning data includes a travel speed, an elevation difference and a position deviation, the historical vehicle body control data includes a vehicle body rotation speed, a motor current and a motor torque, the historical running fluctuation data includes a rotation speed fluctuation amount, a current fluctuation amount and a torque fluctuation amount, and the historical ground surface monitoring data includes a ground flatness. At this time, the number of determined covariance coefficients is 13. Assuming that the first 8 covariance coefficients are to be selected, the 13 covariance coefficients are arranged in descending order, the first 8 covariance coefficients are a covariance coefficient for the vehicle body roll angle, a covariance coefficient for the yaw angle, a covariance coefficient for the vehicle body pitch angle, a covariance coefficient for the travel speed, a covariance coefficient for the rotation speed fluctuation amount, a covariance coefficient for the current fluctuation amount, a covariance coefficient for the torque fluctuation amount and a covariance coefficient for the ground flatness, and then it can be determined that all the vehicle body roll angles, all the yaw angles, all the vehicle body pitch angles, all the travel speeds, all the rotation speed fluctuation amounts, all the current fluctuation amounts, all the torque fluctuation amounts and all the ground flatnesses are the historical data meeting the condition, so as to construct the corresponding target training data set.

[0057] The processor can input the target training data set into the preset neural network, so that the preset neural network outputs corresponding historical running state parameters, to construct the agricultural machinery getting-stuck prediction model. In an embodiment, the preset neural network can be a neural network constructed based on different deep learning frameworks such as PyTorch, Tensorflow, Caffe, Keras, etc. The target training data set includes multiple groups of training data, each group of training data including historical data meeting preset conditions, each group of training data can be used as input layer data of the preset neural network, and CNN convolution operation can be performed on the input layer data to extract convolution features of the input layer data, thereby reducing the complexity of modeling of the preset neural network. Then, the convolution features of the preset neural network can be input into the average value pooling layer of the preset neural network, so as to extract weight coefficients between the convolution features in the preset neural network, to suppress overfitting of the preset neural network. Then, the convolution features and the weight coefficients in the preset neural network can be input into the full connection layer of the preset neural network, and the historical running state parameters corresponding to the training data are used as output labels of the full connection layer, so as to train the preset neural network.

[0058] When the training completion condition of the preset neural network is met, the processor can determine the trained preset neural network as the agricultural machinery getting-stuck prediction model. The training completion condition can be set according to requirements, for example, the training completion condition can be set based on at least one of prediction accuracy, preset error, and iteration number. If the training completion condition is set based on the preset accuracy, it can specifically be that the preset accuracy reaches a preset accuracy range. If the training completion condition is set based on the preset error, it can specifically be that the preset error reaches a preset error range. If the training completion condition is set based on the iteration number, it can specifically be that the iteration number reaches a preset number.

[0059] In the embodiments of the present application, the getting-stuck prediction method further includes: obtaining a current running data set of the agricultural machinery under a current working condition; inputting the current running data set into the agricultural machinery getting-stuck prediction model to obtain a running state prediction parameter of the agricultural machinery in a corresponding front driving area; and predicting whether the agricultural machinery gets stuck in the front driving area according to the running state prediction parameter.

[0060] In the case of obtaining the agricultural machinery stuck prediction model, if the agricultural machinery is in any one working condition, the current running data set of the agricultural machinery can be obtained. The current running data can be obtained through the corresponding sensors installed on the agricultural machinery. The current running data set can be obtained based on the types of the above-mentioned qualified historical data. For example, if the types of the qualified historical data include the vehicle roll angle, the vehicle deflection angle, the vehicle pitch angle, the travel speed, the speed fluctuation, the current fluctuation, the torque fluctuation and the ground flatness, the current running data of the agricultural machinery can include the current vehicle roll angle, the current deflection angle, the current vehicle pitch angle, the current travel speed, the current speed fluctuation, the current current fluctuation, the current torque fluctuation and the current ground flatness of the agricultural machinery. The current running data set of the agricultural machinery can be obtained through the corresponding sensors installed on the agricultural machinery.

[0061] The processor can input the current running data set of the agricultural machinery in the current working condition to the agricultural machinery stuck prediction model to obtain the travel state prediction parameter of the corresponding travel front area of the agricultural machinery. The travel state prediction parameter can reflect the travel state of the agricultural machinery in the corresponding travel front area. The processor can predict whether the agricultural machinery is stuck in the travel front area according to the travel state prediction parameter of the agricultural machinery.

[0062] In the embodiments of the present application, predicting whether the agricultural machinery is stuck in the travel front area according to the travel state prediction parameter includes: in the case that the travel state prediction parameter is a first value, determining that the agricultural machinery will not be stuck in the travel front area; in the case that the travel state prediction parameter is a second value or a third value, determining that the agricultural machinery will be stuck in the travel front area.

[0063] In the case that the travel state prediction parameter of the agricultural machinery is the first value, the agricultural machinery may not be in the stuck area at this time, and the processor can determine that the agricultural machinery may not be stuck in the travel front area. In the case that the travel state prediction parameter of the agricultural machinery is the second value or the third value, the processor can determine that the agricultural machinery may have a certain risk of being stuck in the travel front area. The first value, the second value and the third value can be customized according to requirements, for example, the first value can be 0, the second value can be 1, and the third value can be 2.

[0064] In the embodiment of the present application, the vehicle trapping prediction method further comprises: in the case that it is determined that the agricultural machine is not trapped in the area in front of the driving area, controlling the agricultural machine to enter a high-speed operation mode; in the case that it is determined that the agricultural machine is trapped in the area in front of the driving area, determining a trapping level of the agricultural machine in the area in front of the driving area, the trapping level comprising a first level and a second level, the trapping degree corresponding to the second level being higher than the trapping degree corresponding to the first level; in the case that the trapping level is the first level, controlling the agricultural machine to enter a trapping prevention gear mode to enable the agricultural machine to pass through the corresponding trapping area; and in the case that the trapping level is the second level, controlling the agricultural machine to enter a shutdown mode to control the agricultural machine to stop operation.

[0065] In the case that it is determined that the agricultural machine is not trapped in the area in front of the driving area, the processor can control the agricultural machine to enter a high-speed operation mode. The high-speed operation mode refers to operation in the field at a relatively large travel speed. At this time, the agricultural machine is in a non-trapped state, and the travel speed and the rotating speed can be increased to improve the operation efficiency of the agricultural machine.

[0066] In the case that it is determined that the agricultural machine is trapped in the area in front of the driving area, the processor can determine a trapping level of the agricultural machine, the trapping level comprising a first level. Specifically, if the travel state prediction parameter of the agricultural machine is the second value, at this time, the agricultural machine can be trapped in the area in front of the driving area, but the trapping of the agricultural machine is not serious, i.e., the trapping degree of the agricultural machine in the area in front of the driving area can not reach a preset degree, and the processor can determine that the trapping level of the agricultural machine is the first level.

[0067] In the case that the trapping level is the first level, the processor can control the agricultural machine to enter a trapping prevention gear mode to enable the agricultural machine to pass through the corresponding trapping area. The trapping prevention gear mode can be used to prevent the agricultural machine from being trapped more seriously. In the trapping prevention gear mode, the torque and the rotating speed can be adjusted to enable the agricultural machine to pass through the trapping area quickly. The torque and the rotating speed can be set according to actual needs, for example, the torque can be set as large as possible, and the rotating speed can be set as small as possible, so that the agricultural machine will not sink deeper during movement.

[0068] In a case where it is determined that the agricultural machine is stuck in the area in front of the traveling area, the processor can determine a stuck level of the agricultural machine, the stuck level including a second level, the stuck level corresponding to a stuck degree higher than a stuck degree corresponding to the first level. Specifically, if the traveling state prediction parameter of the agricultural machine is the third value, at this time, the agricultural machine can be stuck in the area in front of the traveling area, but the agricultural machine can be stuck more seriously, that is, the stuck degree of the agricultural machine in the area in front of the traveling area can reach a preset degree, and the processor can determine that the stuck level of the agricultural machine is the second level. In a case where the stuck level is the second level, the processor can control the agricultural machine to enter a stop mode controlled by the processor to control the agricultural machine to stop working.

[0069] In the embodiments of the present application, the stuck prediction method further includes: in a case where it is determined that the agricultural machine is not stuck in the area in front of the traveling area, not sending a warning prompt and an alarm prompt; in a case where it is determined that the agricultural machine is stuck in the area in front of the traveling area and the stuck level is the first level, sending an alarm prompt corresponding to the first level; and in a case where it is determined that the agricultural machine is stuck in the area in front of the traveling area and the stuck level is the second level, sending an alarm prompt corresponding to the second level.

[0070] In a case where it is determined that the agricultural machine is not stuck in the area in front of the traveling area, the processor can not send a warning prompt and an alarm prompt. In a case where it is determined that the agricultural machine is stuck in the area in front of the traveling area and the stuck level is the first level, the processor can send an alarm prompt corresponding to the first level, for example, a warning prompt. In a case where it is determined that the agricultural machine is stuck in the area in front of the traveling area and the stuck level is the second level, the processor can send an alarm prompt corresponding to the second level, for example, an alarm prompt.

[0071] Specifically, the agricultural machine can include a warning device. The warning device can include a voice warning device, an audible and light warning device, etc. In different stuck levels, corresponding prompts can be sent through the voice warning device and / or the audible and light warning device. The prompts can be sent through emails, short messages, and mobile phone applications. After the user receives the warning prompt or the alarm prompt, the agricultural machine can be controlled in time.

[0072] In an embodiment, the corresponding prompts in different stuck levels can set different alarm frequencies, alarm tones, and flashing colors. The above alarm frequencies, alarm tones, and flashing colors based on the alarm prompts corresponding to different stuck levels can be set in combination or separately according to requirements, which is not limited specifically here.

[0073] For example, the pre-warning prompt is sent when the stuck vehicle level is the first level, the alarm prompt is sent when the stuck vehicle level is the second level, and the alarm frequency of the pre-warning prompt can be lower than the alarm frequency of the alarm prompt. When the sound and light alarm device is used, yellow can be flashed when the pre-warning prompt corresponding to the first level is sent, so that the user can send a control instruction to enter the anti-stuck speed change mode when observing that the sound and light alarm device displays yellow, so that the agricultural machine enters the anti-stuck speed change mode. When the alarm prompt corresponding to the second level is sent, red can be flashed, so that the user can send a control instruction to enter the shutdown mode when observing that the sound and light alarm device displays red, so that the agricultural machine enters the shutdown mode.

[0074] Through the above technical solution, the attitude data, positioning data, vehicle body control data and ground surface monitoring data of the agricultural machine are multi-source fused, and the influence of the running fluctuation data of the agricultural machine in the running process on the running state parameter is considered, so that a stuck vehicle prediction model with higher prediction accuracy is established, the efficiency and accuracy of subsequent agricultural machine stuck vehicle prediction are improved, and the operation of the agricultural machine is safer.

[0075] Figure 1 The flowchart of the stuck vehicle prediction method for the agricultural machine in one embodiment is shown. It should be understood that, although Figure 1 the steps in the flowchart are displayed in sequence according to the arrows, these steps are not necessarily executed in sequence according to the arrows. Unless otherwise specified in this article, the execution of these steps has no strict order limitation, and these steps can be executed in other orders. Moreover, Figure 1 at least part of the steps in the flowchart can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily sequential, but can be alternately executed with other steps or at least part of the sub-steps or stages of other steps.

[0076] In one embodiment, a stuck vehicle prediction device for an agricultural machine is provided, comprising:

[0077] a memory configured to store instructions; and

[0078] a processor configured to call the instructions from the memory and capable of realizing the above-mentioned stuck vehicle prediction method for an agricultural machine when executing the instructions.

[0079] In one embodiment, as Figure 2 shown, another stuck vehicle prediction device 200 for an agricultural machine is provided, comprising a multi-sensor real-time monitoring module 201, a multi-source data real-time acquisition module 202, a stuck vehicle prediction and warning module 203, and an operation mode control module 204, wherein:

[0080] A multi-sensor real-time monitoring module 201 is configured to monitor the attitude information of the vehicle body in front, the positioning information of the vehicle body, the control parameters of the vehicle body, and the ground monitoring information in real time.

[0081] A multi-source data real-time acquisition module 202 is configured to acquire the data sources of various sensors in real time and transmit them to the vehicle sinking prediction and early warning module in a standard data format.

[0082] A vehicle sinking prediction and early warning module 203 is configured to establish a vehicle sinking prediction and early warning model based on the data sources of various sensors, and when the latest data stream is obtained, determine the driving state of the agricultural machine by calling the burned vehicle sinking prediction and early warning model, and determine the vehicle sinking level of the area in front of the agricultural machine according to the driving state.

[0083] An operation mode control module 204 is configured to control the actuator to automatically switch to the corresponding optimal operation mode according to the vehicle sinking level sent by the vehicle sinking prediction and early warning module.

[0084] Specifically, the multi-sensor real-time monitoring module 201 can monitor the attitude information of the vehicle body in front in real time through the inertial navigation IMU and the accelerometer, monitor the control parameters of the vehicle body in front in real time through the speed / current / torque sensor, monitor the positioning information of the vehicle body in front in real time through the RTK base station, and monitor the flatness information of the ground in front of the agricultural machine in real time through the sonar sensor.

[0085] The multi-source data real-time acquisition module 202 can acquire the data sources of various sensors and monitoring units in real time and transmit them to the vehicle sinking prediction and early warning module 203 in a standard data format. Among them, the data source transmission can adopt TCP / IP Ethernet, RS485 serial port, can communication and other data transmission protocols. The data sources can include the X, Y and Z coordinates of the vehicle body obtained through the inertial navigation IMU, the vehicle body roll angle Nro, the vehicle body yaw angle Nya, the vehicle body pitch angle Npit, the vehicle body speed n, the motor torque T, and the motor current I obtained through the speed / current / torque sensor, the current vehicle body latitude value B, the longitude value L, the elevation value H, the driving speed v, the elevation difference DH, and the position deviation DP obtained through the RTK base station, and the flatness Sur_flat of the ground in front obtained through the sonar sensor.

[0086] The vehicle sinking prediction and early warning module 203 can establish a vehicle sinking prediction and early warning model based on the historical data collected during the agricultural machine walking process under different working conditions (muddy ground, hard ground) and different control modes (anti-sinking, working, neutral), and is deployed in the STM32H750_Pro development board chip. Specifically, the torque fluctuation T_std, current fluctuation I_std, and speed fluctuation n_std of the agricultural machine equipment can be calculated according to the vehicle body speed n, motor torque T, and motor current I in the historical data, respectively.

[0087] Afterwards, as shown in Table 1 below, the travel speed v of the agricultural equipment, the position deviation DP, the cumulative elevation difference HP, the roll angle Nro, the yaw angle Nya, the pitch angle Npit, the wheel speed n, the wheel speed fluctuation n_std, the motor current I, the motor current fluctuation I_std, the transmission torque T, the transmission torque fluctuation T_std, and the ground flatness Sur_flat in front of the travel can be selected as input, and the working state can be selected as the output label to establish a stuck vehicle prediction classification model.

[0088] Table 1: Stuck vehicle prediction warning model input and output

[0089]

[0090]

[0091] By covariance analysis method, the covariance coefficient Ci between the 13 input parameters in Nx and the output parameter Ny is calculated one by one. If the covariance coefficient Ci is larger, the correlation between the input parameters and the input parameter Ny is stronger, and vice versa. According to the covariance coefficient Ci, the top 8 input parameters with the strongest correlation with the stuck vehicle risk are selected, and the top 8 input parameters are used as the input layer data Inp of the stuck vehicle prediction classification model.

[0092] Afterwards, the stuck vehicle prediction risk model can be trained based on the input layer data Inp. The training steps include: 1) CNN convolution operation can be performed on the input layer data Inp of the stuck vehicle prediction risk classification model to extract the convolution features Cov of the input layer 8 parameters, thereby reducing the complexity of the stuck vehicle prediction classification model modeling; 2) the convolution features Cov of the stuck vehicle prediction classification model are input into the average pooling layer of the neural network, thereby extracting the weight coefficient w between the convolution features Cov in the stuck vehicle prediction classification model, and further suppressing the overfitting of the stuck vehicle prediction classification model; 3) the convolution features Cov and the weight coefficient w in the above stuck vehicle prediction classification model are input into the full connection layer, and the model output parameter Ny is used as the output label of the full connection layer, and the minimum prediction error is used as the evaluation standard to establish the optimal stuck vehicle prediction warning model. The stuck vehicle prediction warning model can support different deep learning frameworks such as PyTorch, Tensorflow, Caffe, Keras, etc.

[0093] After the stuck vehicle prediction warning model is trained, the corresponding model file can be generated. The model file format is.H5 or.pt / .pth or.SaveModel or.onnx. Thereafter, relying on the STM32H750_Pro development board, the model file of the stuck vehicle prediction warning can be burned into the chip of the development board, and the model compression rate can be set to realize the reduction of model capacity and the reduction of operation internal loss.

[0094] When the agricultural machine generates the latest data stream during the traveling, the multi-source data real-time acquisition module 202 transmits the latest data stream to the vehicle sinking prediction and early warning module 203, the vehicle sinking prediction and early warning module 203 performs edge computing by calling the burned vehicle sinking prediction and early warning model, and outputs the traveling state of the agricultural machine in real time. The vehicle sinking prediction and early warning module 203 can determine the vehicle sinking level of the area in front of the agricultural machine according to the traveling state, and send it to the operation mode control module 204.

[0095] The operation mode control module 204 can control the actuator to automatically switch to the best operation mode according to the vehicle sinking level sent by the vehicle sinking prediction and early warning module 203. For example, if the traveling state of the agricultural machine is normal, the corresponding vehicle sinking level is normal operation level, at this time, the high-speed operation mode can be automatically switched. If the traveling state of the agricultural machine is a yellow warning state of vehicle sinking, the corresponding vehicle sinking level is a yellow warning level, at this time, the anti-sinking speed mode can be automatically switched to quickly pass through the deep sinking area of the mud foot through large torque and low speed. If the traveling state of the agricultural machine is a red alarm state of vehicle sinking, the corresponding vehicle sinking level is a red alarm level, at this time, the stop mode can be automatically switched, and an audible and visual alarm can be sent to prompt the agricultural machine driver to take corresponding measures.

[0096] The vehicle sinking prediction device for agricultural machinery 200 includes a processor and a memory, the above-mentioned multi-sensor real-time monitoring module 201, multi-source data real-time acquisition module 202, vehicle sinking prediction and early warning module 203 and operation mode control module 204 are stored in the memory as program units, and the corresponding functions are realized by the processor executing the above-mentioned program modules stored in the memory.

[0097] The processor contains a core, and the corresponding program unit is called from the memory by the core. The core can be set to one or more, and the core parameters are adjusted to realize the vehicle sinking prediction method for agricultural machinery.

[0098] The memory can include non-permanent memory in a computer readable medium, random access memory (RAM) and / or non-volatile memory such as read-only memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.

[0099] In one embodiment, a storage medium is provided, which stores a program, and the program is executed by the processor to realize the above-mentioned vehicle sinking prediction method for agricultural machinery.

[0100] In one embodiment, a processor is provided, and the processor is used to run a program, wherein the program is executed to perform the above-mentioned vehicle sinking prediction method for agricultural machinery.

[0101] In the embodiments of the present application, an agricultural machine is provided, comprising:

[0102] a vehicle body control prediction sensor, the vehicle body control prediction sensor comprising a rotating speed sensor, a current sensor and a torque sensor, the rotating speed sensor being configured to acquire a rotating speed of a vehicle body of the agricultural machine, the current sensor being configured to acquire a motor current of the agricultural machine, and the torque sensor being configured to acquire a motor torque of the agricultural machine;

[0103] a vehicle body posture prediction device, the vehicle body posture prediction device comprising an inertial measurement unit (IMU), the IMU being configured to acquire at least one of a vehicle body position, a vehicle body roll angle, a vehicle body yaw angle and a vehicle body pitch angle of the agricultural machine;

[0104] a vehicle body positioning prediction device, the vehicle body positioning prediction device comprising a real-time kinematic (RTK) base station, the RTK base station being configured to acquire at least one of a latitude value, a longitude value, an elevation value, a travel speed, an elevation difference and a position deviation of the agricultural machine;

[0105] a ground surface monitoring sensor, the ground surface monitoring sensor comprising a sonar sensor, the sonar sensor being configured to acquire a ground flatness of an area in front of the agricultural machine; and

[0106] the above-mentioned vehicle sinking prediction device for the agricultural machine.

[0107] In one embodiment, a computer device can be provided, which can be a server, and an internal structure diagram of the computer device can be as shown in Figure 3 The computer device comprises a processor A01, a network interface A02, a memory (not shown in the figure) and a database (not shown in the figure) connected through a system bus. Among them, the processor A01 of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises an internal memory A03 and a non-volatile storage medium A04. The non-volatile storage medium A04 stores an operating system B01, a computer program B02 and a database (not shown in the figure). The internal memory A03 provides an environment for the operating system B01 and the computer program B02 in the non-volatile storage medium A04 to run. The database of the computer device is configured to store data such as vehicle sinking prediction results. The network interface A02 of the computer device is configured to communicate with an external terminal through a network connection. The computer program B02 is executed by the processor A01 to implement a vehicle sinking prediction method for an agricultural machine.

[0108] Those skilled in the art can understand, Figure 3 the structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can comprise more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0109] The embodiment of the present application provides a device, the device comprising a processor, a memory and a program stored on the memory and executable on the processor, and the processor implements the following steps when executing the program: obtaining a historical running data set of an agricultural machine under different working conditions, the historical running data set comprising at least one of historical attitude data, historical positioning data, historical vehicle body control data and historical ground monitoring data; determining historical running fluctuation data of the agricultural machine under each working condition according to the historical vehicle body control data under each working condition; and constructing an agricultural machine stuck vehicle prediction model according to the historical running data set under different working conditions, the historical running fluctuation data and historical running state parameters of the agricultural machine under different working conditions, wherein the agricultural machine stuck vehicle prediction model is used to predict whether the agricultural machine is stuck in a corresponding region in front of the agricultural machine.

[0110] In the embodiment of the present application, constructing the agricultural machine stuck vehicle prediction model according to the historical running data set under different working conditions, the historical running fluctuation data and the historical running state parameters of the agricultural machine under different working conditions comprises: determining covariance coefficients between the historical running data set and the historical running fluctuation data and historical running state parameters under corresponding working conditions respectively; screening a target training data set from the historical running data set and the historical running fluctuation data based on the covariance coefficients; and inputting the target training data set into a preset neural network to make the preset neural network output corresponding historical running state parameters, so as to construct the agricultural machine stuck vehicle prediction model.

[0111] In the embodiment of the present application, screening the target training data set from the historical running data set and the historical running fluctuation data based on the covariance coefficients comprises: taking all historical data corresponding to the first N covariance coefficients with the largest values in all covariance coefficients as historical data meeting a condition, wherein N is a positive integer; and constructing the target training data set according to the historical data meeting the condition.

[0112] In the embodiment of the present application, the historical vehicle body control data comprises a vehicle body speed, a motor current and a motor torque of the agricultural machine in a running process, and the historical running fluctuation data comprises a speed fluctuation amount, a current fluctuation amount and a torque fluctuation amount of the agricultural machine in the running process; and determining the historical running fluctuation data of the agricultural machine under each working condition according to the historical vehicle body control data under each working condition comprises: determining the speed fluctuation amount of the agricultural machine under each working condition according to the vehicle body speed of the agricultural machine; determining the current fluctuation amount of the agricultural machine under each working condition according to the motor current of the agricultural machine; and determining the torque fluctuation amount of the agricultural machine under each working condition according to the motor torque of the agricultural machine.

[0113] In the embodiment of the present application, the vehicle sinking prediction method further comprises: obtaining a current running data set of the agricultural machine under a current working condition; inputting the current running data set into the agricultural machine vehicle sinking prediction model to obtain a travel state prediction parameter of the agricultural machine in a region in front of the agricultural machine; and predicting whether the agricultural machine will sink in the region in front of the agricultural machine according to the travel state prediction parameter.

[0114] In the embodiment of the present application, predicting whether the agricultural machine will sink in the region in front of the agricultural machine according to the travel state prediction parameter comprises: determining that the agricultural machine will not sink in the region in front of the agricultural machine when the travel state prediction parameter is a first value; and determining that the agricultural machine will sink in the region in front of the agricultural machine when the travel state prediction parameter is a second value or a third value.

[0115] In the embodiment of the present application, the vehicle sinking prediction method further comprises: controlling the agricultural machine to enter a high-speed working mode when it is determined that the agricultural machine will not sink in the region in front of the agricultural machine; determining a sinking level of the agricultural machine when it is determined that the agricultural machine will sink in the region in front of the agricultural machine, the sinking level comprising a first level and a second level, and a sinking degree corresponding to the second level being higher than a sinking degree corresponding to the first level; controlling the agricultural machine to enter an anti-sinking gear mode to enable the agricultural machine to pass through a corresponding sinking region when the sinking level is the first level; and controlling the agricultural machine to enter a shutdown mode to control the agricultural machine to stop working when the sinking level is the second level.

[0116] In the embodiment of the present application, the vehicle sinking prediction method further comprises: not sending a warning prompt and an alarm prompt when it is determined that the agricultural machine will not sink in the region in front of the agricultural machine; sending an alarm prompt corresponding to the first level when it is determined that the agricultural machine will sink in the region in front of the agricultural machine and the sinking level is the first level; and sending an alarm prompt corresponding to the second level when it is determined that the agricultural machine will sink in the region in front of the agricultural machine and the sinking level is the second level.

[0117] The present application also provides a computer program product adapted to execute a program for initializing steps of a vehicle sinking prediction method for an agricultural machine when executed on a data processing device.

[0118] Those skilled in the art will understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0119] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 means for functionally implementing the steps listed in the flowchart block or blocks.

[0120] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 means for functionally implementing the steps listed in the flowchart block or blocks.

[0121] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 means for functionally implementing the steps listed in the flowchart block or blocks.

[0122] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0123] The memory can include non-persistent memory and / or volatile memory, such as a random access memory (RAM) including a cache area for the temporary storage of data. The memory can also include non-volatile memory, such as read only memory (ROM) for storing structural information and / or instruction code. Both volatile and non-volatile memory can be implemented as a flash memory, a magnetic memory, an optical memory, and / or any non-transitory computer readable storage medium.

[0124] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can implement information storage by any method or technology. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0125] It should also be noted that the terms "comprising", "containing", or any other variant thereof are intended to cover a non-exclusive inclusion, such that a process, method, article or apparatus that comprises a list of elements does not only include those elements, but can also include other elements not expressly listed or inherent to such process, method, article or apparatus. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or apparatus that includes the element.

[0126] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the scope of claims of the present application.

Claims

1. A stuck prediction method for an agricultural machine, characterized in that, The stuck vehicle prediction method comprises: obtaining a historical running data set of the agricultural machine under different working conditions, the historical running data set comprising at least one of historical attitude data, historical positioning data, historical vehicle body control data and historical ground monitoring data; determining historical running fluctuation data of the agricultural machine under each working condition according to the historical vehicle body control data under each working condition; constructing an agricultural machine stuck vehicle prediction model according to the historical running data set, the historical running fluctuation data and historical running state parameters of the agricultural machine under different working conditions, wherein the agricultural machine stuck vehicle prediction model is used to predict whether the agricultural machine is stuck in a corresponding region in front of the agricultural machine.

2. The stuck-vehicle prediction method for an agricultural machine according to claim 1, characterized by, The construction of the agricultural machine stuck vehicle prediction model according to the historical running data set, the historical running fluctuation data and the historical running state parameters of the agricultural machine under different working conditions comprises: determining covariance coefficients between the historical running data set and the historical running fluctuation data and historical running state parameters under corresponding working conditions respectively; based on the covariance coefficients, screening a target training data set from the historical running data set and the historical running fluctuation data; inputting the target training data set into a preset neural network to make the preset neural network output corresponding historical running state parameters to construct the agricultural machine stuck vehicle prediction model.

3. The stuck-vehicle prediction method for an agricultural machine according to claim 2, characterized by, The screening of the target training data set from the historical running data set and the historical running fluctuation data based on the covariance coefficients comprises: taking all historical data corresponding to the first N covariance coefficients with the largest values in all covariance coefficients as the historical data meeting the condition, wherein N is a positive integer; constructing the target training data set according to the historical data meeting the condition.

4. The stuck-vehicle prediction method for an agricultural machine according to claim 1, characterized by, The historical vehicle body control data comprises vehicle body speed, motor current and motor torque of the agricultural machine during running, and the historical running fluctuation data comprises speed fluctuation, current fluctuation and torque fluctuation of the agricultural machine during running. The determination of the historical running fluctuation data of the agricultural machine under each working condition according to the historical vehicle body control data under each working condition comprises: determining the speed fluctuation of the agricultural machine under each working condition according to the vehicle body speed of the agricultural machine; determining the current fluctuation of the agricultural machine under each working condition according to the motor current of the agricultural machine; determining the torque fluctuation of the agricultural machine under each working condition according to the motor torque of the agricultural machine.

5. The stuck-vehicle prediction method for an agricultural machine according to claim 1, characterized by, The stuck vehicle prediction method further comprises: obtaining a current running data set of the agricultural machine under a current working condition; inputting the current running data set into the agricultural machine stuck vehicle prediction model to obtain running state prediction parameters of the agricultural machine in a corresponding region in front of the agricultural machine; predicting whether the agricultural machine is stuck in the region in front of the agricultural machine according to the running state prediction parameters.

6. The stuck-vehicle prediction method for an agricultural machine according to claim 5, characterized by, The prediction of whether the agricultural machine is stuck in the region in front of the agricultural machine according to the running state prediction parameters comprises: in a case where the traveling state prediction parameter is the first value, determining that the agricultural machine will not get stuck in the travel front area; in a case where the traveling state prediction parameter is the second value or the third value, determining that the agricultural machine will get stuck in the travel front area.

7. The stuck-vehicle prediction method for an agricultural machine according to claim 6, characterized by, The stuck prediction method further comprises: in a case where it is determined that the agricultural machine does not get stuck in the travel front area, controlling the agricultural machine to enter a high-speed operation mode; in a case where it is determined that the agricultural machine gets stuck in the travel front area, determining a stuck level of the agricultural machine in the travel front area, the stuck level comprising a first level and a second level, the second level corresponding to a higher stuck degree than the first level; in a case where the stuck level is the first level, controlling the agricultural machine to enter a stuck prevention gear mode to enable the agricultural machine to pass through a corresponding stuck area; in a case where the stuck level is the second level, controlling the agricultural machine to enter a shutdown mode to control the agricultural machine to stop operation.

8. The stuck-vehicle prediction method for an agricultural machine according to claim 7, characterized by, The stuck prediction method further comprises: in a case where it is determined that the agricultural machine does not get stuck in the travel front area, not sending a pre-warning prompt and an alarm prompt; in a case where it is determined that the agricultural machine gets stuck in the travel front area and the stuck level is the first level, sending an alarm prompt corresponding to the first level; in a case where it is determined that the agricultural machine gets stuck in the travel front area and the stuck level is the second level, sending an alarm prompt corresponding to the second level.

9. A stuck prediction device for an agricultural machine, characterized by, The stuck prediction device comprises: a memory configured to store instructions; and a processor configured to call the instructions from the memory and enable the stuck prediction method for an agricultural machine according to any one of claims 1 to 8 to be implemented when the instructions are executed.

10. A machine-readable storage medium, characterized in that, The machine-readable storage medium has instructions stored thereon for causing a machine to perform the stuck prediction method for an agricultural machine according to any one of claims 1 to 8. The machine-readable storage medium has instructions stored thereon for causing a machine to perform the stuck prediction method for an agricultural machine according to any one of claims 1 to 8.

Images