Method, device and storage medium for predicting operating parameters

Abstract

This disclosure provides a method, apparatus, device, and storage medium for predicting operating parameters. The method includes: acquiring multiple pre-trained prediction models and a gated neural network; inputting the current first operating parameter of a target object and the current control parameters for the target object into the multiple prediction models and the gated neural network to obtain specific initial second operating parameters of the target object under multiple different environment types, and specific probabilities of the target object being in different environment types; and predicting the current target second operating parameter of the target object based on the specific initial second operating parameters of the target object under multiple different environment types and the specific probabilities of the target object being in different environment types.

Description

Technical Field

[0001] This disclosure relates to the field of IT application technology, and in particular to a method, apparatus, device, and storage medium for predicting operating parameters. Background Technology

[0002] Currently, a single, global machine learning model is typically used to characterize the dynamics of agricultural machinery in the entire agricultural environment. Based on this characterization, the model predicts the trajectory of the machinery under human control, allowing operators to adjust control parameters to ensure the machinery follows the desired path. However, due to the complexity and variability of the agricultural environment, agricultural machinery exhibits significantly different dynamic characteristics in different scenarios, such as smooth roads, rugged mountain paths, and soft soil. A single global model struggles to accurately describe these diverse dynamic characteristics. Therefore, global machine learning models perform poorly in diverse agricultural scenarios, resulting in inaccurate predicted trajectories. Consequently, operators are unable to determine appropriate control parameters and accurately control the machinery to follow the desired trajectory. Summary of the Invention

[0003] This disclosure provides a method, apparatus, device, and storage medium for predicting operating parameters to address the problem in related technologies where the use of a single global model that is difficult to accurately describe the diverse dynamic characteristics under different scenarios results in low accuracy of the predicted agricultural machinery operating trajectory in diverse agricultural scenarios. This makes it impossible for relevant personnel to determine appropriate control quantities, thereby failing to accurately control the agricultural machinery to operate according to the desired trajectory.

[0004] In a first aspect, embodiments of this disclosure provide a method for predicting operating parameters, the method comprising:

[0005] Multiple pre-trained prediction models and a gated neural network are acquired; wherein, the prediction models are used to predict initial second operating parameters of the target object based on control parameters for the target object and first operating parameters of the target object; different prediction models are used to predict the initial second operating parameters of the target object under different environment types; the gated neural network is used to determine the probability of the target object being in the different environment types based on the first operating parameters of the target object; the initial second operating parameters predicted by the different prediction models and the probability of the target object being in the different environment types determined by the gated neural network are used to predict the target second operating parameters of the target object;

[0006] The current first operating parameters of the target object and the current control parameters for the target object are input into multiple prediction models and the gated neural network to obtain specific initial second operating parameters of the target object under multiple different environment types, and specific probabilities of the target object being in the different environment types;

[0007] Based on the specific initial second operating parameters of the target object under the multiple different environment types, and the specific probability of the target object being in the different environment types, the current target second operating parameters of the target object are predicted.

[0008] Secondly, embodiments of this disclosure provide an apparatus for predicting operating parameters, the apparatus comprising:

[0009] A first acquisition module is used to acquire multiple pre-trained prediction models and a gated neural network; wherein, the prediction models are used to predict the initial second operating parameters of the target object based on control parameters for the target object and the first operating parameters of the target object; different prediction models are used to predict the initial second operating parameters of the target object under different environment types; the gated neural network is used to determine the probability that the target object is in the different environment types based on the first operating parameters of the target object; the initial second operating parameters predicted by the different prediction models and the probability that the target object is in the different environment types determined by the gated neural network are used to predict the target second operating parameters of the target object;

[0010] The second acquisition model is used to input the current first operating parameters of the target object and the current control parameters for the target object into multiple prediction models and the gated neural network to obtain specific initial second operating parameters of the target object under multiple different environment types, and specific probabilities of the target object being in the different environment types;

[0011] The prediction module is used to predict the current target second operating parameters of the target object based on the specific initial second operating parameters of the target object under multiple different environment types, and the specific probability of the target object being in the different environment types.

[0012] Thirdly, embodiments of this disclosure provide a device for predicting operating parameters, comprising: a processor; and a memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the steps of the method described in the first aspect above.

[0013] Fourthly, embodiments of this disclosure provide a computer-readable storage medium for storing computer-executable instructions that, when executed by a processor, implement the steps of the method described in the first aspect.

[0014] Fifthly, embodiments of this disclosure provide a computer program product, the computer program product including a computer program, which, when executed by a processor, implements the steps of the method described in the first aspect above.

[0015] The at least one technical solution provided by the embodiments of the present invention can achieve the following technical effects:

[0016] In this embodiment of the invention, multiple pre-trained prediction models and gated neural networks can be obtained first. The prediction models can be used to predict the initial second operating parameters of the target object based on the control parameters and the first operating parameters of the target object. Different prediction models are used to predict the initial second operating parameters of the target object under different environment types. The gated neural network can be used to determine the probability of the target object being in different environment types based on the first operating parameters of the target object. The initial second operating parameters predicted by the different prediction models and the probability of the target object being in different environment types determined by the gated neural network can be used to predict the target second operating parameters of the target object. Then, the current first operating parameters of the target object and the current control parameters of the target object can be input into the multiple prediction models and the gated neural network to obtain the specific initial second operating parameters of the target object under multiple different environment types and the specific probability of the target object being in different environment types. Based on the obtained specific initial second operating parameters and specific probabilities, the current target second operating parameters of the target object can be predicted.

[0017] As described above, the embodiments of the present invention can acquire multiple prediction models for different environment types, and can predict the initial second operating parameters of the target object under the control parameters based on the acquired prediction models and the current operating parameters and control parameters of the target object. Furthermore, the embodiments of the present invention can also combine a gated neural network to determine the probability of the target object being in different environment types based on the current operating parameters of the target object, and, combined with the probability of the target object being in different environment types, selectively combine the initial second operating parameters of the target object under the control of the input control parameters in different environment types to predict the current target second operating parameters of the target object. This allows for high prediction accuracy in diverse environments, effectively addressing complex and diverse environments, and effectively solving the problem in related technologies where the use of a single global model that is difficult to accurately describe the diverse dynamic characteristics under different scenarios leads to low accuracy in predicting the trajectory of agricultural machinery in diverse agricultural scenarios, making it impossible for relevant personnel to determine appropriate control quantities and thus accurately control the agricultural machinery to operate according to the desired trajectory. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in one or more embodiments of this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 One of the flowcharts of a method for predicting operating parameters provided in an embodiment of the present invention;

[0020] Figure 2 A second schematic flowchart of a method for predicting operating parameters provided in an embodiment of the present invention;

[0021] Figure 3 A third schematic flowchart of a method for predicting operating parameters provided in an embodiment of the present invention;

[0022] Figure 4 The fourth flowchart illustrates a method for predicting operating parameters according to an embodiment of the present invention.

[0023] Figure 5 Fifth flowchart illustrating a method for predicting operating parameters according to an embodiment of the present invention;

[0024] Figure 6 A schematic diagram of the module composition of a predictor of operating parameters 600 provided in one embodiment of the present invention;

[0025] Figure 7A schematic diagram of the hardware structure of a device for predicting operating parameters provided in one embodiment of the present invention. Detailed Implementation

[0026] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this disclosure, and to make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0027] The technical solutions provided by the various embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0028] Please see Figure 1 , Figure 1 One of the flowcharts for a method to predict operating parameters provided in an embodiment of the present invention is shown below. Figure 1 As shown, the method includes the following steps:

[0029] Step 102: Obtain multiple pre-trained prediction models and gated neural networks; wherein, the prediction models are used to predict the initial second operating parameters of the target object based on the control parameters for the target object and the first operating parameters of the target object; different prediction models are used to predict the initial second operating parameters of the target object under different environment types; the gated neural network is used to determine the probability of the target object being in different environment types based on the first operating parameters of the target object; the initial second operating parameters predicted by different prediction models, and the probability of the target object being in different environment types determined by the gated neural network, are used to predict the target second operating parameters of the target object.

[0030] Step 104: Input the current first operating parameters of the target object and the current control parameters of the target object into multiple prediction models and gated neural networks to obtain the specific initial second operating parameters of the target object under multiple different environment types, as well as the specific probability of the target object being in different environment types.

[0031] Step 106: Based on the specific initial second operating parameters of the target object under multiple different environment types, and the specific probability of the target object being in different environment types, predict the current target second operating parameters of the target object.

[0032] In this embodiment of the invention, multiple pre-trained prediction models can be obtained first. These prediction models can be used to predict the initial second operating parameters of the target object based on the control parameters for the target object and the first operating parameters of the target object.

[0033] In one embodiment of the present invention, the operating parameters may include the horizontal coordinate, the vertical coordinate, the heading angle, the speed, and the front wheel steering angle.

[0034] In one example, the runtime parameters can be represented as , which can represent the x-coordinate, y-coordinate, heading angle, speed, and front wheel angle of the j-th time step for the i-th environment type, respectively. The time step can be set by the relevant personnel or a default value can be used. This example does not restrict this.

[0035] In this embodiment of the invention, different prediction models can be used to predict the initial second operating parameters of the same target object under different environmental types.

[0036] In one embodiment of the present invention, when obtaining multiple prediction models, the historical first running parameters of the target object can be obtained first, and then the historical first running parameters can be clustered to obtain the historical first running parameters corresponding to multiple environment types. In the clustering process, existing clustering methods can be used, such as using the GMM (Gaussian Mixture Model) clustering algorithm. The present invention does not limit this.

[0037] In one example, it can be obtained The first operating parameter points of each agricultural machine are collected and grouped together to form a dataset. ,in, The total number of data points, and Then, GMM clustering can be performed on dataset D to obtain K clusters. Each data point here It can be viewed as a sample in a five-dimensional space.

[0038] Then, in this embodiment of the invention, a specified number of induced operating parameters can be obtained from the historical operating parameters corresponding to any environment type by means of maximum mutual information, and an induced dataset corresponding to any environment type can be generated based on each induced operating parameter and its corresponding control parameter.

[0039] In one example, for each cluster We can select data points from each cluster using the maximum mutual information method. The selected data points were used as parameters for the induced operation, and the selected data points were used as parameters for the induced operation. The data points constitute the set of parameters for induced operation. ,in, Input is allowed; The range space of the entire input parameters; This can be the corresponding output; These are the input control parameters corresponding to the induced operation parameters.

[0040] Then, the prediction model corresponding to any environment type can be trained based on the induction dataset corresponding to any environment type.

[0041] In one example, when determining each cluster Corresponding induced operation parameter set Then, the induced running parameter set corresponding to each cluster can be used. This is used to train the prediction model corresponding to each cluster.

[0042] In another example, the trained prediction model can be called an SGPR (Sparse Gaussian Process Regression) model.

[0043] As can be seen from the above, the embodiments of the present invention can cluster the historical operating parameters of the target object according to the clustering algorithm to obtain multiple different environment types. Then, representative induced operating parameters can be selected from the historical operating parameters corresponding to each environment type, and the prediction model corresponding to each environment type can be trained according to the selected induced operating parameters, thereby obtaining a prediction model adapted to different environment types, improving the prediction accuracy in different environments, and thus effectively enhancing the ability to cope with diverse environments.

[0044] In one embodiment of the present invention, a pre-trained gated neural network can be obtained. This gated neural network can be used to determine the probability that a target object is in different environment types based on a first operating parameter of the target object. The probability of the target object being in different environment types determined by the gated neural network can be combined with the initial second operating parameters predicted by different prediction models to predict the target object's second operating parameters.

[0045] In this scheme, five-dimensional operating parameters can be obtained. Then, the five-dimensional operating parameters are input into the gated neural network to obtain the output. Let x be the probability that x belongs to each cluster, where T can represent the transpose of the matrix.

[0046] In one example, the training loss function for training a gated neural network can be:

[0047]

[0048] in, It is an indicator function; if the data points Belongs to clustering If the value of the indicator function is 1, then the value of the indicator function is 1; if it does not belong to a cluster... If the value of the indicator function is 0, then the value of the indicator function is 0.

[0049] like Figure 2 As shown, in this embodiment of the invention, the prediction model and the gated neural network can be pre-trained. During training, offline training can be performed using collected data. Specifically, the first operating parameters and corresponding control parameters of the original target object, such as agricultural machinery, can be obtained first, and the obtained data can be used as the original dataset. Then, the obtained data is clustered to obtain representative datasets corresponding to different environment types. For the representative datasets, the gated neural network and the prediction model can be trained separately. When training the gated neural network, the training can be performed based on the first operating parameters in the dataset and the environment type to which the first operating parameters belong, combined with the aforementioned loss function. When training the prediction model, the induced point operating parameters corresponding to each environment type can be determined first through maximum mutual information. Then, the prediction model is trained based on the induced point operating parameters to optimize the hyperparameters of the prediction model and obtain the trained prediction model. After obtaining the trained gated neural network and prediction model, a local composite prediction model can be obtained based on the trained gated neural network and each prediction model. After obtaining all local composite prediction models, the final prediction result can be determined based on the output results of all local composite prediction models. Since the prediction results under different environmental types can be determined with emphasis through gated neural networks, and the prediction results under various environmental types can be combined to determine the final prediction result, the scenario adaptability of the solution and the prediction accuracy for diverse scenarios can be effectively improved.

[0050] After obtaining multiple pre-trained prediction models and gated neural networks, the current first operating parameters of the target object and the current control parameters for the target object can be input into the multiple prediction models and gated neural networks to obtain the specific initial second operating parameters of the target object under multiple different environment types, as well as the specific probability of the target object being in different environment types.

[0051] Then, the target object's current target second operating parameters can be predicted based on the target object's specific initial second operating parameters under multiple different environment types, and the specific probability of the target object being in different environment types.

[0052] Specifically, when predicting a target object, for any environment type, the specific initial second operating parameters of the target object under that environment type and the product of the specific probability of the target object being under that environment type can be obtained. Then, based on the sum of the obtained multiple products, the current target second operating parameters of the target object are predicted.

[0053] In one example, the specific probability of the target object being in any environment type can be used as the weight value of a specific initial second running parameter. The specific probability is multiplied by the specific initial second running parameter, and the multiplication values ​​corresponding to different environment types are summed. The target object's current target second running parameter is predicted based on the sum.

[0054] In one example, the following formula can be obtained:

[0055]

[0056] in, This represents the prediction model, specifically the initial second running parameters output by the prediction model after z is input. In this example, Can be with Figure 2 The corresponding local composite prediction model in the text is, i.e. The value can correspond to Figure 2 The output of the local composite prediction model.

[0057] like Figure 3 As shown above, this scheme uses the specific probability of the target object being in any environment type as the weight value of the specific initial second operating parameter in that environment type. This allows for a focused combination of environment type to predict the target's second operating parameter, thus maintaining high prediction accuracy in diverse environments and effectively responding to complex and varied environments.

[0058] In embodiments of the present invention, such as Figure 4 As shown, when the current first operating parameters of the target object and the current control parameters of the target object are input into multiple prediction models and gated neural networks, and the specific initial second operating parameters of the target object under multiple different environment types and the specific probability of the target object being in different environment types are obtained, it can be determined whether to add the input data to the induced dataset.

[0059] Specifically, based on a gated neural network, the specific probability of a target object being in each environment type can be determined. Then, the specific environment type corresponding to the maximum specific probability can be determined, along with the first posterior variance of the prediction model corresponding to that specific environment type, and the second posterior variance of the prediction model corresponding to that specific environment type after adding the target object's current first operating parameters to the induced dataset corresponding to that specific environment type. When the difference between the first and second posterior variances is greater than a preset threshold for adding the difference, the target object's current first operating parameters are added to the induced dataset corresponding to that specific environment type.

[0060] In one example, when new data points Upon arrival, the probability of belonging to each cluster is first calculated using a gating network. Then, for the cluster with the highest probability... We can consider whether to Add its induced running parameter set .

[0061] Specifically, it can be calculated that the addition The reduction in posterior variance of the post-predictive model at this point:

[0062]

[0063] in, and Clustering before and after addition The corresponding prediction model is The posterior variance at that location. If Greater than the preset difference threshold Then join in .

[0064] In another embodiment of the present invention, to avoid an excessive number of induced data points in the induced dataset, redundant and poorly representative points can be removed from the induced running parameter set. Specifically, during removal, the increase in the posterior variance of the prediction model at that point after removal can be calculated for each candidate removal point. This increase is the difference between the posterior variance of the prediction model at that point after removal and the posterior variance of the prediction model at that point before removal. If the difference is less than a preset removal difference threshold, the point is removed.

[0065] In one example, to avoid having too many induced points, we also consider starting from... Remove redundant information points from the list. For each candidate point to remove... The increase in posterior variance of the prediction model at that point after removal can be calculated:

[0066]

[0067] in and The prediction models before and after removal are respectively The posterior variance at [location]. When Less than the preset removal difference threshold At that time, the removal will be performed.

[0068] like Figure 4 As shown, after updating the induced running parameter set, the hyperparameters of the prediction model corresponding to each cluster can be updated by maximizing the corrected log marginal likelihood. :

[0069]

[0070] In this embodiment of the invention, after predicting the target object's current target second operating parameters based on the target object's specific initial second operating parameters under multiple different environment types and the specific probability of the target object being in different environment types, the data distribution of the current first operating parameters can also be determined based on the target object's current first operating parameters.

[0071] In one example, a Gaussian process model of residual dynamics can be given. The propagation of the approximate state distribution through the composite system model:

[0072]

[0073] in, It can be a state evolution function, or an existing state evolution function can be used; It can be a residual coefficient matrix, which can be determined based on the difference between the predicted values ​​and the actual values ​​of a predetermined prediction model; This is a noise level, which can be set by relevant personnel based on the current environmental type.

[0074] Then, based on the data distribution of the current first operating parameter, the predetermined state evolution formula, the predetermined uncertainty propagation formula, and the predetermined residual mean and residual variance, the predicted variance of the specific initial second operating parameter can be determined. Based on the diagonal element of the predicted variance of the specific initial second operating parameter, the predetermined inverse cumulative distribution function of the standard normal distribution, and the predetermined confidence level parameter, the constraint bounds corresponding to the specific initial second operating parameter can be determined.

[0075] In one example, assume the initial state follows a Gaussian distribution. We can use a first-order Taylor expansion approximation model near the mean trajectory to predict the state, mean, and variance at each time step k+i in the prediction time domain:

[0076]

[0077]

[0078] in, ; .

[0079] Then, using the Gaussian distribution of the state, a tightened constraint bound is calculated:

[0080]

[0081] in, This corresponds to the uncertainty of lateral error. diagonal element; It can be the inverse cumulative distribution function of the standard normal distribution; These are parameters related to the confidence level or constraint conditions, used to determine the strictness or confidence level when calculating constraint boundaries. They can be default values ​​or set by relevant personnel.

[0082] The joint opportunity constraint is simplified into a set of deterministic constraints:

[0083]

[0084] In embodiments of the present invention, such as Figure 5 As shown, after determining the constraint bounds corresponding to the specific initial second operating parameters, the control strategy for the current first operating parameters of the target object can be determined based on the specific initial second operating parameters, the specific probability, and the constraint bounds corresponding to the specific initial second operating parameters, through a pre-determined stochastic model predictive control optimization algorithm.

[0085] In one example, combining the prediction results of the aforementioned determined prediction model, the propagation of approximate uncertainty, and the simplified chance constraint, the optimization problem of the control strategy for the target object can be formulated as follows:

[0086]

[0087]

[0088]

[0089]

[0090]

[0091]

[0092]

[0093]

[0094] in, , These are the predicted mean and covariance, respectively. , These are the state and control reference trajectories, respectively. , , It is a weighted matrix; , , , State and control boundaries; , Let J be the posterior mean and covariance of the local GPRs; The j-th local GPR weight is the output of the hybrid expert gating network; Lateral tracking error Chance constraint confidence level is .

[0095] In this embodiment of the invention, the above-mentioned optimization problem can be solved using nonlinear programming methods such as SQP (Sequential Quadratic Programming) or the interior-point method to obtain the optimal control sequence. It allows you to control the target object to run in a scrolling horizon manner, executing at each time step. And repeat the optimization.

[0096] In this embodiment of the invention, multiple pre-trained prediction models and gated neural networks can be obtained first. The prediction models can be used to predict the initial second operating parameters of the target object based on the control parameters and the first operating parameters of the target object. Different prediction models are used to predict the initial second operating parameters of the target object under different environment types. The gated neural network can be used to determine the probability of the target object being in different environment types based on the first operating parameters of the target object. The initial second operating parameters predicted by the different prediction models and the probability of the target object being in different environment types determined by the gated neural network can be used to predict the target second operating parameters of the target object. Then, the current first operating parameters of the target object and the current control parameters of the target object can be input into the multiple prediction models and the gated neural network to obtain the specific initial second operating parameters of the target object under multiple different environment types and the specific probability of the target object being in different environment types. Based on the obtained specific initial second operating parameters and specific probabilities, the current target second operating parameters of the target object can be predicted.

[0097] As described above, the embodiments of the present invention can acquire multiple prediction models for different environment types, and can predict the initial second operating parameters of the target object under the control parameters based on the acquired prediction models and the current operating parameters and control parameters of the target object. Furthermore, the embodiments of the present invention can also combine a gated neural network to determine the probability of the target object being in different environment types based on the current operating parameters of the target object, and, combined with the probability of the target object being in different environment types, selectively combine the initial second operating parameters of the target object under the control of the input control parameters in different environment types to predict the current target second operating parameters of the target object. This allows for high prediction accuracy in diverse environments, effectively addressing complex and diverse environments, and effectively solving the problem in related technologies where the use of a single global model that is difficult to accurately describe the diverse dynamic characteristics under different scenarios leads to low accuracy in predicting the trajectory of agricultural machinery in diverse agricultural scenarios, making it impossible for relevant personnel to determine appropriate control quantities and thus accurately control the agricultural machinery to operate according to the desired trajectory.

[0098] Corresponding to the above-described method for predicting operating parameters, this embodiment of the invention also provides an apparatus for predicting operating parameters. Figure 6 A schematic diagram of the module composition of the operating parameter prediction device 600 provided in an embodiment of the present invention is shown below. Figure 6 As shown, the predictor 600 for the operating parameters includes:

[0099] The first acquisition module 601 is used to acquire multiple pre-trained prediction models and a gated neural network; wherein, the prediction models are used to predict the initial second operating parameters of the target object based on the control parameters for the target object and the first operating parameters of the target object; different prediction models are used to predict the initial second operating parameters of the target object under different environment types; the gated neural network is used to determine the probability that the target object is in the different environment types based on the first operating parameters of the target object; the initial second operating parameters predicted by the different prediction models and the probability that the target object is in the different environment types determined by the gated neural network are used to predict the target second operating parameters of the target object;

[0100] The second acquisition module 602 is used to input the current first operating parameters of the target object and the current control parameters for the target object into multiple prediction models and the gated neural network to obtain specific initial second operating parameters of the target object under multiple different environment types, and specific probabilities of the target object being in the different environment types;

[0101] The prediction module 603 is used to predict the current target second operating parameters of the target object based on the specific initial second operating parameters of the target object under the multiple different environment types, and the specific probability of the target object being in the different environment types.

[0102] Optionally, the first acquisition module 601 is used for:

[0103] Obtain the historical first running parameters of the target object;

[0104] Clustering the historical first operating parameters yields historical first operating parameters corresponding to multiple environment types;

[0105] By using maximum mutual information, a specified number of induced operating parameters are obtained from the historical operating parameters corresponding to any environment type, and an induced dataset corresponding to any environment type is generated based on each induced operating parameter and its corresponding control parameter.

[0106] Based on the induced dataset corresponding to any of the environment types, a prediction model corresponding to any of the environment types is trained.

[0107] Optionally, the prediction module 603 is used for:

[0108] For any environment type, obtain the product of the specific initial second operating parameters of the target object under that environment type and the specific probability of the target object being under that environment type;

[0109] Based on the sum of multiple obtained products, the current target second operating parameters of the target object are predicted.

[0110] Optionally, the device further includes:

[0111] The first determining module 604 is used to determine the data distribution of the current first operating parameters based on the current first operating parameters of the target object after predicting the current target second operating parameters of the target object based on the specific initial second operating parameters of the target object under multiple different environment types and the specific probability of the target object being in the different environment types.

[0112] The second determining module 605 is used to determine the prediction variance of the specific initial second operating parameter based on the data distribution of the current first operating parameter, the predetermined state evolution formula, the predetermined uncertainty propagation formula, and the predetermined residual mean and residual variance.

[0113] The third determining module 606 is used to determine the constraint bounds corresponding to the specific initial second operating parameters based on the diagonal element of the predicted variance of the specific initial second operating parameters, the inverse cumulative distribution function of the pre-determined standard normal distribution, and the pre-determined confidence parameter.

[0114] Optionally, the device further includes ( Figure 6 (not shown in the image)

[0115] The fourth determining module 607 is used to determine, after determining the constraint bound corresponding to the specific initial second operating parameter, a control strategy for the current first operating parameter of the target object based on the specific initial second operating parameter, the specific probability, and the constraint bound corresponding to the specific initial second operating parameter, through a pre-determined stochastic model predictive control optimization algorithm.

[0116] Optionally, the second acquisition module 602 is used for:

[0117] Based on the gated neural network, the specific probability of the target object being in each environment type is determined;

[0118] Determine the specific environment type corresponding to the maximum specific probability;

[0119] The first posterior variance of the prediction model corresponding to the specific environment type is determined, and the second posterior variance of the prediction model corresponding to the specific environment type is determined after adding the current first running parameters of the target object to the induced dataset corresponding to the specific environment type.

[0120] When the difference between the first posterior variance and the second posterior variance is greater than a preset threshold for adding the difference, the current first running parameter of the target object is added to the induced dataset corresponding to the specific environment type.

[0121] In this embodiment of the invention, multiple pre-trained prediction models and gated neural networks can be obtained first. The prediction models can be used to predict the initial second operating parameters of the target object based on the control parameters and the first operating parameters of the target object. Different prediction models are used to predict the initial second operating parameters of the target object under different environment types. The gated neural network can be used to determine the probability of the target object being in different environment types based on the first operating parameters of the target object. The initial second operating parameters predicted by the different prediction models and the probability of the target object being in different environment types determined by the gated neural network can be used to predict the target second operating parameters of the target object. Then, the current first operating parameters of the target object and the current control parameters of the target object can be input into the multiple prediction models and the gated neural network to obtain the specific initial second operating parameters of the target object under multiple different environment types and the specific probability of the target object being in different environment types. Based on the obtained specific initial second operating parameters and specific probabilities, the current target second operating parameters of the target object can be predicted.

[0122] As described above, the embodiments of the present invention can acquire multiple prediction models for different environment types, and can predict the initial second operating parameters of the target object under the control parameters based on the acquired prediction models and the current operating parameters and control parameters of the target object. Furthermore, the embodiments of the present invention can also combine a gated neural network to determine the probability of the target object being in different environment types based on the current operating parameters of the target object, and, combined with the probability of the target object being in different environment types, selectively combine the initial second operating parameters of the target object under the control of the input control parameters in different environment types to predict the current target second operating parameters of the target object. This allows for high prediction accuracy in diverse environments, effectively addressing complex and diverse environments, and effectively solving the problem in related technologies where the use of a single global model that is difficult to accurately describe the diverse dynamic characteristics under different scenarios leads to low accuracy in predicting the trajectory of agricultural machinery in diverse agricultural scenarios, making it impossible for relevant personnel to determine appropriate control quantities and thus accurately control the agricultural machinery to operate according to the desired trajectory.

[0123] Corresponding to the above-described method for predicting operating parameters, this embodiment of the invention also provides a device for predicting operating parameters. Figure 7 A schematic diagram of the hardware structure of a device for predicting operating parameters provided in one embodiment of the present invention.

[0124] The device for predicting these operating parameters can be a terminal device or server for predicting trajectories, as provided in the above embodiments.

[0125] The device for predicting operating parameters can vary considerably depending on its configuration or performance. It may include one or more processors 701 and a memory 702, which may store one or more application programs or data. The memory 702 may be temporary or persistent storage. The application programs stored in the memory 702 may include one or more modules (not shown in the figures), each module including a series of computer-executable instructions for the device. Furthermore, the processor 701 may be configured to communicate with the memory 702 and execute the series of computer-executable instructions in the memory 702 on the device for predicting operating parameters. The device for predicting operating parameters may also include one or more power supplies 703, one or more wired or wireless network interfaces 704, one or more input / output interfaces 705, and one or more keyboards 706.

[0126] Specifically, in this embodiment, the device for predicting operating parameters includes a memory and one or more programs, wherein one or more programs are stored in the memory, and one or more programs may include one or more modules, and each module may include a series of computer-executable instructions for the device for predicting operating parameters, and is configured to be executed by one or more processors as described above.

[0127] In this embodiment of the invention, multiple pre-trained prediction models and gated neural networks can be obtained first. The prediction models can be used to predict the initial second operating parameters of the target object based on the control parameters and the first operating parameters of the target object. Different prediction models are used to predict the initial second operating parameters of the target object under different environment types. The gated neural network can be used to determine the probability of the target object being in different environment types based on the first operating parameters of the target object. The initial second operating parameters predicted by the different prediction models and the probability of the target object being in different environment types determined by the gated neural network can be used to predict the target second operating parameters of the target object. Then, the current first operating parameters of the target object and the current control parameters of the target object can be input into the multiple prediction models and the gated neural network to obtain the specific initial second operating parameters of the target object under multiple different environment types and the specific probability of the target object being in different environment types. Based on the obtained specific initial second operating parameters and specific probabilities, the current target second operating parameters of the target object can be predicted.

[0128] As described above, the embodiments of the present invention can acquire multiple prediction models for different environment types, and can predict the initial second operating parameters of the target object under the control parameters based on the acquired prediction models and the current operating parameters and control parameters of the target object. Furthermore, the embodiments of the present invention can also combine a gated neural network to determine the probability of the target object being in different environment types based on the current operating parameters of the target object, and, combined with the probability of the target object being in different environment types, selectively combine the initial second operating parameters of the target object under the control of the input control parameters in different environment types to predict the current target second operating parameters of the target object. This allows for high prediction accuracy in diverse environments, effectively addressing complex and diverse environments, and effectively solving the problem in related technologies where the use of a single global model that is difficult to accurately describe the diverse dynamic characteristics under different scenarios leads to low accuracy in predicting the trajectory of agricultural machinery in diverse agricultural scenarios, making it impossible for relevant personnel to determine appropriate control quantities and thus accurately control the agricultural machinery to operate according to the desired trajectory.

[0129] Another embodiment of this disclosure also provides a computer-readable storage medium for storing computer-executable instructions that, when executed by a processor, implement the above-described process.

[0130] The storage medium in this embodiment can implement the various processes of the above-described method embodiment for predicting operating parameters and achieve the same effects and functions, which will not be repeated here.

[0131] Another embodiment of this disclosure also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described process.

[0132] The computer program product in this disclosure embodiment can implement the various processes of the above-described method embodiment for predicting operating parameters and achieve the same effects and functions, which will not be repeated here.

[0133] In various embodiments of this disclosure, the computer-readable storage medium includes read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, etc.

[0134] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must also be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0135] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0136] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0137] For ease of description, the above apparatus is described by dividing it into various functional units. Of course, in implementing the embodiments of this disclosure, the functions of each unit can be implemented in one or more software and / or hardware.

[0138] Those skilled in the art will understand that one or more embodiments of this disclosure can be provided as a method, system, or computer program product. Therefore, one or more embodiments of this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, one or more embodiments of this disclosure 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-ROM, optical storage, etc.) containing computer-usable program code.

[0139] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0140] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0141] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0142] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0143] One or more embodiments of this disclosure can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a particular task or implement a particular abstract data type. One or more embodiments of this disclosure can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can reside in local and remote computer storage media, including storage devices.

[0144] The various embodiments in this disclosure are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0145] The above description is merely an embodiment of this disclosure and is not intended to limit the scope of this disclosure. Various modifications and variations can be made to this disclosure by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of the claims of this disclosure.

Claims

1. A method for predicting operating parameters, characterized in that, The method includes: Multiple pre-trained prediction models and a gated neural network are acquired; wherein, the prediction models are used to predict initial second operating parameters of the target object based on control parameters for the target object and first operating parameters of the target object; different prediction models are used to predict the initial second operating parameters of the target object under different environment types; the gated neural network is used to determine the probability that the target object is in the different environment types based on the first operating parameters of the target object; the initial second operating parameters predicted by the different prediction models and the probability that the target object is in the different environment types determined by the gated neural network are used to predict the target second operating parameters of the target object; the target object includes agricultural machinery; The current first operating parameters of the target object and the current control parameters for the target object are input into multiple prediction models and the gated neural network to obtain specific initial second operating parameters of the target object under multiple different environment types, and specific probabilities of the target object being in the different environment types; Based on the specific initial second operating parameters of the target object under the multiple different environment types, and the specific probability of the target object being in the different environment types, the current target second operating parameters of the target object are predicted; The acquisition of multiple pre-trained prediction models includes: Obtain the historical first running parameters of the target object; Clustering the historical first operating parameters yields historical first operating parameters corresponding to multiple environment types; By using maximum mutual information, a specified number of induced operating parameters are obtained from the historical first operating parameters corresponding to any environment type, and an induced dataset corresponding to any environment type is generated based on each induced operating parameter and its corresponding control parameter. Train a prediction model for any environment type based on the induced dataset corresponding to any environment type. The step of predicting the current target second operating parameters of the target object based on the specific initial second operating parameters of the target object under multiple different environment types, and the specific probability of the target object being in the different environment types, includes: For any environment type, obtain the product of the specific initial second operating parameters of the target object under that environment type and the specific probability of the target object being under that environment type; Based on the sum of multiple obtained products, the current target second operating parameters of the target object are predicted.

2. The method according to claim 1, characterized in that, After predicting the current target second operating parameters of the target object based on the specific initial second operating parameters of the target object under the multiple different environment types, and the specific probability of the target object being in the different environment types, the method further includes: Based on the current first operating parameters of the target object, determine the data distribution of the current first operating parameters; Based on the data distribution of the current first operating parameter, the predetermined state evolution formula, the predetermined uncertainty propagation formula, and the predetermined residual mean and residual variance, the prediction variance of the specific initial second operating parameter is determined. The constraint bounds corresponding to the specific initial second operating parameters are determined based on the diagonal elements of the predicted variance of the specific initial second operating parameters, the inverse cumulative distribution function of the pre-determined standard normal distribution, and the pre-determined confidence parameters.

3. The method according to claim 2, characterized in that, After determining the constraint bounds corresponding to the specific initial second operating parameter, the method further includes: Based on the specific initial second operating parameters, the specific probability, and the constraint bounds corresponding to the specific initial second operating parameters, a control strategy for the current first operating parameters of the target object is determined through a pre-determined stochastic model predictive control optimization algorithm.

4. The method according to claim 1, characterized in that, The step of inputting the current first operating parameters of the target object and the current control parameters for the target object into multiple prediction models and the gated neural network to obtain specific initial second operating parameters of the target object under multiple different environment types, and specific probabilities of the target object being in the different environment types, includes: Based on the gated neural network, the specific probability of the target object being in each environment type is determined; Determine the specific environment type corresponding to the maximum specific probability; The first posterior variance of the prediction model corresponding to the specific environment type is determined, and the second posterior variance of the prediction model corresponding to the specific environment type is determined after adding the current first running parameters of the target object to the induced dataset corresponding to the specific environment type. When the difference between the first posterior variance and the second posterior variance is greater than a preset threshold for adding the difference, the current first running parameter of the target object is added to the induced dataset corresponding to the specific environment type.

5. A device for predicting operating parameters, characterized in that, The device includes: A first acquisition module is used to acquire multiple pre-trained prediction models and a gated neural network; wherein, the prediction models are used to predict initial second operating parameters of the target object based on control parameters for the target object and first operating parameters of the target object; different prediction models are used to predict the initial second operating parameters of the target object under different environment types; the gated neural network is used to determine the probability that the target object is in the different environment types based on the first operating parameters of the target object; the initial second operating parameters predicted by the different prediction models and the probability that the target object is in the different environment types determined by the gated neural network are used to predict the target second operating parameters of the target object; the target object includes agricultural machinery; The second acquisition model is used to input the current first operating parameters of the target object and the current control parameters for the target object into multiple prediction models and the gated neural network to obtain specific initial second operating parameters of the target object under multiple different environment types, and specific probabilities of the target object being in the different environment types; The prediction module is used to predict the current target second operating parameters of the target object based on the specific initial second operating parameters of the target object under multiple different environment types, and the specific probability of the target object being in the different environment types; Wherein, the first acquisition module is used for: Obtain the historical first running parameters of the target object; Clustering the historical first operating parameters yields historical first operating parameters corresponding to multiple environment types; By using maximum mutual information, a specified number of induced operating parameters are obtained from the historical operating parameters corresponding to any environment type, and an induced dataset corresponding to any environment type is generated based on each induced operating parameter and its corresponding control parameter. Train a prediction model for any environment type based on the induced dataset corresponding to any environment type. The prediction module is used for: For any environment type, obtain the product of the specific initial second operating parameters of the target object under that environment type and the specific probability of the target object being under that environment type; Based on the sum of multiple obtained products, the current target second operating parameters of the target object are predicted.

6. A device for predicting operating parameters, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store computer-executable instructions that, when executed by a processor, implement the steps of the method described in any one of claims 1 to 4.

8. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the method described in any one of claims 1 to 4.

Images