Pose information estimation method and apparatus, storage medium, and device

Abstract

The present disclosure relates to a posture information estimation method, device, storage medium and equipment, the method comprising: acquiring multi-source posture detection data collected by sensors arranged on a coal machine equipment, the multi-source posture detection comprising a plurality of sub-posture detection data, different sub-posture detection data corresponding to different sensors; determining weighted posture detection data based on weight information of each sub-posture detection data; performing range analysis on the weighted posture detection data to obtain first posture data; inputting the weighted posture detection data into a pre-trained posture prediction model to obtain predicted posture data; performing Kalman filtering based on the predicted posture data and the weighted posture detection data to obtain corrected posture data; and determining posture information of the coal machine equipment based on the corrected posture data. The coal machine equipment can obtain posture information with high accuracy in a high-vibration and high-impact environment.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of coal machine equipment data processing, in particular, to a posture information estimation method and device, a storage medium and equipment. BACKGROUND

[0002] For coal mine equipment, especially coal machine equipment, the sensing detection during the mining process is usually affected by vibration and impact, resulting in inaccurate detection values and low sensing detection accuracy. SUMMARY

[0003] This summary is provided to introduce a selection of concepts that are further described below in the detailed description. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

[0004] According to a first aspect of an embodiment of the present disclosure, a posture information estimation method is provided, the method comprising:

[0005] Obtaining multi-source posture detection data collected by sensors arranged on coal machine equipment, the multi-source posture detection data comprising a plurality of sub-posture detection data, different sub-posture detection data corresponding to different sensors;

[0006] Determining weighted posture detection data based on weight information of each sub-posture detection data;

[0007] Performing range analysis on the weighted posture detection data to obtain first posture data;

[0008] Inputting the weighted posture detection data into a pre-trained posture prediction model to obtain predicted posture data;

[0009] Performing Kalman filtering based on the predicted posture data and the weighted posture detection data to obtain corrected posture data;

[0010] Determining posture information of the coal machine equipment based on the corrected posture data.

[0011] Optionally, the method further comprises:

[0012] Determining reliability influence factors of each sub-posture detection data based on installation positions of different posture detection sensors;

[0013] Calibrating initial weight information based on prior experience of the reliability influence factors;

[0014] Iterating the initial weight information based on historical posture detection data and historical true posture information corresponding to the historical posture detection data to obtain the weight information.

[0015] Optionally, the posture prediction model is trained by the following way:

[0016] inputting the sample detection data set into a first data verification model to obtain sample verification data output by the first data verification model for each sample detection data set;

[0017] obtaining a reliable index of each sample detection data set based on the sample detection data set and the corresponding sample verification data;

[0018] regarding the sample detection data set whose reliable index does not satisfy the preset threshold condition as a reliable sample detection data for data cleaning;

[0019] iterating the first posture prediction model based on the reliable sample detection data to obtain the posture prediction model.

[0020] Optionally, the obtaining of the reliable index of each sample detection data set based on the sample detection data set and the corresponding sample verification data comprises:

[0021] inputting the sample detection data set and the corresponding sample verification data into the first posture prediction model and a second posture prediction model respectively to obtain first predicted posture data output by the first posture prediction model and second predicted posture data output by the second posture prediction model, wherein the sample detection data set is called from prior predicted posture data used for training the second posture prediction model;

[0022] determining a prediction deviation between the first predicted posture data and the second predicted posture data, and determining a reliable index of the sample detection data set based on the prediction deviation, wherein the reliable index is negatively correlated with the prediction deviation.

[0023] Optionally, the obtaining of the reliable index of each sample detection data set based on the sample detection data set and the corresponding sample verification data comprises:

[0024] after extracting a detection data verification set from the sample detection data set, inputting the detection data verification set into a reliable index verification model to obtain verification data output by the reliable index verification model, wherein the reliable index verification model comprises sub-verification modules configured as different verification parameters;

[0025] regarding the detection data verification set corresponding to verification data meeting a verification rule as an example set, iterating a first verification module to obtain a target verification module, wherein the verification rule comprises that a similarity distance between the verification data and the sample verification data is lower than a preset threshold value.

[0026] Substitute the target verification module for the sub-verification module corresponding to the verification data conforming to the verification rule, and take the reliable index verification model obtained after the substitution as a target reliable index verification model;

[0027] Input the sample detection data set into the target reliable index verification model to obtain target verification data output by the target reliable index verification model;

[0028] Based on the target verification data and the sample verification data, obtain the reliable index of each sample detection data set.

[0029] Optionally, the obtaining of the reliable index of each sample detection data set based on the sample detection data set and the corresponding sample verification data comprises:

[0030] After determining the detection data verification set in the sample detection data set, input the detection data verification set into a reliable index verification model to obtain verification data output by the reliable index verification model, wherein the reliable index verification model comprises sub-verification modules configured with different verification parameters;

[0031] Input the detection data verification set into a first verification model to obtain first verification data output by the first verification model;

[0032] Iterate the first verification model based on the verification data and the first verification data to obtain a target verification model after training;

[0033] Input the sample detection data set into the target verification model to obtain sample verification data output by the target verification model;

[0034] Based on the sample verification data and the corresponding sample verification data, obtain the reliable index of each sample detection data set.

[0035] Optionally, the Kalman filtering based on the predicted posture data and the weighted posture detection data to obtain the corrected posture data further comprises:

[0036] Perform Kalman filtering based on a first weight of the predicted posture data and a second weight of the weighted posture detection data to obtain the corrected posture data.

[0037] According to a first aspect of an embodiment of the present disclosure, a posture information estimation device is provided, and the device comprises:

[0038] The acquisition module is configured to acquire multi-source attitude detection data collected by sensors arranged on the coal mining equipment, wherein the multi-source attitude detection includes a plurality of sub-attitude detection data, and different sub-attitude detection data corresponds to different sensors.

[0039] The first determination module is configured to determine weighted attitude detection data based on weight information of each sub-attitude detection data.

[0040] The second determination module is configured to perform range analysis on the weighted attitude detection data to obtain first attitude data.

[0041] The prediction module is configured to input the weighted attitude detection data into a pre-trained attitude prediction model to obtain predicted attitude data.

[0042] The filtering module is configured to perform Kalman filtering based on the predicted attitude data and the weighted attitude detection data to obtain corrected attitude data.

[0043] The third determination module is configured to determine attitude information of the coal mining equipment based on the corrected attitude data.

[0044] According to a third aspect of the embodiments of the present disclosure, a computer readable medium is provided, and the computer readable medium stores a computer program, and the computer program is executed by a processing device to implement the steps of the method in any one of the first aspect of the present disclosure.

[0045] According to a fourth aspect of the embodiments of the present disclosure, an electronic device is provided, and the electronic device includes:

[0046] A storage device stores a computer program.

[0047] A processing device is configured to execute the computer program in the storage device to implement the steps of the method in any one of the first aspect of the present disclosure.

[0048] Through the above technical solution, the multi-source attitude detection data collected by the sensors arranged on the coal mining equipment is acquired, the weighted attitude detection data is determined based on the weight information of each sub-attitude detection data, the range analysis is performed on the weighted attitude detection data to obtain the first attitude data, the weighted attitude detection data is input into the pre-trained attitude prediction model to obtain the predicted attitude data, the Kalman filtering is performed based on the predicted attitude data and the weighted attitude detection data to obtain the corrected attitude data, and finally the attitude information of the coal mining equipment is determined based on the corrected attitude data. Through the weighted fusion and the range analysis of the data, the problem that the multi-source data cannot be reused can be avoided, the unity of the data can be ensured, and then the attitude prediction model is used for prediction and the Kalman filtering is performed, so that the coal mining equipment can obtain the attitude information with high accuracy in the high-vibration and high-impact environment.

[0049] Other features and advantages of the present disclosure will be made apparent from the following detailed description. BRIEF DESCRIPTION OF DRAWINGS

[0050] The above and other features, advantages and aspects of embodiments of the present disclosure will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings in which:

[0051] Figure 1 is a flowchart of a posture information estimation method according to an exemplary embodiment.

[0052] Figure 2 is a block diagram of a posture information estimation apparatus according to an exemplary embodiment.

[0053] Figure 3 is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0054] Embodiments of the present disclosure will be described more fully hereinafter with reference to the accompanying drawings, in which some embodiments of the present disclosure are shown. This present disclosure may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and fully convey the scope of the present disclosure to those skilled in the art.

[0055] It should be understood that each of the steps of the method embodiments of the present disclosure can be performed in a different order, and / or in parallel. In addition, the method embodiments can include additional steps and / or omit performing the steps shown. The scope of the present disclosure is not limited in this respect.

[0056] The term "include," and derivations thereof, means "including, but not limited to." The term "based on" means "based, at least in part, on." The term "one embodiment" means "at least one embodiment." The term "another embodiment" means "at least one additional embodiment." The term "some embodiments" means "at least some embodiments." Related terms have corresponding meanings.

[0057] Note that the terms "first," "second," and the like, used in the present disclosure are merely used to distinguish one element from another, and do not limit the sequence or the interrelation of the elements.

[0058] It should be noted that the modification of "one", "multiple" mentioned in the present disclosure is illustrative but not restrictive, and those skilled in the art should understand that "one or more" should be understood unless otherwise explicitly indicated in the context.

[0059] The names of the messages or information exchanged between the plurality of devices in the embodiments of the present disclosure are only for illustrative purposes, and are not used to limit the scope of the messages or information.

[0060] It can be understood that, before using the technical solutions disclosed in the embodiments of the present disclosure, the type, use range, use scenario, etc. of the personal information involved in the present disclosure should be informed to the user and the authorization of the user should be obtained through appropriate means according to relevant laws and regulations.

[0061] For example, in response to receiving the active request of the user, the prompt information is sent to the user to explicitly prompt the user that the operation requested to be performed will require obtaining and using the personal information of the user. Thus, the user can voluntarily choose whether to provide the personal information to the software or hardware such as electronic device, application program, server or storage medium, etc. that performs the operation of the technical solutions of the present disclosure according to the prompt information.

[0062] As an optional but not limited implementation manner, in response to receiving the active request of the user, the prompt information can be sent to the user in the form of a pop-up window, and the prompt information can be presented in the form of text in the pop-up window. In addition, the pop-up window can also carry selection controls for the user to select "agree" or "disagree" to provide the personal information to the electronic device.

[0063] It can be understood that the above notification and user authorization process is only illustrative, and does not limit the implementation manner of the present disclosure, and other manners meeting the relevant laws and regulations can also be applied to the implementation manner of the present disclosure.

[0064] At the same time, it can be understood that the data (including but not limited to the data itself, the acquisition or use of the data) involved in the present technical solution should comply with the requirements of the relevant laws and regulations and relevant provisions.

[0065] In order to enable those skilled in the art to better understand the improvements of the technical solutions provided by the present disclosure, the present disclosure will first further introduce the related art. In the related art, for coal mine equipment, especially coal machine equipment, the sensing detection in the mining process is usually affected by vibration and impact, resulting in inaccurate detection value and low sensing detection precision.

[0066] On the other hand, the results of the measurement of the same physical parameter by multiple acquisition sources have low data utilization on existing equipment, and are generally used according to the single utilization principle of data accuracy to improve the reliability and durability of detection. In addition, unlike single data, multiple source data has errors such as redundancy and contradiction between data, which is generally difficult to use and utilize directly.

[0067] Figure 1 is a flowchart of a posture information estimation method according to an exemplary embodiment, which can be applied to coal machine equipment, for example, can be a central processor of coal machine equipment, or other arbitrary electronic devices with information processing capability, or the method can also be applied to a server to obtain data collected by coal machine equipment through a communication network and perform the following steps, which are not limited by the disclosure, as shown in Figure 1 The method comprises the following steps:

[0068] S101, acquiring multi-source posture detection data collected by sensors arranged on coal machine equipment, wherein the multi-source posture detection data comprises a plurality of sub-posture detection data, and different sub-posture detection data corresponds to different sensors.

[0069] For example, a plurality of sensors can be arranged at different positions of the coal machine equipment, for example, can be arranged in an electric control box and a rocker arm, etc. The data collected by each sensor can be used for posture detection of the coal machine equipment, thereby ensuring reliable operation of the coal machine equipment.

[0070] S102, determining weighted posture detection data based on weight information of each sub-posture detection data.

[0071] It can be understood that the sensors arranged at different positions may be affected differently by vibration or impact, and therefore the data collected by different sensors can be pre-set with corresponding weights to ensure that the posture information determined based on these data is more accurate.

[0072] S103, performing range analysis on the weighted posture detection data to obtain first posture data.

[0073] The range analysis method is also called intuitive analysis method, and the range can be used to represent the dispersion degree of a group of data, and reflects the variation range and dispersion amplitude of the variable distribution. The difference between the standard values of any two units in the population cannot exceed the range. At the same time, it can reflect the fluctuation range of a group of data. The larger the range, the greater the dispersion degree, and vice versa, the smaller the dispersion degree. The first posture data can be data obtained after removing data with large range. In this way, data with low accuracy can be removed, so that the dispersion degree of the obtained first posture data is lower, and the accuracy of the posture information is ensured.

[0074] S104, input the weighted attitude detection data into a pre-trained attitude prediction model to obtain predicted attitude data.

[0075] The attitude prediction model can predict the attitude of the coal machine equipment based on the input attitude detection data. The attitude prediction model can modify unreasonable or unreliable data in the input data based on a neural network, so that the obtained predicted attitude data is more reliable, and the influence of unreliable data in the multi-source attitude detection data on the attitude information is avoided.

[0076] S105, perform Kalman filtering based on the predicted attitude data and the weighted attitude detection data to obtain corrected attitude data.

[0077] The Kalman filtering can be an algorithm for optimal estimation of system state by using linear system state equation and observing data of system input and output. Since the observation data includes the influence of noise and interference in the system, the optimal estimation can also be regarded as a filtering process.

[0078] In an optional embodiment, the Kalman filtering based on the predicted attitude data and the weighted attitude detection data to obtain corrected attitude data further includes: performing Kalman filtering according to a first weight of the predicted attitude data and a second weight of the weighted attitude detection data to obtain the corrected attitude data.

[0079] The first weight and the second weight can be determined based on prior experience, for example, the first weight and the second weight can be determined based on historical real attitude information, historical predicted attitude data and historical weighted attitude detection data, and the specific values of the first weight and the second weight are not limited in the present disclosure.

[0080] S106, determine the attitude information of the coal machine equipment based on the corrected attitude data.

[0081] The corrected attitude data is more accurate detection data obtained by processing the weighted attitude detection data of the multi-source attitude detection data based on the weight, and based on the data, more accurate attitude information of the coal machine equipment can be obtained, and the reliable operation of the coal machine equipment is ensured.

[0082] In the embodiment of the present disclosure, multi-source attitude detection data collected by sensors arranged on the coal machine equipment is acquired; weighted attitude detection data is determined based on weight information of each sub-attitude detection data; first attitude data is obtained by performing range analysis on the weighted attitude detection data; the weighted attitude detection data is input into a pre-trained attitude prediction model to obtain predicted attitude data; the predicted attitude data and the weighted attitude detection data are subjected to Kalman filtering to obtain corrected attitude data; and finally, the attitude information of the coal machine equipment is determined based on the corrected attitude data. By performing weighted fusion and range analysis on the data, the problem that multi-source data cannot be reused can be avoided, and the unity of the data can be ensured, and then the attitude prediction model is used for prediction and Kalman filtering, so that the coal machine equipment can obtain attitude information with high accuracy in a high-vibration and high-impact environment.

[0083] In an optional embodiment, the method further comprises:

[0084] Based on the installation positions of the different attitude detection sensors, the reliability influence factors of each sub-attitude detection data are determined; based on prior experience of the reliability influence factors, initial weight information is calibrated; based on historical attitude detection data and historical real attitude information corresponding to the historical attitude detection data, the initial weight information is iterated to obtain the weight information.

[0085] It can be understood that after the attitude information is obtained based on step S106, the weight information can be further iterated to maintain the accuracy of the weight information.

[0086] In the embodiment of the present disclosure, the weight information can be iterated based on the historical attitude detection data and the corresponding historical real attitude information, which can effectively ensure the accuracy of the weight information, and thus the attitude information of the coal machine equipment determined is more accurate.

[0087] Further, in the related art, the attitude prediction model can be a model for predicting the real attitude data of the coal machine equipment based on low-reliability detection data during the operation of the coal machine equipment. When training the model, a large amount of historical detection data needs to be labeled, and the labeled historical detection data is input into the neural network model to complete the training of the model. However, due to the large amount of historical detection data collected, not only is the labeling workload large, but also the low-value detection data of repeated scenes after labeling is used as training samples to train the model, and the accuracy of the model obtained is low.

[0088] On the basis of the above-mentioned embodiments, the present disclosure further provides an attitude prediction model training method. Optionally, the method comprises the following steps, and the execution subject of the method can be the same as the execution subject of the method. Figure 1The execution subject of the method shown can or can not be the same, and the present disclosure does not limit this.

[0089] In step S201, the sample detection data set is input into the first data verification model to obtain sample verification data output by the first data verification model for each sample detection data set.

[0090] For example, the position or sensor identifier of the sensor in the collected detection data can be labeled for the sample detection data set, and the real pose of the coal machine equipment can also be labeled to obtain the sample verification data.

[0091] Optionally, the sample detection data set can be an initial data set without labels, and the first data verification model can label the data in the initial data set based on the data type, for example, if the first data verification model determines that some data in the data set is the height of the rocker arm, the data can be initially labeled to obtain the sample verification data.

[0092] In step S202, based on the sample detection data set and the corresponding sample verification data, a reliable index of each sample detection data set is obtained.

[0093] In one embodiment, the reliable index of each sample detection data set is obtained based on the sample detection data set and the corresponding sample verification data, including:

[0094] The sample detection data set and the corresponding sample verification data are respectively input into the first pose prediction model and the second pose prediction model to obtain first predicted pose data output by the first pose prediction model and second predicted pose data output by the second pose prediction model, wherein the sample detection data set is called from prior predicted pose data used to train the second pose prediction model.

[0095] The first data verification model, the first pose prediction model, and the second pose prediction model described above can all be neural network models, and the above models can all be preliminary trained models. The first data verification model can be a classification model such as a random forest model, and the first pose prediction model and the second pose prediction model can be BP neural network models, etc. The network structure of the embodiments of the present disclosure is not limited.

[0096] In the embodiments of the present disclosure, the second pose prediction model can be an offline model, the first pose prediction model is trained as a pose prediction model loaded on the coal machine equipment, the second pose prediction model is obtained by iterating a large amount of data in the prior predicted pose data, and the result output by the second pose prediction model has high accuracy. The first pose prediction model partially sacrifices accuracy to improve recall rate and running efficiency.

[0097] determining a prediction deviation between the first predicted pose data and the second predicted pose data, and determining a reliability indicator of the sample detection data set based on the prediction deviation, wherein the reliability indicator is negatively correlated with the prediction deviation.

[0098] For example, the prediction deviation can be a similarity distance between the prediction data of the two models.

[0099] In one embodiment, the reliability indicator of each sample detection data set is determined based on the sample detection data set and the corresponding sample verification data, including:

[0100] determining a detection data verification set from the sample detection data set, and inputting the detection data verification set into a reliability indicator verification model to obtain verification data output by the reliability indicator verification model, wherein the reliability indicator verification model includes sub-verification modules configured as different verification parameters.

[0101] In one embodiment, the sub-verification modules configured as different verification parameters can be in series. Each sample detection data set is input into a sub-verification module, and if the sample detection data set does not activate the sub-verification module of the verification parameter, the sub-verification module continues to read the next sample detection data set, and the sample detection data set that fails to activate the sub-verification module continues to pass down to the next sub-verification module, until after activating any sub-verification module, the sample detection data set no longer continues to pass down, and the verification data is obtained. In this way, the sample detection data set can be quickly activated.

[0102] In another embodiment, the sub-verification modules configured as different verification parameters can be in parallel. Each sample detection data set is input into each sub-verification module, and it is determined whether the sub-verification module of the verification parameter can be activated, and if all sub-verification modules output verification data, the next sample detection data set is read.

[0103] In the embodiments of the present disclosure, the reliability indicator verification model includes a verification parameter for type classification of any data defined by a human, and when the detection data verification set passes through the sub-verification modules of different verification parameters in sequence, if the verification parameter is met, the detection data verification set is labeled with the label of the reliability indicator verification model.

[0104] The detection data verification set that meets the verification rule corresponding to the verification data is taken as an example set, and the first verification module is iterated to obtain a target verification module, wherein the verification rule includes that a similarity distance between the verification data and the sample verification data is lower than a preset threshold and / or a verification time delay does not meet a preset time delay requirement.

[0105] In the embodiments of the present disclosure, the verification delay is the sum of the sub-verification delays corresponding to the sub-verification modules. Wherein, when the proportion of the verification data output by the sub-verification module and different from the label in the sample verification data reaches a preset threshold, it is determined that the accuracy of the sub-verification module is low.

[0106] Wherein, the example set and the sample verification data corresponding to the sample are input into the first verification module, and the first verification module is iterated.

[0107] The target verification module is substituted for the sub-verification module corresponding to the verification data meeting the verification rule, and the reliable index verification model obtained after the substitution is taken as a target reliable index verification model.

[0108] For example, the sub-verification module whose verification delay does not meet the preset delay requirement is retained, and the sub-verification module whose verification delay does not meet the preset delay requirement is substituted for the target verification module obtained by training.

[0109] The sample detection data set is input into the target reliable index verification model to obtain target verification data output by the target reliable index verification model.

[0110] Based on the target verification data and the sample verification data, the reliable index of each sample detection data set is obtained.

[0111] In the embodiments of the present disclosure, the reliable index of each sample detection data set is obtained by comparing the target verification data and the sample verification data.

[0112] In one of the embodiments, the reliable index of each sample detection data set is obtained based on the sample detection data set and the corresponding sample verification data, including:

[0113] A detection data verification set is determined from the sample detection data set, and the detection data verification set is input into a reliable index verification model to obtain verification data output by the reliable index verification model, wherein the reliable index verification model includes sub-verification modules configured with different verification parameters.

[0114] The detection data verification set is input into a first verification model to obtain first verification data output by the first verification model.

[0115] Based on the verification data and the first verification data, the first verification model is iterated to obtain a target verification model after training.

[0116] The sample detection data set is input into the target verification model to obtain sample verification data output by the target verification model.

[0117] Based on the sample verification data and the corresponding sample verification data, a reliability index of each sample detection data set is obtained.

[0118] The consistency of the sample verification data of the sample detection data set and the corresponding sample verification data is determined.

[0119] The higher the consistency of the sample verification data of the sample detection data set and the corresponding sample verification data, the higher the reliability index of the sample detection data set.

[0120] In step S203, the sample detection data set corresponding to the candidate detection data whose reliability index does not satisfy the preset threshold condition is taken as reliable sample detection data for data cleaning.

[0121] In step S204, the first attitude prediction model is iterated based on the reliable sample detection data, and the attitude prediction model is obtained.

[0122] In the embodiments of the present disclosure, the sample detection data set corresponding to the candidate detection data whose reliability index is greater than or equal to the preset threshold condition is removed, and the sample detection data set corresponding to the candidate detection data whose reliability index does not satisfy the preset threshold condition is taken as reliable sample detection data for data cleaning.

[0123] Based on the same inventive concept, the present disclosure also provides a block diagram of an attitude information estimation device according to an exemplary embodiment, as shown in Figure 2 As shown in Figure 2 The attitude information estimation device 20 comprises:

[0124] The acquisition module 21 is configured to acquire multi-source attitude detection data collected by sensors arranged on the coal machine equipment, wherein the multi-source attitude detection data comprises a plurality of sub-attitude detection data, and different sub-attitude detection data corresponds to different sensors.

[0125] The first determination module 22 is configured to determine weighted attitude detection data based on weight information of each sub-attitude detection data.

[0126] The second determination module 23 is configured to perform range analysis on the weighted attitude detection data to obtain first attitude data.

[0127] The prediction module 24 is configured to input the weighted attitude detection data into a pre-trained attitude prediction model to obtain predicted attitude data.

[0128] The filtering module 25 is configured to perform Kalman filtering based on the predicted attitude data and the weighted attitude detection data to obtain corrected attitude data.

[0129] The third determining module 26 is configured to determine the attitude information of the coal mining equipment based on the corrected attitude data.

[0130] Optionally, the attitude information estimation device 20 is further configured to:

[0131] determine a reliability influence factor of each sub-attitude detection data based on the installation positions of different attitude detection sensors;

[0132] calibrate initial weight information based on prior experience of the reliability influence factor;

[0133] iteratively obtain the weight information based on historical attitude detection data and historical true attitude information corresponding to the historical attitude detection data.

[0134] Optionally, the attitude information estimation device 20 further comprises:

[0135] an input module configured to input the sample detection data set into the first data verification model to obtain sample verification data output by the first data verification model for each sample detection data set;

[0136] a fourth determining module configured to obtain a reliability index of each sample detection data set based on the sample detection data set and the corresponding sample verification data;

[0137] a fifth determining module configured to take a sample detection data set whose reliability index does not satisfy the preset threshold condition as reliable sample detection data for data cleaning;

[0138] a sixth determining module configured to iteratively obtain the attitude prediction model based on the reliable sample detection data.

[0139] Optionally, the fourth determining module is configured to:

[0140] input the sample detection data set and the corresponding sample verification data into the first attitude prediction model and the second attitude prediction model respectively to obtain first predicted attitude data output by the first attitude prediction model and second predicted attitude data output by the second attitude prediction model, wherein the sample detection data set is called from prior predicted attitude data obtained by training the second attitude prediction model;

[0141] determine a prediction deviation between the first predicted attitude data and the second predicted attitude data, and determine a reliability index of the sample detection data set based on the prediction deviation, wherein the reliability index is negatively correlated with the prediction deviation.

[0142] Optionally, the fourth determining module is configured to;

[0143] After extracting the detection data verification set from the sample detection data set, input the detection data verification set into a reliable index verification model, obtain verification data output by the reliable index verification model, and the reliable index verification model includes sub-verification modules configured as different verification parameters;

[0144] Take the detection data verification set corresponding to the verification data meeting the verification rule as an example set, iterate the first verification module, and obtain a target verification module, wherein the verification rule includes that the similarity distance between the verification data and the sample verification data is lower than a preset threshold;

[0145] Replace the sub-verification module corresponding to the verification data meeting the verification rule with the target verification module, and take the obtained reliable index verification model after the replacement as a target reliable index verification model;

[0146] Input the sample detection data set into the target reliable index verification model, and obtain target verification data output by the target reliable index verification model;

[0147] Based on the target verification data and the sample verification data, obtain the reliable index of each sample detection data set.

[0148] Optionally, a fourth determination module is configured to:

[0149] After determining the detection data verification set in the sample detection data set, input the detection data verification set into a reliable index verification model, obtain verification data output by the reliable index verification model, and the reliable index verification model includes sub-verification modules configured as different verification parameters;

[0150] Input the detection data verification set into a first verification model, and obtain first verification data output by the first verification model;

[0151] Based on the verification data and the first verification data, iterate the first verification model to obtain a target verification model after training;

[0152] Input the sample detection data set into the target verification model, and obtain sample verification data output by the target verification model;

[0153] Based on the sample verification data and the corresponding sample verification data, obtain the reliable index of each sample detection data set.

[0154] Optionally, a filtering module 25 is configured to:

[0155] The corrected attitude data is obtained by performing Kalman filtering based on the first weight of the predicted attitude data and the second weight of the weighted attitude detection data.

[0156] The following is for reference. Figure 3 The block diagram shown illustrates an electronic device according to an exemplary embodiment, which shows an electronic device 300 suitable for implementing embodiments of the present disclosure (e.g., Figure 1 The diagram below shows the structure of the coal mining equipment in the corresponding embodiment. The terminal devices in this disclosure may include, but are not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle terminals (e.g., vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0157] like Figure 3 As shown, the electronic device 300 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0158] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0159] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program according to embodiments of the present disclosure. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program comprising program code for performing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the methods of embodiments of the present disclosure are performed.

[0160] It should be noted that the computer-readable medium described above in the present disclosure can be a computer-readable signal medium or a computer-readable storage medium or any combination thereof. The computer-readable storage medium can be, for example but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or apparatus, or any suitable combination thereof. More specific examples of the computer-readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program used by or in connection with an instruction execution system, apparatus, or device. In the present disclosure, the computer-readable signal medium can include a data signal propagated in baseband or propagated as a carrier wave in a propagated data signal, in which the computer-readable program code is carried. Such a propagated data signal can take a variety of forms, including but not limited to electro-magnetic, optical, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium that can be used to carry or store program code for use by or in connection with an instruction execution system, apparatus, or device. The program code contained in the computer-readable medium can be transmitted by any suitable medium, including but not limited to wire, cable, RF (radio frequency), or any suitable combination thereof.

[0161] In some embodiments, the client, server, or other computing machines utilized by the system can communicate over any known or future developed network protocol, such as the HyperText Transfer Protocol (HTTP), and can be interconnected with any form or medium of digital data communication (for example, a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the Internet, and peer-to-peer networks (for example, ad hoc peer-to-peer networks), as well as any current or future developed network.

[0162] The computer-readable medium described above can be included in the electronic device described above; alternatively, it can exist separately from the electronic device and be assembled into the electronic device.

[0163] The computer-readable medium described above carries one or more programs, which, when executed by the electronic device, cause the electronic device to: acquire multi-source attitude detection data collected by sensors arranged on the coal machine equipment, the multi-source attitude detection including a plurality of sub-attitude detection data, different sub-attitude detection data corresponding to different sensors; determine weighted attitude detection data based on weight information of each sub-attitude detection data; perform range analysis on the weighted attitude detection data to obtain first attitude data; input the weighted attitude detection data into a pre-trained attitude prediction model to obtain predicted attitude data; perform Kalman filtering based on the predicted attitude data and the weighted attitude detection data to obtain corrected attitude data; and determine attitude information of the coal machine equipment based on the corrected attitude data.

[0164] Computer program code for carrying out operations of the present disclosure can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network ("LAN") or a wide area network ("WAN"), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0165] The computer program product of the first aspect can include one or more non-transitory computer-readable media storing instructions that, when executed, cause one or more processors to perform the operations of the method of the first aspect. The computer program product of the first aspect can include a computer-readable medium storing instructions that, when executed, cause one or more processors to perform the operations of the method of the first aspect.

[0166] The modules involved in the embodiments of the present disclosure can be implemented in the form of software or in the form of hardware. In some cases, the name of the module does not constitute a limitation on the module itself.

[0167] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, non-limiting examples of exemplary types of hardware logic components that can be used include field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SOCs), complex programmable logic devices (CPLDs), etc.

[0168] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable storage media can include, without limitation, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media can include one or more lines of electrical wire, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the foregoing.

[0169] The above description merely illustrates the preferred embodiment of the disclosure and a principle of applied technologies. It should be understood by those skilled in the art that the disclosed range of the disclosure is not limited to the technical solutions formed by the specific combinations of the technical features described above, and should also cover other technical solutions formed by the combinations of the technical features described above or their equivalent features without departing from the disclosed concept. For example, the technical solutions formed by the mutual replacement of the above-described features and the technical features with similar functions disclosed in the disclosure (but not limited to) can be formed.

[0170] Furthermore, although operations are depicted in a particular, sequential order, this should not be understood as requiring or implying that the operations are performed in the order illustrated or sequentially. In certain circumstances, multitasking and parallel processing can be advantageous. Likewise, although specific implementation details are contained in the above discussion, these should not be construed as limiting the scope of the disclosure. Certain features described in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable sub-combination.

[0171] Although the subject matter has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are merely illustrative of the example forms of implementing the claims. As to the means for performing the operations of the apparatus in the above-described embodiments, the specific manner in which the various modules perform the operations has been described in detail in the embodiments related to the method, and will not be described here.

Claims

1. A method of estimating attitude information, characterized by, The method comprises: Obtaining multi-source posture detection data collected by sensors arranged on the coal machine equipment, wherein the multi-source posture detection comprises a plurality of sub-posture detection data, and different sub-posture detection data corresponds to different sensors; Determining weighted posture detection data based on weight information of each sub-posture detection data; Performing range analysis on the weighted posture detection data to obtain first posture data; Inputting the weighted posture detection data into a pre-trained posture prediction model to obtain predicted posture data; Performing Kalman filtering based on the predicted posture data and the weighted posture detection data to obtain corrected posture data; Determining posture information of the coal machine equipment based on the corrected posture data; The posture prediction model is trained in the following manner: Inputting sample detection data sets into a first data verification model to obtain sample verification data output by the first data verification model for each sample detection data set; Obtaining a reliability index of each sample detection data set based on the sample detection data set and the corresponding sample verification data; Regarding sample detection data sets whose reliability index does not meet a preset threshold condition as reliable sample detection data for data cleaning; Iterating a first posture prediction model based on the reliable sample detection data to obtain the posture prediction model; The method further comprises: Determining reliability influencing factors of each sub-posture detection data based on installation positions of different posture detection sensors; Calibrating initial weight information based on prior experience of the reliability influencing factors; 2. The method of claim 1, wherein, Iterating the initial weight information based on historical posture detection data and historical true posture information corresponding to the historical posture detection data to obtain the weight information. The method further comprises: After extracting a detection data verification set from the sample detection data set, inputting the detection data verification set into a reliability index verification model to obtain verification data output by the reliability index verification model, wherein the reliability index verification model comprises sub-verification modules configured as different verification parameters; ​ 3. The method of claim 1, wherein, ​ ​ The detection data check set corresponding to the verification data meeting the verification rule is taken as an example set, and the first verification module is iterated to obtain a target verification module, wherein the verification rule includes that the similarity distance between the verification data and the sample verification data is lower than a preset threshold; The target verification module is used to replace the sub-verification module corresponding to the verification data meeting the verification rule, and a reliable index verification model obtained after the replacement is taken as a target reliable index verification model; The sample detection data set is input into the target reliable index verification model to obtain target verification data output by the target reliable index verification model; Based on the target verification data and the sample verification data, reliable indexes of each sample detection data set are obtained.

4. The method of claim 1, wherein, The reliable indexes of each sample detection data set are obtained based on the sample detection data set and the corresponding sample verification data, including: After the detection data check set is determined in the sample detection data set, the detection data check set is input into a reliable index verification model to obtain verification data output by the reliable index verification model, and the reliable index verification model includes sub-verification modules configured as different verification parameters; The detection data check set is input into a first check model to obtain first check data output by the first check model; Based on the verification data and the first check data, the first check model is iterated to obtain a target check model after training; The sample detection data set is input into the target check model to obtain sample check data output by the target check model; Based on the sample check data and the corresponding sample verification data, reliable indexes of each sample detection data set are obtained.

5. The method according to any one of claims 1 to 4, characterized in that, The Kalman filtering is performed on the predicted attitude data and the weighted attitude detection data to obtain the corrected attitude data, and the method further includes: The Kalman filtering is performed on the first weight of the predicted attitude data and the second weight of the weighted attitude detection data to obtain the corrected attitude data.

6. An attitude information estimation device characterized by comprising: The device includes: The acquisition module is configured to acquire multi-source attitude detection data collected by sensors arranged on the coal machine equipment, and the multi-source attitude detection includes a plurality of sub-attitude detection data, and different sub-attitude detection data correspond to different sensors; The first determination module is configured to determine weighted attitude detection data based on weight information of each sub-attitude detection data; The second determination module is configured to perform range analysis on the weighted attitude detection data to obtain first attitude data; The prediction module is configured to input the weighted attitude detection data into a pre-trained attitude prediction model to obtain predicted attitude data; The filtering module is configured to perform Kalman filtering on the predicted attitude data and the weighted attitude detection data to obtain corrected attitude data; The third determination module is configured to determine attitude information of the coal machine equipment based on the corrected attitude data; The input module is configured to input a sample detection data set into a first data verification model to obtain sample verification data for each sample detection data set output by the first data verification model; a fourth determining module, configured to obtain a reliability index of each of the sample detection data sets based on the sample detection data set and the corresponding sample verification data; a fifth determining module, configured to take the sample detection data set whose reliability index does not satisfy a preset threshold condition as reliable sample detection data for data cleaning; a sixth determining module, configured to perform iteration on the first pose prediction model based on the reliable sample detection data to obtain the pose prediction model; the fourth determining module is at least configured to: input the sample detection data set and the corresponding sample verification data into the first pose prediction model and the second pose prediction model respectively to obtain first predicted pose data output by the first pose prediction model and second predicted pose data output by the second pose prediction model, wherein the sample detection data set is called from prior predicted pose data obtained by training the second pose prediction model; determine a prediction deviation between the first predicted pose data and the second predicted pose data, and determine a reliability index of the sample detection data set based on the prediction deviation, wherein the reliability index is negatively correlated with the prediction deviation.

7. A computer readable medium having stored thereon a computer program, characterized in that The program, when executed by a processing device, implements the steps of the method of any one of claims 1-5.

8. An electronic device, comprising: comprise: a storage device having a computer program stored thereon; a processing device configured to execute the computer program in the storage device to implement the steps of the method of any one of claims 1-5.

Images