Function robot operation data acquisition regulation and control method based on cloud platform driving

Abstract

The invention relates to the technical field of data acquisition regulation and control, in particular to a functional robot operation data acquisition regulation and control method based on cloud platform driving. The method comprises the following steps: acquiring an operation data curve set and a difference curve of each operation data curve in the operation data curve set; obtaining an initial lag value corresponding to the operation data curve according to the difference curve of the operation data curve; obtaining a correlation representation value between any two operation data curves in the operation data curve set according to an envelope line of the operation data curve, a peak value data point on the operation data curve and an initial lag value corresponding to the operation data curve; and according to the correlation characterization value, operation data of the functional robot are collected, regulated and controlled. According to the method, the obtained correlation characterization value is accurate and reliable, so that the reliability and the accuracy when the operation data of the functional robot is collected, regulated and controlled can be ensured according to the accurate and reliable correlation characterization value.

Description

Technical Field

[0001] The present invention relates to the technical field of data acquisition and control, and in particular to a method for acquiring and controlling the operation data of a functional robot driven by a cloud platform. Background Art

[0002] Since functional robots have the functions of improving work efficiency and quality, serving human life, and expanding or extending the scope of human activities and capabilities, they are currently widely used in many fields, such as medical care, agriculture, logistics, and household services. In order to improve the work efficiency and product quality of robots, help enterprises optimize production processes and management strategies, and realize the intelligent and refined production process, it is usually necessary to collect and regulate the operating data of functional robots. In other words, collecting and regulating the operating data of functional robots is crucial to improving the work efficiency and product quality of robots and helping enterprises optimize production processes and management strategies.

[0003] Since the operating data of the functional robot is complex and numerous, in order to ensure the reliability and accuracy of the collection and regulation of the operating data of the functional robot, the prior art usually analyzes the correlation between different operating data types to ensure the reliability and accuracy of the collection and regulation of the operating data of the functional robot; however, the functional robot system usually has a response delay phenomenon, which will cause the collected data to have a time lag feature, and the existence of time lag will cause the correlation between the calculated different operating data types to have low reliability or accuracy, resulting in low reliability and accuracy when the operating data of the functional robot is subsequently collected and regulated based on the correlation. Therefore, how to reliably and accurately obtain the correlation between different operating data types to ensure or improve the reliability and accuracy of the subsequent collection and regulation of the operating data of the functional robot is a problem that needs to be solved urgently. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a method for collecting and controlling the operation data of a functional robot driven by a cloud platform. The technical solutions adopted are as follows: An embodiment of the present invention provides a method for collecting and controlling the operation data of a functional robot driven by a cloud platform, comprising the following steps: Acquire an operation data curve set corresponding to the functional robot in a target monitoring time period and a differential curve of each operation data curve in the operation data curve set; Obtaining an initial hysteresis value corresponding to the operating data curve according to a differential curve of the operating data curve; Obtaining a correlation characterization value between any two operation data curves in the set of operation data curves according to an envelope of the operation data curve, a peak data point on the operation data curve, and an initial hysteresis value corresponding to the operation data curve; According to the correlation characterization value, the operation data of the functional robot is collected and regulated.

[0005] Beneficial effect: The present invention first obtains the operation data curve set corresponding to the target monitoring time period of the functional robot and the differential curve of each operation data curve in the operation data curve set; then, according to the differential curve of the operation data curve, the initial hysteresis value corresponding to the operation data curve is obtained; then, according to the envelope of the operation data curve, the peak data point on the operation data curve and the initial hysteresis value corresponding to the operation data curve, the correlation characterization value between any two operation data curves in the operation data curve set is obtained; finally, according to the correlation characterization value, the operation data of the functional robot is collected and regulated. The correlation characterization value between any two operation data curves obtained by the present invention is relatively accurate and reliable, and based on the accurate and reliable correlation characterization value, the reliability and accuracy of the subsequent collection and regulation of the operation data of the functional robot can be guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0007] Figure 1 The present invention provides a method for collecting and controlling the operation data of a functional robot driven by a cloud platform. DETAILED DESCRIPTION

[0008] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the embodiments of the present invention.

[0009] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0010] This embodiment provides a method for collecting and controlling the operation data of a functional robot driven by a cloud platform, which is described in detail as follows: like Figure 1 As shown, the cloud platform-driven functional robot operation data collection and control method includes the following steps: Step S001, obtaining an operation data curve set corresponding to a functional robot in a target monitoring time period and a differential curve of each operation data curve in the operation data curve set.

[0011] At present, usually the correlation between different operating data types is analyzed to ensure the reliability and accuracy of the subsequent collection and regulation of the operating data of the functional robot; however, because the functional robot system usually has a response delay, this phenomenon will cause the collected data to have a time lag feature, and the existence of time lag will cause the calculated correlation between different operating data types to have low reliability or accuracy, resulting in low reliability when the operating data of the functional robot is subsequently collected and regulated based on the correlation. The response delay or time lag means that the data collected at the current moment may actually be data at a historical moment before the current moment. In order to ensure the reliability of the subsequent collection and regulation of the operating data of the functional robot, this embodiment requires reliable and accurate Obtaining the correlation between different types of operation data, that is, the main purpose of this embodiment is to reliably and accurately obtain the correlation between different types of operation data, that is, to reliably and accurately obtain the correlation characterization value between any two operation data curves, wherein one operation data curve represents one type of operation data, and before obtaining the correlation characterization value between any two operation data curves, it is necessary to first obtain the operation data curve set corresponding to the target monitoring time period of the functional robot and the differential curves of each operation data curve in the operation data curve set, that is, the operation data curve set and the differential curves of each operation data curve in the operation data curve set are the basis for the subsequent determination of the correlation characterization value, so in this embodiment, the specific process of obtaining the operation data curve set and the differential curves of each operation data curve in the operation data curve set is: First, obtain an operation data type set; and in specific applications, the implementer needs to set the operation data type set according to actual conditions. For example, the operation data types in the operation data type set set in this embodiment include at least the temperature data of the functional robot, the movement speed of the functional robot, the position data of the functional robot, the posture angle of the functional robot, the degree of freedom of the functional robot, and the acceleration of the functional robot. That is, in this embodiment, the operation data type set consists of the temperature data of the functional robot, the movement speed of the functional robot, the position data of the functional robot, the posture angle of the functional robot, the degree of freedom of the functional robot, and the acceleration of the functional robot, and different operation data types can be collected through different sensors or built-in devices; in addition, for the sake of ease of understanding and analysis, this embodiment will take any functional robot as an example for analysis.

[0012] However, different types of operating data may not all be represented by one-dimensional data, and there may be cases where they are represented by multi-dimensional data. Therefore, in this embodiment, the collected data will be represented by vectors in the future; for example, the temperature data of the functional robot is one-dimensional data, and the temperature data of the functional robot collected subsequently is represented by a one-dimensional vector; the movement speed and acceleration of the functional robot are both two-dimensional data, respectively composed of size and direction, and the movement speed and acceleration of the robot collected subsequently are both represented by two-dimensional vectors; the position data of the functional robot is two-dimensional data or three-dimensional data. If it is two-dimensional data Then the position data refers to the coordinate information of the functional robot in the geodetic coordinate system, and the subsequently collected position data is also represented by a two-dimensional vector. If it is three-dimensional data, it includes the longitude, latitude and elevation of the functional robot's location, and the subsequently collected position data of the robot is also represented by a three-dimensional vector; the attitude angle of the functional robot is three-dimensional data, namely the roll angle, pitch angle and yaw angle, so the subsequently collected attitude angle of the robot is also represented by a three-dimensional vector; the degree of freedom of the functional robot is one-dimensional data, so the subsequently collected degree of freedom of the robot is also represented by a one-dimensional vector.

[0013] Therefore, based on the above analysis, it can be known that this embodiment will then obtain the operation data vectors corresponding to various operation data types in the operation data type set at each monitoring time in the target monitoring time period; for example, if a certain operation data type is the temperature data of a functional robot, then the operation data vector corresponding to the operation data type has only one parameter, which is the temperature data, and if a certain operation data type is the position data of a functional robot, then the operation data vector corresponding to the operation data type may have three parameters, namely longitude, latitude and elevation.

[0014] Therefore, in the target monitoring time period, multiple vectors can be obtained for each operating data type; then, all operating data vectors corresponding to the same operating data type are sorted in order of acquisition time, and the sorted sequence is recorded as the operating data vector sequence corresponding to the corresponding operating data type; for example, for any operating data type A0 in the operating data type set, a set constructed by all operating data vectors corresponding to the operating data type A0 obtained in the target monitoring time period is recorded as a to-be-sorted set, and all operating data vectors in the to-be-sorted set are sorted in order of acquisition time, and the sorted sequence is recorded as the operating data vector sequence corresponding to the operating data type A0.

[0015] In specific applications, the implementer needs to set the target monitoring time period according to the actual situation. For example, in this embodiment, the time period from the start of the functional robot's operation to the current moment is recorded as the target monitoring time period; and in this embodiment, the collection time intervals of different operating data types are also set to be consistent, that is, if there are two types of operating data that need to be collected in this embodiment, then the vectors corresponding to these two types of operating data need to be collected at each monitoring moment.

[0016] Then, a dimensionality reduction algorithm is used to reduce the dimensionality of the operating data vectors in the operating data vector sequence corresponding to each operating data type to obtain one-dimensional data corresponding to each operating data vector in the operating data vector sequence, and the one-dimensional data corresponding to the operating data vector is recorded as the reduced dimensionality data of the corresponding operating data vector; that is, this embodiment requires that all vectors be reduced to one-dimensional data, and in order to reduce the amount of calculation, this embodiment requires that no dimensionality reduction operation be performed on vectors consisting of only one parameter; and in specific applications, the implementer needs to select a dimensionality reduction algorithm based on actual conditions, such as this embodiment can use the principal component analysis algorithm to achieve dimensionality reduction.

[0017] Next, a sequence constructed by the dimensionality reduction data of all the operating data vectors in the operating data vector sequence corresponding to various operating data types is obtained, and recorded as the dimensionality reduction data sequence corresponding to the corresponding operating data type; then, the dimensionality reduction data sequence corresponding to various operating data types is preprocessed, and the preprocessed sequence is recorded as the target operating data sequence corresponding to the corresponding operating data type, and the data in the target operating data sequence are all target operating data; and the preprocessing process includes but is not limited to adjusting the outliers in the dimensionality reduction data sequence, normalizing the dimensionality reduction data in the dimensionality reduction data sequence, etc.

[0018] After obtaining the target operation data sequence, the time mark value of the target operation data is obtained immediately, and the specific acquisition method is: for any target operation data W in any target operation data sequence, if the target operation data W is the operation data vector obtained at the rth monitoring moment in the target monitoring time period, and is the data after dimension reduction and preprocessing, then this embodiment records the time corresponding to the rth monitoring moment as the time mark value of the target operation data W.

[0019] Therefore, the time stamp value of each target operating data can be obtained through the above process. After the time stamp value of the target operating data is obtained, the present embodiment will obtain the operating data curves corresponding to the various operating data types according to the target operating data sequences corresponding to the various operating data types and the time stamp values ​​of the various target operating data in the target operating data sequences. The specific process of obtaining the operating data curves is as follows: For the target operation data sequence A1 corresponding to any operation data type A0 in the operation data type set: first, construct the data space of the target operation data sequence A1, and the horizontal coordinate of the data space corresponding to the target operation data sequence A1 is the time mark value, and the vertical coordinate is the value of the target operation data that is the same as the operation data type A0; then map all the target operation data in the target operation data sequence A1 and the time mark value of the target operation data to the data space to obtain all the data points in the data space, and connect all the data points in the data space in order from small to large in terms of the horizontal coordinate value, and record the connected curve as the operation data curve corresponding to the operation data type A0, and the data points on the operation data curve are composed of the target operation data in the target operation data sequence A1 and the time mark value of the target operation data, and the number of data points in the data space corresponding to the target operation data sequence A1 is consistent with the total number of data in the target operation data sequence A1.

[0020] Therefore, this embodiment can obtain the operating data curve corresponding to each operating data type in the operating data type set according to the above process; then the set constructed by the operating data curves corresponding to all operating data types in the operating data type set is recorded as the operating data curve set corresponding to the functional robot in the target monitoring time period.

[0021] After obtaining the operation data curve set, the differential curves of each operation data curve in the operation data curve set will be obtained, and the specific process of obtaining the differential curve is as follows: For any operating data curve C in the operating data curve set: first, according to two adjacent data points on the operating data curve C, a differential vector sequence corresponding to the operating data curve C is obtained, and the differential vector consists of a first mark value and a differential value, and the differential vectors in the differential vector sequence are arranged in ascending order according to the first mark value; then a mapping space of the differential vector sequence is constructed, and the horizontal coordinate of the mapping space is the first mark value, and the vertical coordinate value is the differential value; then each differential vector in the differential vector sequence is mapped to the mapping space, and all data points in the mapping space after the mapping is completed are recorded as differential points, and the differential points in the mapping space are connected in sequence according to the ascending order of the horizontal coordinate value, and then the connected curve is recorded as the differential curve C0 of the operating data curve C.

[0022] In addition, the method for obtaining the first mark value and the differential value in the differential vector is: the value obtained by subtracting the vertical coordinate value of the bth data point on the running data curve C from the vertical coordinate value of the b+1th data point on the running data curve C is used as the differential value in the bth differential vector in the differential vector sequence corresponding to the running data curve C, and the first mark value in the bth differential vector in the differential vector sequence corresponding to the running data curve C is recorded as b.

[0023] Therefore, this embodiment can obtain the operation data curve set and the difference curve of each operation data curve in the operation data curve set through the above process, and the difference curve is used to subsequently determine the initial hysteresis value corresponding to the operation data curve.

[0024] Step S002: obtaining an initial hysteresis value corresponding to the operation data curve according to a differential curve of the operation data curve.

[0025] Next, this embodiment will obtain the initial hysteresis value corresponding to the operation data curve according to the differential curve of the operation data curve, and the initial hysteresis value corresponding to the operation data curve is an important indicator for the subsequent acquisition of the correlation characterization value; in addition, for ease of understanding, the following will take the acquisition process of the initial hysteresis value corresponding to the operation data curve C as an example for description, that is, the specific acquisition process of the initial hysteresis value corresponding to the operation data curve C is: First, the difference curve of the running data curve C is recorded as curve D, and then the change characterization value corresponding to each peak difference point on the curve D is obtained; and the specific process of obtaining the change characterization value corresponding to the peak difference point is: For any peak differential point c on curve D: first obtain the tangent slope of each differential point on curve D; then on curve D, record the set constructed by all differential points located on the left side of the peak differential point c as the first set; then obtain the cumulative sum of the tangent slopes of all differential points in the first set, and record it as the comprehensive slope; then calculate the difference between the tangent slope of the peak differential point c and the comprehensive slope, and record it as the first difference; then obtain the hyperbolic tangent function value of the first difference, and record it as the first characterization value; then obtain the mean of the ordinate values ​​of all differential points in the first set, and record it as the first mean; then record the ratio of the ordinate value of the peak differential point c to the absolute value of the first mean as the first ratio; then obtain the hyperbolic tangent function value of the first ratio, and record it as the second characterization value; finally, obtain the mean of the first characterization value and the second characterization value, and record it as the change characterization value corresponding to the peak differential point c.

[0026] In addition, in this embodiment, the specific expression for obtaining the change characterization value corresponding to the peak difference point c is: in, is the change characterization value corresponding to the peak difference point c, tanh() is the hyperbolic tangent function, is the tangent slope of the peak difference point c, C0 is the total number of difference points in the first set, is the tangent slope of the qth difference point in the first set, is the comprehensive slope, is the first difference, is the ordinate value of the peak difference point c, is the ordinate value of the qth differential point in the first set, is the first mean, is the first ratio. And when The closer the value is to 1, the more likely it is that the running data type corresponding to curve D will have a response delay or time lag during acquisition. The closer the value is to 1, the less likely it is that the running data type corresponding to curve D will have a response delay or time lag during acquisition.

[0027] Therefore, according to the above process, the change characterization value corresponding to each peak differential point can be obtained. After obtaining the change characterization value, it is then determined whether the change characterization values ​​corresponding to all the peak differential points on curve D do not belong to the preset judgment interval. If so, it indicates that the operating data type corresponding to curve D is less likely to have a response delay or time lag during collection, so 0 is used as the initial lag value corresponding to the operating data curve C; and if it is determined that there is a peak differential point on curve D whose change characterization value belongs to the preset judgment interval, then it indicates that the operating data type corresponding to curve D is more likely to have a response delay or time lag during collection. Therefore, this embodiment needs to obtain a set constructed by the horizontal coordinates of all the peak differential points whose change characterization values ​​belong to the preset judgment interval on curve D, and record it as the set to be selected, and select the smallest horizontal coordinate value in the set to be selected as the initial lag value corresponding to the operating data curve C.

[0028] The essential meaning of the peak differential point in this embodiment is the same as that of the peak point, and in specific applications, the implementer can determine the peak differential point according to actual conditions. For example, for any differential point on the differential curve, if the value obtained by multiplying the first-order derivative of the adjacent differential point on the left side of the differential point by the first-order derivative of the adjacent differential point on the right side of the differential point is less than 0, then the differential point is the peak differential point.

[0029] In addition, in a specific application, the implementer needs to set the preset judgment interval according to the actual situation. For example, in this embodiment, the preset judgment interval is set to (0.6, 1].

[0030] Step S003, obtaining a correlation characterization value between any two operation data curves in the operation data curve set according to the envelope of the operation data curve, the peak data point on the operation data curve and the initial hysteresis value corresponding to the operation data curve.

[0031] Next, this embodiment will further determine the correlation characterization value between any two operation data curves in the operation data curve set by analyzing the envelope of the operation data curve, the peak data point on the operation data curve, and the initial hysteresis value corresponding to the operation data curve; and for the sake of ease of understanding, this embodiment will be described by taking the acquisition process of the correlation characterization value between the i-th operation data curve and the j-th operation data curve in the data curve set as an example, i is not equal to j, then the specific acquisition process of the correlation characterization value between the i-th operation data curve and the j-th operation data curve is: First, according to the initial lag values ​​corresponding to the ith operation data curve and the jth operation data curve, the to-be-determined lag difference interval between the ith operation data curve and the jth operation data curve is obtained, and subsequently the target lag difference is determined by analyzing each to-be-determined lag difference in the to-be-determined lag difference interval, and the target lag difference is an important parameter for accurately obtaining the correlation characterization value between the ith operation data curve and the jth operation data curve. Therefore, in this embodiment, the specific acquisition process of the to-be-determined lag difference interval is as follows: the value obtained by subtracting the initial lag value corresponding to the jth operation data curve from the initial lag value corresponding to the ith operation data curve is recorded as the initial difference; then the value obtained by subtracting the preset first constant from the initial difference is obtained, and recorded as the first characteristic value, and the value obtained by adding the preset first constant to the initial difference is obtained, and recorded as the second eigenvalue; then, the value interval composed of the first eigenvalue and the second eigenvalue is recorded as the lag difference interval to be determined between the ith operation data curve and the jth operation data curve, and all integers in the lag difference interval to be determined are recorded as lag differences to be determined; in addition, in this embodiment, the initial difference is not directly used as the subsequent target lag difference, but the lag difference interval to be determined is constructed by presetting the first constant, and the target lag difference is determined based on the preferred degree of each lag difference to be determined in the lag difference interval to be determined. This is to obtain a more reliable and accurate target lag difference, because the target lag difference can reflect the time lag difference between the operation data type corresponding to the ith operation data curve and the operation data type corresponding to the jth operation data curve, and the time lag difference is the key to the subsequent accurate determination of the correlation.

[0032] In a specific application, the implementer needs to determine the preset first constant according to the actual situation. For example, in this embodiment, the preset first constant is set to 10.

[0033] After obtaining the lag difference interval to be determined, the present embodiment will then obtain the degree of preference corresponding to each lag difference to be determined in the lag difference interval to be determined based on the lag difference interval to be determined, the envelope of the i-th operating data curve, the envelope of the j-th operating data curve, the peak data point on the i-th operating data curve, and the peak data point on the j-th operating data curve. The degree of preference is the basis for subsequently determining the target lag difference. After obtaining the degree of preference, the characteristic data point sequence pairs corresponding to each lag difference to be determined in the lag difference interval to be determined are then obtained.

[0034] Then, in the interval of lag difference values ​​to be determined, the lag difference value to be determined corresponding to the maximum preferred degree is selected as the target lag difference value between the ith operation data curve and the jth operation data curve; then, the two characteristic data point sequences constituting the characteristic data point sequence pair corresponding to the target lag difference value are respectively recorded as the first sequence and the second sequence, and the sequence constructed by the ordinate values ​​of all data points in the first sequence is recorded as the first sequence to be analyzed, and the sequence constructed by the ordinate values ​​of all data points in the second sequence is recorded as the second sequence to be analyzed; then, the Spearman correlation coefficient between the first sequence to be analyzed and the second sequence to be analyzed is calculated and recorded as the initial correlation coefficient, and then the absolute value of the initial correlation coefficient is used as the correlation representation value between the ith operation data curve and the jth operation data curve, and the larger the correlation representation value, the stronger the correlation between the operation data type corresponding to the ith operation data curve and the operation data type corresponding to the jth operation data curve; and since the process of obtaining the Spearman correlation coefficient between the two sequences is a well-known technology, it will not be described in detail in this embodiment.

[0035] Next, this embodiment will specifically describe the process of acquiring the characteristic data point sequence pair corresponding to the to-be-determined lag difference value, and for ease of understanding, this embodiment will take the process of acquiring the characteristic data point sequence pair corresponding to the h-th to-be-determined lag difference value in the to-be-determined lag difference value interval as an example for description, that is, the process of acquiring the characteristic data point sequence pair corresponding to the h-th to-be-determined lag difference value is: First, the hth lag difference to be determined is recorded as G; if G is judged to be a positive number, it means that the time delay or response delay of the operating data type corresponding to the i-th operating data curve during collection is greater than the time delay or response delay of the operating data type corresponding to the j-th operating data curve during collection. In order to ensure the accuracy of subsequent correlation calculations, the i-th operating data curve or the j-th operating data curve should be moved. If only the i-th operating data curve is moved, the i-th operating data curve should be moved to the left. If only the j-th operating data curve is moved, the j-th operating data curve should be moved to the right. After the movement is completed, the v+1th point on the i-th operating data curve is aligned with the vth point on the j-th operating data curve. When determining the preference and correlation of the h-th lag difference to be determined later, it is only necessary to analyze the existence The points with alignment relationship can be used, so if G is judged to be a positive number, then the remaining data points except the first G data points on the ith running data curve and the remaining data points except the last G data points on the jth running data curve are all recorded as characteristic data points, that is, if G is judged to be a positive number, the first G data points on the ith running data curve and the last G data points on the jth running data curve are all data points without alignment relationship, that is, they are not characteristic data points, and will also affect the subsequent correlation analysis; for example, if there are 5 data points on the ith running data curve and the jth running data curve, and the value of G is 1, then at this time, the first point on the ith running data curve and the last point on the jth running data curve are both data points without alignment relationship, that is, they are not characteristic data points.

[0036] In addition, before the i-th operating data curve and the j-th operating data curve are moved, the i-th operating data curve and the j-th operating data curve should be aligned in time, that is, the horizontal coordinates of the v-th point on the i-th operating data curve and the v-th point on the j-th operating data curve are consistent before moving, but when there is a time delay or response delay, that is, when the value of G is not 0, in order to ensure the accuracy of subsequent correlation calculations, time compensation is required to eliminate the impact of time delay or response delay on correlation accuracy, that is, to eliminate time delay or response delay by moving the i-th operating data curve or moving the j-th operating data curve.

[0037] Based on the above analysis, it can be seen that if G is judged to be a negative number, then the remaining data points on the i-th operating data curve except the last G data points and the remaining data points on the j-th operating data curve except the first G data points are all recorded as characteristic data points; if G is judged to be 0, then all data points on the i-th operating data curve and all data points on the j-th operating data curve are all recorded as characteristic data points.

[0038] Next, a sequence constructed by all characteristic data points on the i-th operating data curve is obtained and recorded as characteristic data point sequence h1, and a sequence constructed by all characteristic data points on the j-th operating data curve is obtained and recorded as characteristic data point sequence h2; then, a sequence pair consisting of characteristic data point sequence h1 and characteristic data point sequence h2 is recorded as a characteristic data point sequence pair corresponding to the h-th lag difference to be determined; and the positional relationship between each data point in the characteristic data point sequence is consistent with its positional relationship on the corresponding operating data curve.

[0039] Therefore, according to the above process of obtaining the characteristic data point sequence pair corresponding to the h-th lag difference value to be determined, the present embodiment can obtain the characteristic data point sequence pair corresponding to each lag difference value to be determined; next, the present embodiment describes the process of obtaining the preferred degree corresponding to each lag difference value to be determined in the lag difference value interval to be determined. Similarly, the present embodiment also describes the process of obtaining the preferred degree corresponding to the h-th lag difference value to be determined as an example. Then, the process of obtaining the preferred degree corresponding to the h-th lag difference value to be determined is: First, the upper envelope and lower envelope of the ith operating data curve and the upper envelope and lower envelope of the jth operating data curve are obtained, and the process of obtaining the upper envelope and lower envelope of the curve is a well-known technology, so this embodiment will not be described in detail; then, according to the upper envelope and lower envelope of the ith operating data curve and the upper envelope and lower envelope of the jth operating data curve, the envelope data point pairs corresponding to each data point on the ith operating data curve and the envelope data point pairs corresponding to each data point on the jth operating data curve are obtained; then, according to the hth lag difference to be determined, the ith operating data curve and the jth operating data curve, the first set of data point pairs to be analyzed and the second set of data point pairs to be analyzed are obtained, and one of the two data points in the data point pairs to be analyzed is located on the ith operating data curve, and the other is located on the jth operating data curve.

[0040] Next, the sum of the total number of peak data points on the i-th running data curve and the total number of peak data points on the j-th running data curve is obtained, and recorded as the first quantity characterization value. Then, the number of all data points in the second set of data point pairs to be analyzed is obtained, and recorded as the second quantity characterization value, and the second quantity characterization value is twice the total number of data point pairs in the second data point pairs to be analyzed; then, the ratio of the second quantity characterization value to the first quantity characterization value is obtained, and recorded as the peak data point overlap rate.

[0041] Then, obtain the similarity characterization values ​​of each first pair of data points to be analyzed in the set of first pairs of data points to be analyzed, and record the mean of the similarity characterization values ​​of all pairs of data points to be analyzed in the set of first pairs of data points to be analyzed as the second mean; then negative correlation mapping is performed on the second mean, and the value obtained after negative correlation mapping is recorded as the envelope amplitude similarity, that is, the envelope amplitude similarity is exp(-E), and exp() is an exponential function with a constant e as the base, and E is the second mean. Then, weighted summation is performed on the peak data point overlap rate and the envelope amplitude similarity, and the value obtained by weighted summation is recorded as the preferred degree corresponding to the hth lag difference to be determined. The specific expression of the preferred degree corresponding to the hth lag difference to be determined is: ,in, is the peak data point coincidence rate, is the envelope amplitude similarity ratio, is the first weight value, is the second weight value; and in a specific application, the implementer needs to set the first weight value and the second weight value according to the actual situation. For example, in this embodiment, the first weight value and the second weight value are set to 0.45 and 0.55 respectively.

[0042] In addition, when the preference degree corresponding to the h-th lag difference value to be determined is greater, it indicates that after moving based on the h-th lag difference value to be determined, the alignment degree between the i-th operating data curve and the j-th operating data curve is better, and the probability that the subsequent h-th lag difference value to be determined is the target lag difference value is also greater; when the preference degree corresponding to the h-th lag difference value to be determined is smaller, it indicates that after moving based on the h-th lag difference value to be determined, the alignment degree between the i-th operating data curve and the j-th operating data curve is worse, and the probability that the subsequent h-th lag difference value to be determined is the target lag difference is also smaller.

[0043] In the above process of obtaining the preferred degree corresponding to the hth lag difference to be determined, the specific process of obtaining the envelope data point pair corresponding to the data point is: for the g-th data point on the ith operating data curve: first, in the upper envelope line of the ith operating data curve, obtain the data point with the same horizontal coordinate value as the g-th data point, and record it as the upper envelope data point of the g-th data point; then, in the lower envelope line of the ith operating data curve, obtain the data point with the same horizontal coordinate value as the g-th data point, and record it as the lower envelope data point of the g-th data point; then, the data point pair consisting of the upper envelope data point and the lower envelope data point of the obtained g-th data point is recorded as the envelope data point pair corresponding to the g-th data point; since the method for obtaining the envelope data point pair corresponding to the data point on the j-th operating data curve is the same as the method for obtaining the envelope data point pair corresponding to the g-th data point, it will not be described in detail.

[0044] In the above process of obtaining the preferred degree corresponding to the hth lag difference to be determined, the method for obtaining the first set of data point pairs to be analyzed and the second set of data point pairs to be analyzed is: first, obtain the characteristic marking value of each characteristic data point in the characteristic data point sequence h1 and the characteristic marking value of each characteristic data point in the characteristic data point sequence h2, and the characteristic marking value of the kth characteristic data point in the characteristic data point sequence h1 is k, and the characteristic marking value of the kth characteristic data point in the characteristic data point sequence h2 is also k; then, in the characteristic data point sequence h1 and the characteristic data point sequence h2, the data pairs consisting of two data points with the same characteristic marking value are all recorded as first data point pairs to be analyzed, and the set constructed by all the obtained first data point pairs to be analyzed is recorded as the first set of data point pairs to be analyzed. For any first data point pair to be analyzed in the first set of data point pairs to be analyzed, if both data points in the first data point pair to be analyzed are peak data points, then the first data point pair to be analyzed is recorded as the second data point pair to be analyzed; then the set constructed by all second data point pairs to be analyzed in the first set of data point pairs to be analyzed is recorded as the second set of data point pairs to be analyzed.

[0045] In addition, the method for determining the peak data point in this embodiment is the same as the method for determining the peak difference point described above, and therefore will not be described in detail.

[0046] In the above process of obtaining the preferred degree corresponding to the hth lag difference to be determined, the process of obtaining the similarity characterization value of the first data point pair to be analyzed is, for any first data point pair H to be analyzed in the first data point pair set to be analyzed: first, the two data points constituting the first data point pair H to be analyzed are respectively recorded as the first data point and the second data point, and the envelope data point pair corresponding to the first data point and the envelope data point pair corresponding to the second data point are respectively recorded as the first envelope data point pair and the second envelope data point pair; then, the absolute value of the difference between the ordinate values ​​of the two data points constituting the first envelope data point pair is obtained, and recorded as the first envelope difference, and the absolute value of the difference between the ordinate values ​​of the two data points constituting the second envelope data point pair is obtained, and recorded as the second envelope difference; then, the absolute value of the difference between the first envelope difference and the second envelope difference is obtained, and recorded as the similarity characterization value of the data point pair H to be analyzed.

[0047] Therefore, this embodiment can obtain the correlation characterization value between any two operating data curves in the operating data curve set through the above process; and the larger the correlation characterization value, the stronger the correlation between the two operating data types corresponding to the two operating data curves, and vice versa; and the correlation characterization value obtained according to the above method is relatively accurate and reliable.

[0048] Step S004: collecting and regulating the functional robot according to the correlation characterization value.

[0049] Next, this embodiment collects and controls the functional robot according to the correlation characterization value between any two operating data curves in the operating data curve set, and the functional robot can be collected and controlled more reliably and accurately based on the correlation characterization value obtained above, and collecting and controlling the functional robot based on correlation is a known technology, so this embodiment only briefly describes the collection and control process.

[0050] For any operating data curve C in the operating data curve set: the set constructed by the operating data curves in the operating data curve set except the operating data curve C is recorded as a subset; and the set constructed by the correlation characterization values ​​between the operating data curve C and each operating data curve in the subset is recorded as a characterization value set; then, it is determined whether there is a correlation characterization value greater than a preset correlation threshold in the characterization value set. If so, it is determined that the correlation between the operating data type corresponding to the operating data curve C and other operating data types is strong, and the operating data type corresponding to the operating data curve C is marked as strongly correlated. If it is determined that there is no correlation characterization value greater than the preset correlation threshold in the characterization value set, then it is determined that the correlation between the operating data type corresponding to the operating data curve C and other operating data types is weak, and the operating data type corresponding to the operating data curve C is marked as weakly correlated.

[0051] In a specific application, the implementer needs to set the preset correlation threshold according to the actual situation. For example, in this embodiment, the preset correlation threshold is set to 0.7.

[0052] Therefore, this embodiment can obtain the marked conditions of the strength of the correlation of various operating data types in the operating data type set through the above process, and then collect and regulate the operating data of the functional robot in the next monitoring time period according to the marked conditions of the strength of the correlation. Specifically, in the next monitoring time period, when collecting all operating data types marked as strongly correlated, the same or unified interface is used for collection. Such a setting can not only optimize the data collection strategy, reduce redundant data, reduce the complexity of subsequent data processing, reduce the amount of data required to be processed in the data collection stage, improve the efficiency of data processing, reduce unnecessary data collection and ensure the consistency of data format and structure, and enhance decision support, but also contribute to the subsequent regulation of the functional robot. The choice of a unified interface for all operating data types with strong correlation during collection needs to be determined by the implementer according to the actual situation. In the next monitoring time period, when collecting the operating data types marked as weakly correlated, normal collection can be performed. If the operating data type A0 is marked as weakly correlated, then in the next monitoring time period, the method for collecting the operating data type A0 in the historical monitoring time period can be used for collection. After completing the operation data collection of the functional robot for the next monitoring time period, the collected data will be transmitted to the central control system or the cloud platform, and the data will be analyzed at the central control system or the cloud platform to obtain analysis results, which include but are not limited to the identification of key performance indicators and abnormal patterns, the prediction of the operation status and performance of the functional robot, etc., and then the operation parameters of the robot will be adjusted based on the analysis results, such as adjusting the path, modifying task allocation, etc., and the adjustment process is realized through the control system.

[0053] In addition, by collecting and regulating the operating data of functional robots, the operating efficiency and accuracy of the robots can be improved, potential problems can be identified to avoid equipment failures, and reasonable maintenance and servicing plans can be formulated based on the data to extend the service life, monitor and regulate the operation of the robots, ensure safe operation, and avoid accidents.

[0054] In summary, this embodiment first obtains the operation data curve set corresponding to the target monitoring time period of the functional robot and the differential curves of each operation data curve in the operation data curve set; then, according to the differential curve of the operation data curve, the initial hysteresis value corresponding to the operation data curve is obtained; then, according to the envelope of the operation data curve, the peak data point on the operation data curve and the initial hysteresis value corresponding to the operation data curve, the correlation characterization value between any two operation data curves in the operation data curve set is obtained; finally, according to the correlation characterization value, the operation data of the functional robot is collected and regulated. And the correlation characterization value between any two operation data curves obtained by this embodiment is relatively accurate and reliable, and based on the accurate and reliable correlation characterization value, the reliability and accuracy of the subsequent collection and regulation of the operation data of the functional robot can be guaranteed.

[0055] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for collecting and controlling the operation data of a functional robot driven by a cloud platform, characterized in that: The method comprises the following steps: Acquire an operation data curve set corresponding to the functional robot in a target monitoring time period and a differential curve of each operation data curve in the operation data curve set; Obtaining an initial hysteresis value corresponding to the operating data curve according to a differential curve of the operating data curve; Obtaining a correlation characterization value between any two operation data curves in the set of operation data curves according to an envelope of the operation data curve, a peak data point on the operation data curve, and an initial hysteresis value corresponding to the operation data curve; According to the correlation characterization value, the operation data of the functional robot is collected and regulated.

2. The method for collecting and controlling the operation data of a functional robot driven by a cloud platform as claimed in claim 1, characterized in that: The method for obtaining the operation data curve set includes: At each monitoring moment in the target monitoring time period, obtain the operation data vectors corresponding to the various operation data types in the operation data type set; sort all the operation data vectors corresponding to the same operation data type in chronological order to obtain the operation data vector sequence corresponding to the various operation data types; reduce the dimension of the operation data vectors in the operation data vector sequence to obtain one-dimensional data corresponding to the operation data vector, and record the one-dimensional data corresponding to the operation data vector as the reduced dimension data of the corresponding operation data vector; record the sequence constructed by the reduced dimension data of all the operation data vectors in the operation data vector sequence corresponding to the operation data type as the reduced dimension data sequence corresponding to the corresponding operation data type; preprocess the reduced dimension data sequence corresponding to the operation data type, and record the preprocessed sequence as the target operation data sequence corresponding to the corresponding operation data type; For any target operation data W in any target operation data sequence, if the target operation data W is the data of the operation data vector obtained at the r-th monitoring moment in the target monitoring time period after dimension reduction and preprocessing, the time corresponding to the r-th monitoring moment is recorded as the time mark value of the target operation data W; For a target operation data sequence A1 corresponding to any operation data type A0: construct a data space of the target operation data sequence A1, wherein the horizontal coordinate of the data space is the time mark value and the vertical coordinate is the target operation data; map all target operation data and the time mark value of the target operation data in the target operation data sequence A1 to the data space, obtain all data points in the data space, and sequentially connect all data points in the data space, and record the connected curve as the operation data curve corresponding to the operation data type A0; A set constructed by the operation data curves corresponding to all the operation data types in the operation data type set is recorded as an operation data curve set.

3. The method for collecting and controlling the operation data of a functional robot driven by a cloud platform as claimed in claim 1, characterized in that: The method for obtaining the differential curve of the operation data curve comprises: For any operation data curve C: according to two adjacent data points on the operation data curve C, a differential vector sequence corresponding to the operation data curve C is obtained, wherein the differential vector consists of a first mark value and a differential value; a mapping space of the differential vector sequence is constructed, wherein the horizontal axis of the mapping space is the first mark value and the vertical axis is the differential value; each differential vector in the differential vector sequence is mapped to the mapping space, and all data points in the mapping space after the mapping is completed are recorded as differential points, and the differential points in the mapping space are connected in sequence from small to large in the horizontal axis value, and the connected curve is recorded as the differential curve C0 of the operation data curve C, the differential value in the b-th differential vector in the differential vector sequence is the vertical axis difference between the b+1-th data point and the b-th data point on the operation data curve C, and the first mark value in the b-th differential vector in the differential vector sequence is b.

4. The method for collecting and controlling the operation data of a functional robot driven by a cloud platform as claimed in claim 3, characterized in that: The method for obtaining the initial hysteresis value corresponding to the operation data curve includes: For the operating data curve C: the differential curve of the operating data curve C is recorded as curve D, and the change characterization value of each peak differential point on the curve D is obtained; if the change characterization values ​​corresponding to all the peak differential points on the curve D do not belong to the preset judgment interval, 0 is used as the initial lag value corresponding to the operating data curve C; if there are peak differential points on the curve D whose change characterization values ​​belong to the preset judgment interval, then the set constructed by the horizontal coordinates of all the peak differential points on the curve D whose change characterization values ​​belong to the preset judgment interval is recorded as the set to be selected, and the smallest horizontal coordinate value in the set to be selected is selected as the initial lag value corresponding to the operating data curve C.

5. The method for collecting and controlling the operation data of a functional robot driven by a cloud platform as claimed in claim 4, characterized in that: The method for obtaining the change characterization value of the peak difference point includes: For any peak differential point c on the curve D: on the curve D, the set constructed by all differential points located on the left side of the peak differential point c is recorded as the first set; the cumulative sum of the tangent slopes of all differential points in the first set is recorded as the comprehensive slope; the difference between the tangent slope of the peak differential point c and the comprehensive slope is recorded as the first difference, and the hyperbolic tangent function value of the first difference is recorded as the first characterization value; the mean of the ordinate values ​​of all differential points in the first set is obtained, and recorded as the first mean; the ratio of the ordinate value of the peak differential point c to the absolute value of the first mean is recorded as the first ratio, and the hyperbolic tangent function value of the first ratio is recorded as the second characterization value; the mean of the first characterization value and the second characterization value is recorded as the change characterization value corresponding to the peak differential point c.

6. The method for collecting and controlling the operation data of a functional robot driven by a cloud platform as claimed in claim 1, characterized in that: The method for obtaining the correlation characterization value between any two operation data curves in the operation data curve set includes: For the i-th operating data curve in the data curve set and the j-th operating data curve in the data curve set, i is not equal to j: Obtain a hysteresis difference interval to be determined between the i-th operation data curve and the j-th operation data curve, and record all integers in the hysteresis difference interval to be determined as hysteresis differences to be determined; obtain the preference degree of each hysteresis difference to be determined in the hysteresis difference interval to be determined according to the hysteresis difference interval to be determined, the envelope of the i-th operation data curve, the envelope of the j-th operation data curve, the peak data point on the i-th operation data curve, and the peak data point on the j-th operation data curve; obtain a pair of characteristic data point sequences corresponding to each hysteresis difference to be determined in the hysteresis difference interval to be determined; In the interval of hysteresis difference values ​​to be determined, the hysteresis difference value to be determined corresponding to the maximum preferred degree is selected as the target hysteresis difference value; two feature data point sequences constituting the feature data point sequence pair corresponding to the target hysteresis difference value are respectively recorded as a first sequence and a second sequence, a sequence constructed by the ordinate values ​​of all data points in the first sequence is recorded as a first sequence to be analyzed, and a sequence constructed by the ordinate values ​​of all data points in the second sequence is recorded as a second sequence to be analyzed; Using the absolute value of the Spearman correlation coefficient between the first sequence to be analyzed and the second sequence to be analyzed as a correlation representation value between the i-th operating data curve and the j-th operating data curve; The method for obtaining the to-be-determined hysteresis difference interval includes: recording the value obtained by subtracting the initial hysteresis value corresponding to the j-th operation data curve from the initial hysteresis value corresponding to the ith operation data curve as the initial difference; recording the value obtained by subtracting a preset first constant from the initial difference as the first eigenvalue, recording the value obtained by adding the preset first constant to the initial difference as the second eigenvalue, and recording the value interval composed of the first eigenvalue and the second eigenvalue as the to-be-determined hysteresis difference interval between the ith operation data curve and the j-th operation data curve.

7. The method for collecting and controlling the operation data of a functional robot driven by a cloud platform as claimed in claim 6, characterized in that: The method for obtaining the characteristic data point sequence pair corresponding to the to-be-determined lag difference value comprises: For the hth hysteresis difference value to be determined in the hysteresis difference value interval to be determined: The hth lag difference to be determined is recorded as G; if G is a positive number, the remaining data points on the i-th operation data curve except the first G data points and the remaining data points on the j-th operation data curve except the last G data points are all recorded as characteristic data points; if G is a negative number, the remaining data points on the i-th operation data curve except the last G data points and the remaining data points on the j-th operation data curve except the first G data points are all recorded as characteristic data points; if G is 0, Then all the data points on the i-th operating data curve and all the data points on the j-th operating data curve are recorded as characteristic data points; the sequence constructed by all the characteristic data points on the i-th operating data curve is recorded as characteristic data point sequence h1, and the sequence constructed by all the characteristic data points on the j-th operating data curve is recorded as characteristic data point sequence h2; the sequence pair composed of the characteristic data point sequence h1 and the characteristic data point sequence h2 is recorded as the characteristic data point sequence pair corresponding to the h-th lag difference to be determined.

8. The method for collecting and controlling the operation data of a functional robot driven by a cloud platform as claimed in claim 7, characterized in that: The method for obtaining the preferred degree corresponding to each hysteresis difference value to be determined in the hysteresis difference value interval to be determined includes: For the hth hysteresis difference to be determined: Obtaining the upper envelope and lower envelope of the ith operation data curve, and the upper envelope and lower envelope of the jth operation data curve; obtaining envelope data point pairs corresponding to each data point on the ith operation data curve and envelope data point pairs corresponding to each data point on the jth operation data curve according to the upper envelope and lower envelope of the ith operation data curve and the upper envelope and lower envelope of the jth operation data curve; Obtaining a first set of data point pairs to be analyzed and a second set of data point pairs to be analyzed according to the hth hysteresis difference to be determined, the i-th operation data curve, and the j-th operation data curve; The sum of the total number of peak data points on the i-th operation data curve and the total number of peak data points on the j-th operation data curve is recorded as a first quantity characterization value, and the number of all data points in the second set of data point pairs to be analyzed is recorded as a second quantity characterization value; and the ratio of the second quantity characterization value to the first quantity characterization value is recorded as a peak data point overlap rate; For any first data point pair H to be analyzed in the first set of data point pairs to be analyzed, two data points constituting the first data point pair H to be analyzed are respectively recorded as a first data point and a second data point, an envelope data point pair corresponding to the first data point and an envelope data point pair corresponding to the second data point are respectively recorded as a first envelope data point pair and a second envelope data point pair, an absolute value of a vertical coordinate difference between the two data points constituting the first envelope data point pair is obtained and recorded as a first envelope difference, an absolute value of a vertical coordinate difference between the two data points constituting the second envelope data point pair is obtained and recorded as a second envelope difference, and an absolute value of a difference between the first envelope difference and the second envelope difference is recorded as a similarity characterization value of the data point pair H to be analyzed; The mean of the similarity characterization values ​​of all the pairs of data points to be analyzed in the first set of pairs of data points to be analyzed is recorded as a second mean; negative correlation mapping is performed on the second mean, and the value obtained after the negative correlation mapping is recorded as the envelope amplitude similarity rate; A weighted sum is taken for the peak data point coincidence rate and the envelope amplitude similarity rate, and the value obtained by the weighted sum is recorded as the preferred degree corresponding to the hth lag difference value to be determined.

9. The method for collecting and controlling the operation data of a functional robot based on a cloud platform as claimed in claim 8, characterized in that: The method for obtaining the envelope data point pair corresponding to the data point includes: For the g-th data point on the i-th operating data curve: in the upper envelope of the i-th operating data curve, obtain the data point with the same horizontal coordinate value as the g-th data point, and record it as the upper envelope data point of the g-th data point; in the lower envelope of the i-th operating data curve, obtain the data point with the same horizontal coordinate value as the g-th data point, and record it as the lower envelope data point of the g-th data point; record the data point pair formed by the upper envelope data point and the lower envelope data point of the g-th data point as the envelope data point pair corresponding to the g-th data point; The method for acquiring the envelope data point pair corresponding to the data point on the j-th operating data curve is the same as the method for acquiring the envelope data point pair corresponding to the g-th data point.

10. The method for collecting and controlling the operation data of a functional robot driven by a cloud platform as claimed in claim 8, characterized in that: The method for obtaining the first set of data point pairs to be analyzed and the second set of data point pairs to be analyzed includes: Acquire a characteristic mark value of each data point in the characteristic data point sequence h1 and a characteristic mark value of each data point in the characteristic data point sequence h2, wherein the characteristic mark value of the k-th data point in the characteristic data point sequence h1 and the characteristic data point sequence h2 is both k; In the characteristic data point sequence h1 and the characteristic data point sequence h2, data point pairs formed by two data points with the same characteristic tag value are all recorded as first data point pairs to be analyzed, and a set constructed by all the first data point pairs to be analyzed is recorded as a first data point pair set to be analyzed; For any first data point pair to be analyzed in the first set of data point pairs to be analyzed, if both data points in the first data point pair to be analyzed are peak data points, the first data point pair to be analyzed is recorded as a second data point pair to be analyzed; A set constructed by all second data point pairs to be analyzed in the first data point pair set to be analyzed is recorded as a second data point pair set to be analyzed.

Images