A robot-based cluster textile machine information acquisition method and system

Abstract

The present invention relates to the technical field of information collection, and specifically discloses a robot-based cluster textile machine information acquisition method and system, the method comprising receiving time-containing signals uploaded by each module, generating a signal matrix containing identification tags and time tags; identifying the signal matrix containing identification tags and time tags, and determining the abnormality of each textile machine; determining the information collection path of the robot based on the abnormality of each textile machine, and sending the information collection path to the robot; when the robot returns, reading the collected data of the robot; the present invention adds a robot between a data terminal and the textile machine, the robot moves to the vicinity of the textile machine to perform high-frequency data collection, and the data terminal performs low-frequency data collection on all textile machines, thereby adjusting the data collection process of the robot. In this architecture, the collection frequency of the data terminal can be very low, greatly reducing resource consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of information collection, and in particular to a robot-based cluster textile machine information acquisition method and system. Background Art

[0002] Intelligent textile machines are highly efficient textile machines that integrate modern technologies such as advanced sensing, automated control, artificial intelligence, and big data analytics with traditional textile equipment. These machines not only significantly improve the efficiency and quality of textile production but also enable automated monitoring and intelligent management.

[0003] The automated monitoring and intelligent management process of intelligent textile machines is inseparable from the information acquisition process. The existing information acquisition process is a remote acquisition process with the help of wireless networks. This method is very convenient, but when the number of textile machines is large, each textile machine needs to perform high-frequency data collection, which consumes a lot of resources. How to optimize the data collection architecture, provide a more flexible collection method, and reduce resource consumption is the technical problem that the technical solution of the present invention wants to solve. Summary of the Invention

[0004] The object of the present invention is to provide a robot-based cluster textile machine information acquisition method and system to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A robot-based cluster textile machine information acquisition method and system, the method comprising:

[0007] Query the modules with communication functions in the textile machine, and create a matrix template of the textile machine according to the position information of the modules with communication functions;

[0008] receiving time-containing signals uploaded by each module based on a preset first frequency, and counting the signals of all modules of the same textile machine at the same time based on a matrix template to generate a signal matrix containing an identification tag and a time tag; wherein the identification tag is a unique tag of the textile machine;

[0009] Identify the signal matrix containing identification tags and time tags to determine the abnormality of each textile machine;

[0010] Determining an information collection area for each textile machine in a map according to the abnormality of each textile machine; the area of ​​the information collection area is inversely proportional to the abnormality;

[0011] determining an information collection path for the robot based on the information collection area of ​​each textile machine, and sending the information collection path to the robot;

[0012] When the robot returns, the collected data of the robot is read; wherein, the robot obtains the signal of each module in each textile machine based on the second frequency on the information collection path and stores it locally; the second frequency is not less than the first frequency.

[0013] As a further solution of the present invention, the step of receiving the time-containing signals uploaded by each module based on the preset first frequency, counting the signals of all modules of the same textile machine at the same time based on the matrix template, and generating a signal matrix containing identification tags and time tags includes:

[0014] Receive the time-containing signal uploaded by each module based on a preset first frequency;

[0015] Based on the preset storage variables, the signal values ​​uploaded by each module are stored in real time, and the uploaded time-containing signals are converted into standard values ​​based on the signal values; the standard values ​​are dimensionless data, and the value range of the standard data corresponding to all modules is the same;

[0016] For each textile machine, the standard values ​​are aggregated based on time, and the time of the aggregated standard values ​​is regularized to obtain the time label of the aggregated standard values;

[0017] Copy the matrix template, and for each type of aggregated standard value, insert the standard value into the copied matrix template according to the corresponding relationship between the standard value and the module to obtain the signal matrix;

[0018] Read the identification of the textile machine, create an identification tag, and insert the identification tag and time tag into the signal matrix.

[0019] As a further solution of the present invention: the step of identifying the signal matrix containing the identification tag and the time tag and determining the abnormality of each textile machine includes:

[0020] Based on the identification tags, a signal matrix containing a time tag is counted and the signal matrix is ​​spliced ​​based on the time sequence to obtain a three-dimensional signal matrix; two dimensions of the three-dimensional signal matrix represent the position of the module in the textile machine, and the other dimension is the time dimension; the three-dimensional signal matrix only contains the identification tags;

[0021] intercepting a three-dimensional signal matrix according to a preset backtracking time, comparing the intercepted three-dimensional signal matrices of different identification labels, and calculating similarity as the similarity between the different identification labels;

[0022] For each identification tag, the mean of its similarity with all other identification tags is calculated, and the abnormality is determined according to the mean; the abnormality is inversely proportional to the mean.

[0023] As a further solution of the present invention: the step of determining the information collection area of ​​each textile machine in the map according to the abnormality degree of each textile machine includes:

[0024] querying a regional map of the textile work area, and marking the outlines of the textile machines in the regional map based on the outline positions of the textile machines;

[0025] Determining an expansion radius according to the abnormality; the expansion radius is inversely proportional to the abnormality;

[0026] The equidistant contour of the textile machine outline is determined based on the expansion radius as the information collection area.

[0027] As a further solution of the present invention, the step of determining the information collection path of the robot based on the information collection area of ​​each textile machine and sending the information collection path to the robot includes:

[0028] A preset number of points are randomly selected in the information collection area as the passing point set of the corresponding textile machine;

[0029] A waypoint is randomly selected from the set of waypoints of each textile machine;

[0030] receiving a start point and an end point input by a management party, and creating an information collection path based on the start point, the end point, and all waypoints; wherein a set of waypoints corresponds to one information collection path;

[0031] Select the optimal path among the information collection paths and send it to the robot.

[0032] As a further embodiment of the present invention, the method further comprises:

[0033] During the movement, the robot obtains the distance between the robot and each textile machine in real time, and determines the dynamic collection frequency according to the inverse proportion of the distance; the dynamic collection frequency has a preset range, and the minimum value of the range is not less than the first frequency; the information collection process of the robot is supplemented with a visual information collection process, and when the distance between the robot and a textile machine is less than the preset distance threshold, the visual information collection process is triggered.

[0034] The technical solution of the present invention also provides a robot-based cluster textile machine information acquisition system, the system comprising:

[0035] A template creation module is used to query the modules with communication functions in the textile machine and create a matrix template of the textile machine according to the position information of the modules with communication functions;

[0036] a signal matrix generation module, configured to receive time-containing signals uploaded by each module based on a preset first frequency, and to count the signals of all modules of the same textile machine at the same time based on a matrix template, thereby generating a signal matrix containing an identification tag and a time tag; wherein the identification tag is a unique tag for the textile machine;

[0037] The abnormality determination module is used to identify the signal matrix containing identification tags and time tags and determine the abnormality of each textile machine;

[0038] A collection area creation module is used to determine the information collection area of ​​each textile machine in the map according to the abnormality degree of each textile machine; the area of ​​the information collection area is inversely proportional to the abnormality degree;

[0039] a path generation and sending module, configured to determine the information collection path of the robot based on the information collection area of ​​each textile machine, and send the information collection path to the robot;

[0040] The data reading module is used to read the collected data of the robot when the robot returns; wherein, the robot obtains the signal of each module in each textile machine based on the second frequency on the information collection path and stores it locally; the second frequency is not less than the first frequency.

[0041] As a further solution of the present invention: the signal matrix generation module includes:

[0042] A signal receiving unit, configured to receive a time signal uploaded by each module based on a preset first frequency;

[0043] A signal conversion unit is used to store the maximum value of the signal uploaded by each module in real time based on a preset storage variable, and convert the uploaded signal containing time into a standard value based on the maximum value of the signal; the standard value is dimensionless data, and the value range of the standard data corresponding to all modules is the same;

[0044] A value aggregation unit is used to aggregate the standard values ​​of each textile machine based on time, and regularize the time of the aggregated standard values ​​to obtain a time label of the aggregated standard values;

[0045] The value insertion unit is used to copy the matrix template, and for each type of aggregated standard value, insert the standard value into the copied matrix template according to the corresponding relationship between the standard value and the module to obtain the signal matrix;

[0046] The tag insertion unit is used to read the identification of the textile machine, create an identification tag, and insert the identification tag and time tag into the signal matrix.

[0047] As a further solution of the present invention: the abnormality determination module includes:

[0048] a matrix splicing unit, configured to count a signal matrix containing a time tag based on the identification tag, and splice the signal matrix based on the time sequence to obtain a three-dimensional signal matrix; two dimensions of the three-dimensional signal matrix represent the position of the module in the textile machine, and the other dimension is the time dimension; and the three-dimensional signal matrix only contains the identification tag;

[0049] a matrix interception unit, configured to intercept a three-dimensional signal matrix according to a preset backtracking time, compare the intercepted three-dimensional signal matrices of different identification labels, and calculate similarities as similarities between the different identification labels;

[0050] The mean calculation application unit is used to calculate the mean of the similarity between each identification tag and all other identification tags, and determine the abnormality according to the mean; the abnormality is inversely proportional to the mean.

[0051] As a further solution of the present invention: the acquisition area creation module includes:

[0052] A contour insertion unit is used to query the regional map of the textile work area and mark the contour of the textile machine in the regional map based on the contour position of the textile machine;

[0053] a radius generating unit, configured to determine an expansion radius according to the abnormality; the expansion radius is inversely proportional to the abnormality;

[0054] The expansion unit is used to determine an equidistant contour of the textile machine contour based on an expansion radius as an information collection area.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] The present invention adds a robot between the data terminal and the textile machine. The robot moves to the vicinity of the textile machine to perform high-frequency data collection, and the data terminal performs low-frequency data collection on all textile machines, thereby adjusting the robot's data collection process. In this architecture, the collection frequency of the data terminal can be very low, and the high-frequency collection task is handed over to the robot closer to the textile machine. When the robot returns, the collected data of the robot is read again. This architecture actually shortens the data collection distance in disguise and greatly reduces resource consumption.

[0057] In fact, when the data collection distance is short enough, more diverse data collection methods can be used. In the most extreme cases, even wired collection methods can be used. For example, a robot can connect to the data port of a textile machine for wired data collection. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention.

[0059] Figure 1 The overall flow chart of the robot-based cluster textile machine information acquisition method is shown.

[0060] Figure 2 The structural diagram of the robot-based cluster textile machine information acquisition system is shown. DETAILED DESCRIPTION

[0061] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0062] Figure 1 The figure is a general flow chart of a clustered textile machine information acquisition method and system based on a robot. In an embodiment of the present invention, a clustered textile machine information acquisition method based on a robot includes:

[0063] Step S100: querying the modules with communication functions in the textile machine, and creating a matrix template of the textile machine according to the position information of the modules with communication functions;

[0064] The application field of this application is the working field of clustered textile machines. The working process of each textile machine is independent, and its discharge port is generally connected to the same conveying equipment; each textile machine will be installed with multiple intelligent devices, which can be some sensors or some signal transfer units, and they need to have communication functions; in actual applications, the textile machine itself will have a built-in communication module, and these sensors or signal transfer units will first send the signal containing their own identification to the communication module, and then the communication module will upload it to the main end.

[0065] Furthermore, the position of the module with communication function in the textile machine is fixed, and the modules with communication function are sorted in sequence according to the position, and then a matrix template is constructed; the models of all textile machines are the same, and the matrix template used is also the same; the modules with communication function are sorted in sequence according to the position can be from top to bottom, from north to south (or from west to east), and generally two directions are set, because the matrix template is a two-dimensional matrix, from top to bottom corresponds to top to bottom in the matrix, and from north to south (or from west to east) corresponds to left to right in the matrix. Based on this, a correspondence between modules and row and column positions is established, such as the first row and the first column corresponds to the first module from top to bottom and the first module from north to south (or from west to east); the first row and the first column corresponds to the second module from top to bottom and the first module from north to south (or from west to east), and so on.

[0066] In fact, there are some other ways, for example, numbering each module with communication function, and then corresponding to the row and column positions of the first row in turn (the number of columns needs to be set in advance). After each row and column position of the first row corresponds to a module, the second row is operated, and so on.

[0067] Step S200: receiving signals containing time information uploaded by each module based on a preset first frequency, counting signals of all modules of the same textile machine at the same time based on a matrix template, and generating a signal matrix containing identification tags and time tags; wherein the identification tag is a unique tag of the textile machine;

[0068] This method is applied to the main end, which receives the time-containing signals uploaded by each module based on a preset first frequency. Step S100 has created a matrix template, and each module has a corresponding row and column position. The signal is filled into the corresponding position to obtain a signal matrix. It should be noted that the signals of the modules of the same textile machine at the same time will be inserted into the same matrix, that is, the signal matrix contains an identification tag and a time tag. The identification tag indicates which textile machine, and the time tag indicates the time at which the signal is obtained. Since the signal acquisition frequency of all modules of all textile machines is the same, the process of determining the time tag is very simple. Assuming that the first frequency is to collect data every one minute and the first collection is at 7:00, then the time tags are 7:00, 7:01, 7:02, and so on. The collection time of all signals is almost close to the time tag, with a difference of at most a few tenths of a second (fluctuation in the data transmission process).

[0069] Step S300: Identify the signal matrix containing the identification tag and the time tag to determine the abnormality degree of each textile machine;

[0070] The signal matrix is ​​classified according to the identification label to obtain the signal matrix of each textile machine at different times, which represents the operating status of the textile machine. By identifying it, it can be determined whether there is an abnormality in each textile machine. The abnormality degree is used to represent this parameter. The higher the abnormality degree, the greater the possibility of an abnormality.

[0071] Step S400: determining an information collection area for each textile machine in the map according to the abnormality degree of each textile machine; the area of ​​the information collection area is inversely proportional to the abnormality degree;

[0072] Step S500: determining an information collection path of the robot based on the information collection area of ​​each textile machine, and sending the information collection path to the robot;

[0073] Step S600: When the robot returns, the collected data of the robot is read; wherein, the robot obtains the signal of each module in each textile machine based on the second frequency on the information collection path and stores it locally; the second frequency is not less than the first frequency.

[0074] The technical solution of the present invention actually provides a gradient information acquisition architecture, in which the main terminal obtains the working information of all textile machines (signals of each module) by means of a wireless network. The acquisition frequency of this process is very low, that is, the above-mentioned first frequency is very low; further, based on the acquired working information, the abnormality degree of each textile machine is determined, and the movement path of the robot is adjusted according to the abnormality degree. When the robot moves on the movement path, it will further obtain information about the textile machine, and the acquisition frequency is higher than the first frequency. Since the robot is mobile, it can be very close to the textile machine, and the data acquisition process is faster. After the robot completes data acquisition, it returns to the main terminal, and the main terminal reads the data obtained by the robot again; in this process, the data acquisition pressure of the main terminal is very small, and more data acquisition tasks are completed by the robot. Since the position of the robot is constantly changing, it can choose to only obtain data from textile machines that are close enough. That is, the robot calculates the distance to each textile machine in real time. When the distance is less than a preset distance threshold, the data of the textile machine is obtained based on a second frequency, which is greater than the first frequency, for example, once per second; of course, the distance threshold will be very large, much larger than the area of ​​each information collection area, which also means that the robot may also obtain data from multiple textile machines simultaneously during movement.

[0075] Specifically, regarding the movement process of the robot, the requirement it must meet is that it needs to pass through each textile machine, that is, each textile machine is its waypoint, and then the information collection path can be obtained based on the conventional path generation scheme; however, on this basis, the present application introduces the parameter of abnormality, and determines the information collection area of ​​each textile machine in the map according to the abnormality of each textile machine, and the area of ​​the information collection area is inversely proportional to the abnormality, which means that the higher the abnormality of the textile machine, the smaller the corresponding information collection area, and the information collection area has an impact on the information collection path. The waypoint is included in the information collection area. The smaller the information collection area, the closer the waypoint is to the textile machine. Combined with the robot working process in the above content, the robot calculates the distance to each textile machine in real time. When the distance is less than the preset distance threshold, data acquisition is performed. The closer the waypoint is to the textile machine, the longer the time it takes for the robot to acquire data from the textile machine. The actual meaning is that the higher the abnormality of the textile machine, the longer the robot will acquire data from it and the more data it acquires. It should be noted that the movement speed of the robot in this application is generally fixed.

[0076] For the above content, an example is given as follows: assuming that the robot's data acquisition distance threshold is 50 meters, it will acquire data from all textile machines within 50 meters. The radius of the information collection area of ​​a certain textile machine is 10 meters, and the radius of the information collection area of ​​a certain textile machine is 1 meter. In the extreme case, the robot leaves the first textile machine when it is 10 meters away from the first textile machine, and the time it takes to acquire data is the time it takes to move 80 meters. The robot leaves the second textile machine when it is 1 meter away from the second textile machine, and the time it takes to acquire data is the time it takes to move 98 meters. Of course, the above is just an example, and the actual situation is much more complicated, but it can still be concluded that the smaller the information collection area, the more likely it is that the robot will take longer to acquire data.

[0077] Regarding step S100, the steps of receiving the time-containing signals uploaded by each module based on the preset first frequency, counting the signals of all modules of the same textile machine at the same time based on the matrix template, and generating a signal matrix containing identification tags and time tags include:

[0078] Receive the time-containing signal uploaded by each module based on a preset first frequency;

[0079] Based on the preset storage variables, the signal values ​​uploaded by each module are stored in real time, and the uploaded time-containing signals are converted into standard values ​​based on the signal values; the standard values ​​are dimensionless data, and the value range of the standard data corresponding to all modules is the same;

[0080] For each textile machine, the standard values ​​are aggregated based on time, and the time of the aggregated standard values ​​is regularized to obtain the time label of the aggregated standard values;

[0081] Copy the matrix template, and for each type of aggregated standard value, insert the standard value into the copied matrix template according to the corresponding relationship between the standard value and the module to obtain the signal matrix;

[0082] Read the identification of the textile machine, create an identification tag, and insert the identification tag and time tag into the signal matrix.

[0083] The above content describes the process of generating the signal matrix. Based on the preset first frequency, the time-containing signals uploaded by each module are received. Since the data type or unit of the signal of each module is different, dimensionless processing is required. The dimensionless processing method is to calculate the difference between the maximum and minimum values. For each signal, the difference between it and the minimum value is calculated. The difference between each signal and the minimum value is divided by the difference between the maximum and minimum values ​​to obtain data in the range of 0 to 1. At this time, the signal has no unit.

[0084] On this basis, the matrix template is a standard template and data cannot be directly filled in it. It is necessary to copy a copy first and then insert the signal into the copy. In addition, the time of inserting the signal into the same copy needs to be the same or almost the same. Therefore, it is necessary to aggregate the standard values ​​based on time first. After inserting the information of all modules of the same textile machine at the same time into the same copy, create an identification tag and a time tag, and insert the identification tag and time tag into the signal matrix.

[0085] Regarding step S300, the step of identifying the signal matrix containing the identification tag and the time tag and determining the abnormality of each textile machine includes:

[0086] Based on the identification tags, a signal matrix containing a time tag is counted and the signal matrix is ​​spliced ​​based on the time sequence to obtain a three-dimensional signal matrix; two dimensions of the three-dimensional signal matrix represent the position of the module in the textile machine, and the other dimension is the time dimension; the three-dimensional signal matrix only contains the identification tags;

[0087] intercepting a three-dimensional signal matrix according to a preset backtracking time, comparing the intercepted three-dimensional signal matrices of different identification labels, and calculating similarity as the similarity between the different identification labels;

[0088] For each identification tag, the mean of its similarity with all other identification tags is calculated, and the abnormality is determined according to the mean; the abnormality is inversely proportional to the mean.

[0089] The abnormality calculation process is an independent analysis process for each textile machine. The signal matrix of the same textile machine can be extracted based on the identification label. The signal matrix contains a time label. For signal matrices with the same identification label, they are sorted based on the time label to obtain a three-dimensional signal matrix; the third dimension represents time.

[0090] The time span of the three-dimensional signal matrix is ​​very long, and relatively old data has no analytical value. Therefore, the administrator needs to set a lookback time in advance, such as one day or one week. Taking the current moment as the starting point, the three-dimensional signal matrix is ​​intercepted based on the lookback time. Then, by comparing the intercepted three-dimensional signal matrices, the differences in the operating status of each textile machine can be determined, which is represented by the similarity parameter.

[0091] For each textile machine, the mean of its similarity with other textile machines is calculated. The larger the mean, the more similar its working state is to other textile machines. At this time, it is normal, so the probability of large-scale abnormalities is extremely low. Even if it occurs, there will be other warning methods. The detection process of this application will be somewhat redundant. The focus of the detection process of this application is to query different textile machines among multiple textile machines and treat them as abnormal textile machines. The advantage of this detection method is that it almost eliminates the impact of the environment, because the impact of the environment on all textile machines is the same (such as power fluctuations). When all textile machines are affected in the same way, the impact on the abnormality judgment process is almost zero, because the abnormality is the degree of abnormality of a textile machine relative to other textile machines, not the degree of abnormality relative to a standard state.

[0092] Specifically, regarding the intercepted three-dimensional signal matrix, since each textile machine is identical, the matrix template is identical, the size of the information matrix at each moment is identical, and since the first frequency is identical, the number of matrices within the same lookback time is also identical. This also means that the size of the intercepted three-dimensional signal matrix for each textile machine is identical. Based on this, a feasible method for calculating similarity is as follows:

[0093] Where A and B represent the intercepted three-dimensional signal matrices corresponding to the two textile machines, S(A,B) represents the similarity, N, M and T represent the sizes of the intercepted three-dimensional signal matrix in three dimensions, ω ijt is the weight term at position (i, j, t), A ijt Represents the value at position (i, j, t) in A, B ijt Represents the value at position (i, j, t) in B; ω ijt The value of is as follows:

[0094] δ is the preset difference absolute value threshold, and α is the preset magnification factor.

[0095] Regarding step S400, the step of determining the information collection area of ​​each textile machine in the map according to the abnormality degree of each textile machine includes:

[0096] querying a regional map of the textile work area, and marking the outlines of the textile machines in the regional map based on the outline positions of the textile machines;

[0097] Determining an expansion radius according to the abnormality; the expansion radius is inversely proportional to the abnormality;

[0098] The equidistant contour of the textile machine outline is determined based on the expansion radius as the information collection area.

[0099] The above content provides a detailed description of the process of creating the information collection area. The shape of the information collection area created in this application is the same as the outline of the textile machine. An expansion radius is determined by the abnormality, and the equidistant outline of the textile machine outline is determined based on the expansion radius, which can be used as the information collection area. Its advantage is that the outline of the textile machine is retained, and the obtained information collection area is more consistent with the actual situation. In fact, a simpler way is to limit the information collection area to a circle, the center of the circle is the centroid of the textile machine, and the radius is determined by the abnormality, and the radius is inversely proportional to the abnormality.

[0100] Regarding step S500, the step of determining the information collection path of the robot based on the information collection area of ​​each textile machine and sending the information collection path to the robot includes:

[0101] A preset number of points are randomly selected in the information collection area as the passing point set of the corresponding textile machine;

[0102] A waypoint is randomly selected from the set of waypoints of each textile machine;

[0103] receiving a start point and an end point input by a management party, and creating an information collection path based on the start point, the end point, and all waypoints; wherein a set of waypoints corresponds to one information collection path;

[0104] Select the optimal path among the information collection paths and send it to the robot.

[0105] This application describes the process of generating an information collection path, which adopts an enumeration-selection scheme. First, a preset number of points are randomly selected in the information collection area as the set of passing points corresponding to the textile machine, such as 10 points. This process is equivalent to selecting 10 points in an area. At this time, each textile machine corresponds to 10 optional positions. Then, one and only one point is selected from the optional positions corresponding to each textile machine as the passing point. The starting point and end point input by the management party are received, and an information collection path is created based on the starting point, end point and all passing points. Since each textile machine corresponds to 10 optional positions, there will be a lot of information collection paths obtained. The optimal path is selected from the many information collection paths as the final path and sent to the robot.

[0106] Among them, the method of selecting the optimal path among many information collection paths is very simple, such as selecting the information collection path with the shortest total length.

[0107] As a preferred embodiment of the technical solution of the present invention, the method further includes:

[0108] During the movement, the robot obtains the distance between the robot and each textile machine in real time, and determines the dynamic collection frequency according to the inverse proportion of the distance; the dynamic collection frequency has a preset range, and the minimum value of the range is not less than the first frequency; the information collection process of the robot is supplemented with a visual information collection process, and when the distance between the robot and a textile machine is less than the preset distance threshold, the visual information collection process is triggered.

[0109] In an example of the technical solution of the present invention, the data acquisition process of the robot is described. In the above content, a specific data acquisition process has been provided, that is, a distance threshold is determined for the robot, and the robot performs high-frequency data collection on all textile machines within the distance threshold, and the data collection frequency is greater than the first frequency.

[0110] On this basis, the above content introduces the dynamic acquisition frequency, which obtains the distance between the robot and each textile machine in real time, and determines the dynamic acquisition frequency according to the inverse proportion of the distance. The greater the distance, the lower the dynamic acquisition frequency, which enables the robot to obtain more data from closer textile machines. For textile machines that are farther away, high-frequency acquisition will only be performed when the robot moves near the textile machine. In addition, due to the high abnormality of the textile machine, the robot will be closer to it, which also means that the robot will obtain more data from textile machines with high abnormality, thereby further gradient processing of the data acquisition process, so that data acquisition resources can be applied to more important textile machines, thereby improving resource utilization.

[0111] It should be noted that although it is a dynamic acquisition frequency, the dynamic acquisition frequency has a range. Generally speaking, between the first frequency and the second frequency, it is the second frequency only when they are very close, and the first frequency when they are far away.

[0112] Figure 2 The structure diagram of a robot-based cluster textile machine information acquisition system is shown. In a preferred embodiment of the technical solution of the present invention, a robot-based cluster textile machine information acquisition system is also provided. The system 10 includes:

[0113] The template creation module 11 is used to query the modules with communication functions in the textile machine and create a matrix template of the textile machine according to the position information of the modules with communication functions;

[0114] The signal matrix generation module 12 is configured to receive time-containing signals uploaded by each module based on a preset first frequency, and to count the signals of all modules of the same textile machine at the same time based on a matrix template to generate a signal matrix containing an identification tag and a time tag; wherein the identification tag is a unique tag for the textile machine;

[0115] The abnormality determination module 13 is used to identify the signal matrix containing the identification tag and the time tag and determine the abnormality of each textile machine;

[0116] A collection area creation module 14 is used to determine an information collection area for each textile machine in the map according to the abnormality degree of each textile machine; the area of ​​the information collection area is inversely proportional to the abnormality degree;

[0117] a path generation and sending module 15 for determining an information collection path of the robot based on the information collection area of ​​each textile machine and sending the information collection path to the robot;

[0118] The data reading module 16 is used to read the collected data of the robot when the robot returns; wherein, the robot obtains the signal of each module in each textile machine based on the second frequency on the information collection path and stores it locally; the second frequency is not less than the first frequency.

[0119] Furthermore, the signal matrix generating module 12 includes:

[0120] A signal receiving unit, configured to receive a time signal uploaded by each module based on a preset first frequency;

[0121] A signal conversion unit is used to store the maximum value of the signal uploaded by each module in real time based on a preset storage variable, and convert the uploaded signal containing time into a standard value based on the maximum value of the signal; the standard value is dimensionless data, and the value range of the standard data corresponding to all modules is the same;

[0122] A value aggregation unit is used to aggregate the standard values ​​of each textile machine based on time, and regularize the time of the aggregated standard values ​​to obtain a time label of the aggregated standard values;

[0123] The value insertion unit is used to copy the matrix template, and for each type of aggregated standard value, insert the standard value into the copied matrix template according to the corresponding relationship between the standard value and the module to obtain the signal matrix;

[0124] The tag insertion unit is used to read the identification of the textile machine, create an identification tag, and insert the identification tag and time tag into the signal matrix.

[0125] Specifically, the abnormality determination module 13 includes:

[0126] a matrix splicing unit, configured to count a signal matrix containing a time tag based on the identification tag, and splice the signal matrix based on the time sequence to obtain a three-dimensional signal matrix; two dimensions of the three-dimensional signal matrix represent the position of the module in the textile machine, and the other dimension is the time dimension; and the three-dimensional signal matrix only contains the identification tag;

[0127] a matrix interception unit, configured to intercept a three-dimensional signal matrix according to a preset backtracking time, compare the intercepted three-dimensional signal matrices of different identification labels, and calculate similarities as similarities between the different identification labels;

[0128] The mean calculation application unit is used to calculate the mean of the similarity between each identification tag and all other identification tags, and determine the abnormality according to the mean; the abnormality is inversely proportional to the mean.

[0129] Furthermore, the acquisition area creation module 14 includes:

[0130] A contour insertion unit is used to query the regional map of the textile work area and mark the contour of the textile machine in the regional map based on the contour position of the textile machine;

[0131] a radius generating unit, configured to determine an expansion radius according to the abnormality; the expansion radius is inversely proportional to the abnormality;

[0132] The expansion unit is used to determine an equidistant contour of the textile machine contour based on an expansion radius as an information collection area.

[0133] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A clustered textile machine information acquisition method based on robots, characterized in that: The method comprises: Query the modules with communication functions in the textile machine, and create a matrix template of the textile machine according to the position information of the modules with communication functions; receiving time-containing signals uploaded by each module based on a preset first frequency, and counting the signals of all modules of the same textile machine at the same time based on a matrix template to generate a signal matrix containing an identification tag and a time tag; wherein the identification tag is a unique tag of the textile machine; Identify the signal matrix containing identification tags and time tags to determine the abnormality of each textile machine; Determining an information collection area for each textile machine in a map according to the abnormality of each textile machine; the area of ​​the information collection area is inversely proportional to the abnormality; determining an information collection path for the robot based on the information collection area of ​​each textile machine, and sending the information collection path to the robot; When the robot returns, the collected data of the robot is read; wherein, the robot obtains the signal of each module in each textile machine based on the second frequency on the information collection path and stores it locally; the second frequency is not less than the first frequency; The step of identifying the signal matrix containing the identification tag and the time tag and determining the abnormality of each textile machine includes: Based on the identification tags, a signal matrix containing a time tag is counted and the signal matrix is ​​spliced ​​based on the time sequence to obtain a three-dimensional signal matrix; two dimensions of the three-dimensional signal matrix represent the position of the module in the textile machine, and the other dimension is the time dimension; the three-dimensional signal matrix only contains the identification tags; intercepting a three-dimensional signal matrix according to a preset backtracking time, comparing the intercepted three-dimensional signal matrices of different identification labels, and calculating similarity as the similarity between the different identification labels; For each identification tag, calculate the mean of its similarity with all other identification tags, and determine the abnormality according to the mean; the abnormality is inversely proportional to the mean; The step of determining the information collection area of ​​each textile machine in the map according to the abnormality degree of each textile machine includes: querying a regional map of the textile work area, and marking the outlines of the textile machines in the regional map based on the outline positions of the textile machines; Determining an expansion radius according to the abnormality; the expansion radius is inversely proportional to the abnormality; The equidistant contour of the textile machine outline is determined based on the expansion radius as the information collection area.

2. The robot-based cluster textile machine information acquisition method according to claim 1, characterized in that: The step of receiving the time-containing signals uploaded by each module based on a preset first frequency, counting the signals of all modules of the same textile machine at the same time based on a matrix template, and generating a signal matrix containing identification tags and time tags includes: Receive the time-containing signal uploaded by each module based on a preset first frequency; Based on the preset storage variables, the signal values ​​uploaded by each module are stored in real time, and the uploaded time-containing signals are converted into standard values ​​based on the signal values; the standard values ​​are dimensionless data, and the value range of the standard data corresponding to all modules is the same; For each textile machine, the standard values ​​are aggregated based on time, and the time of the aggregated standard values ​​is regularized to obtain the time label of the aggregated standard values; Copy the matrix template, and for each type of aggregated standard value, insert the standard value into the copied matrix template according to the corresponding relationship between the standard value and the module to obtain the signal matrix; Read the identification of the textile machine, create an identification tag, and insert the identification tag and time tag into the signal matrix.

3. The robot-based cluster textile machine information acquisition method according to claim 1, characterized in that: The steps of determining the information collection path of the robot based on the information collection area of ​​each textile machine and sending the information collection path to the robot include: A preset number of points are randomly selected in the information collection area as the passing point set of the corresponding textile machine; A waypoint is randomly selected from the set of waypoints of each textile machine; receiving a start point and an end point input by a management party, and creating an information collection path based on the start point, the end point, and all waypoints; wherein a set of waypoints corresponds to one information collection path; Select the optimal path among the information collection paths and send it to the robot.

4. The robot-based cluster textile machine information acquisition method according to claim 1, characterized in that: The method further comprises: During the movement, the robot obtains the distance between the robot and each textile machine in real time, and determines the dynamic collection frequency according to the inverse proportion of the distance; the dynamic collection frequency has a preset range, and the minimum value of the range is not less than the first frequency; the information collection process of the robot is supplemented with a visual information collection process, and when the distance between the robot and a textile machine is less than the preset distance threshold, the visual information collection process is triggered.

5. A robot-based cluster textile machine information acquisition system, characterized in that: The system comprises: A template creation module is used to query the modules with communication functions in the textile machine and create a matrix template of the textile machine according to the position information of the modules with communication functions; a signal matrix generation module, configured to receive time-containing signals uploaded by each module based on a preset first frequency, and to count the signals of all modules of the same textile machine at the same time based on a matrix template, thereby generating a signal matrix containing an identification tag and a time tag; wherein the identification tag is a unique tag for the textile machine; The abnormality determination module is used to identify the signal matrix containing identification tags and time tags and determine the abnormality of each textile machine; A collection area creation module is used to determine the information collection area of ​​each textile machine in the map according to the abnormality degree of each textile machine; the area of ​​the information collection area is inversely proportional to the abnormality degree; a path generation and sending module, configured to determine the information collection path of the robot based on the information collection area of ​​each textile machine, and send the information collection path to the robot; a data reading module, configured to read the collected data of the robot when the robot returns; wherein the robot acquires signals from modules in each textile machine on the information collection path based on a second frequency and stores the signals locally; the second frequency is not less than the first frequency; The abnormality determination module includes: a matrix splicing unit, configured to count a signal matrix containing a time tag based on the identification tag, and splice the signal matrix based on the time sequence to obtain a three-dimensional signal matrix; two dimensions of the three-dimensional signal matrix represent the position of the module in the textile machine, and the other dimension is the time dimension; and the three-dimensional signal matrix only contains the identification tag; a matrix interception unit, configured to intercept a three-dimensional signal matrix according to a preset backtracking time, compare the intercepted three-dimensional signal matrices of different identification labels, and calculate similarities as similarities between the different identification labels; a mean calculation application unit, configured to calculate, for each identification tag, a mean of similarities between it and all other identification tags, and determine an abnormality based on the mean; the abnormality is inversely proportional to the mean; The acquisition area creation module includes: A contour insertion unit is used to query the regional map of the textile work area and mark the contour of the textile machine in the regional map based on the contour position of the textile machine; a radius generating unit, configured to determine an expansion radius according to the abnormality; the expansion radius is inversely proportional to the abnormality; The expansion unit is used to determine an equidistant contour of the textile machine contour based on an expansion radius as an information collection area.

6. The robot-based cluster textile machine information acquisition system according to claim 5, characterized in that: The signal matrix generation module includes: A signal receiving unit, configured to receive a time signal uploaded by each module based on a preset first frequency; A signal conversion unit is used to store the maximum value of the signal uploaded by each module in real time based on a preset storage variable, and convert the uploaded signal containing time into a standard value based on the maximum value of the signal; the standard value is dimensionless data, and the value range of the standard data corresponding to all modules is the same; A value aggregation unit is used to aggregate the standard values ​​of each textile machine based on time, and regularize the time of the aggregated standard values ​​to obtain a time label of the aggregated standard values; The value insertion unit is used to copy the matrix template, and for each type of aggregated standard value, insert the standard value into the copied matrix template according to the corresponding relationship between the standard value and the module to obtain the signal matrix; The tag insertion unit is used to read the identification of the textile machine, create an identification tag, and insert the identification tag and time tag into the signal matrix.

Images