Information processor, information processing method, and information processing program

Abstract

To provide a technology capable of improving calculation accuracy as much as possible when calculating displacement of a prescribed object on the basis of an observation of GNSS with respect to the object.SOLUTION: An information processor includes a control unit which acquires positioning coordinates measuring an observation point installed on an object with a positioning satellite, performs error processing with respect to the positioning coordinates by using a Kalman filter, and calculates displacement of the object on the basis of error correction coordinates to be the positioning coordinates error processed with the Kalman filter. The control unit calculates error correction coordinates on the basis of the state of the positioning coordinates determined on the basis of prescribed preprocessing of the positioning coordinates when performing error processing by using the Kalman filter.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program for calculating the displacement of a given object based on an observation point on the object. [Background technology]

[0002] Measuring the displacement of structures is an important element in evaluating the soundness and load-bearing capacity of structures such as bridges.

[0003] In recent years, technology has been put into practical use to measure changes in the coordinates of an object's observation point using a satellite observation system (GNSS: Global Navigation Satellite System), which can receive radio waves from satellites and determine the position of antennas in three dimensions.

[0004] For example, Patent Document 1 discloses a system that supports slope stability assessment by processing a large amount of observation data using artificial intelligence (AI) based on GNSS observation results. In this system, observed coordinate change data observed during an observation period and the like are input into a change model of the target slope generated by machine learning, and predicted coordinate changes, which are future changes in the coordinates of the target slope, are calculated. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2019-85712 A Summary of the Invention [Problem to be solved by the invention]

[0006] Satellite observation systems (GNSS: Global Navigation Satellite System), which can receive radio waves from satellites and grasp the position of antennas in three dimensions, have become smaller, more accurate, and less expensive in recent years, and can be an effective means of measuring the displacement of structures such as bridges. However, with the current observation data processing method, system identification is difficult due to the influence of positioning accuracy, and diagnosis and prediction of the soundness, load-bearing capacity, etc. of structures are difficult. Here, according to the technology described in Patent Document 1, it seems that the calculation accuracy of the displacement of the target object can be improved by processing a large amount of observation data using artificial intelligence (AI). However, in the technology described in Patent Document 1, time series processing is performed on slopes that show slow displacement, and it may be difficult to apply it to time series processing with the real-time properties required for monitoring the soundness of structures such as bridges.

[0007] On the other hand, observation data contains noise caused by various factors, which can be a factor in reducing diagnostic and prediction performance for the soundness and load-bearing capacity of structures. Therefore, it is considered to use a Kalman filter to remove noise from observation data and estimate true values. However, with conventional error processing using a Kalman filter, it has been difficult to perform accurate error correction for GNSS observation data, which has a lot of error (variation) in the observation results.

[0008] An object of the present disclosure is to provide a technology that can improve the calculation accuracy as much as possible when calculating the displacement of a specific object based on the observation results of the object by a GNSS. [Means for solving the problem]

[0009] The information processing device disclosed herein is an information processing device that calculates the displacement of a given object based on an observation point on the object. The information processing device includes a control unit that executes the following: acquiring positioning coordinates obtained by positioning the observation point installed on the object using a positioning satellite, performing error processing on the positioning coordinates using a Kalman filter, and calculating the displacement of the object based on error-corrected coordinates that are the positioning coordinates error-processed by the Kalman filter. When performing error processing using the Kalman filter, the control unit calculates the error-corrected coordinates based on a state of the positioning coordinates determined based on a predetermined pre-processing of the positioning coordinates.

[0010] In the information processing device of the present disclosure, the control unit executes, as a first process included in the pre-processing, a process of determining whether the positioning coordinates are missing values, and if it is determined that the positioning coordinates are not missing values, a process of regarding the positioning coordinates as missing values ​​when the positioning coordinates are a float solution.Furthermore, in this case, if it is determined in the first process that the positioning coordinates are not missing values, the control unit executes, as a second process included in the pre-processing, a process of calculating an anomaly degree of the positioning coordinates using a sliding window k-nearest neighbor method when the positioning coordinates are a fix solution, and regarding the positioning coordinates whose anomaly degree is equal to or greater than a predetermined threshold as missing values.

[0011] In the information processing device, when performing error processing using the Kalman filter, the control unit may calculate a Kalman gain using a pre-defined solution based on a plurality of error patterns. In this case, the control unit may determine a current error pattern based on the current positioning coordinates and the previous positioning coordinates, and if the current error pattern differs from the previous error pattern, calculate the Kalman gain using the pre-defined solution based on an error pattern that matches the current error pattern among the plurality of error patterns.

[0012] The present disclosure can also be understood from the aspect of an information processing method by a computer. That is, the information processing method of the present disclosure is an information processing method for calculating a displacement of a predetermined object based on an observation point of the object, in which a computer executes the following steps: acquiring positioning coordinates obtained by positioning the observation point installed on the object by a positioning satellite, performing error processing on the positioning coordinates using a Kalman filter, and calculating the displacement of the object based on error-corrected coordinates that are the positioning coordinates error-processed by the Kalman filter, and when performing error processing using the Kalman filter, the computer calculates the error-corrected coordinates based on a state of the positioning coordinates determined based on a predetermined pre-processing of the positioning coordinates.

[0013] The present disclosure can also be understood from the aspect of an information processing program. That is, the information processing program of the present disclosure is an information processing program that calculates the displacement of a predetermined object based on an observation point of the object, and causes a computer to acquire positioning coordinates obtained by positioning the observation point installed on the object by a positioning satellite, perform error processing on the positioning coordinates using a Kalman filter, and calculate the displacement of the object based on error-corrected coordinates that are the positioning coordinates error-processed by the Kalman filter, and calculates the error-corrected coordinates based on a state of the positioning coordinates determined based on a predetermined pre-processing of the positioning coordinates when the computer performs error processing using the Kalman filter. Effect of the Invention

[0014] According to the present disclosure, when calculating the displacement of a specific object based on the observation results of the object by a GNSS, the calculation accuracy can be improved as much as possible. [Brief description of the drawings]

[0015] [Figure 1] FIG. 2 is a diagram illustrating components of a server included in an information processing system according to an embodiment. [Diagram 2]10 is a flowchart showing a process flow for calculating the displacement of a predetermined object based on the observation results of the object by a GNSS. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. The configurations of the following embodiments are merely examples, and the present disclosure is not limited to the configurations of the embodiments.

[0017] (Hardware configuration) The configuration of hardware resources of an information processing system according to the present disclosure that calculates the displacement of a specific object based on the observation results of the GNSS for the object will be described with reference to Fig. 1. Fig. 1 is a diagram showing components of a server 300 included in an information processing system 100 according to an embodiment. The process of calculating the displacement of an object based on the observation results of the GNSS according to the embodiment is executed by the server 300.

[0018] The server 300 may be any electronic device having a processing capability for arithmetic processing such as data acquisition, generation, and updating, and processing, such as a personal computer, a server, a mainframe, or other electronic device. That is, the server 300 may be configured as a computer having a processor such as a CPU or GPU, a main storage device such as a RAM or ROM, and an auxiliary storage device such as an EPROM, a hard disk drive, or a removable medium. The removable medium may be, for example, a USB memory or a disk recording medium such as a CD or DVD. The auxiliary storage device stores an operating system (OS), various programs, various tables, and the like.

[0019] The server 300 has, as functional units, a communication unit 301, a storage unit 302, and a control unit 303, and can realize functions that match the predetermined purpose of each functional unit by loading a program stored in an auxiliary storage device into a working area of ​​a main storage device and executing the program, and controlling each functional unit through the execution of the program. However, some or all of the functions may be realized by hardware circuits such as ASIC and FPGA.

[0020] Here, the communication unit 301 is a communication interface for connecting the server 300 to a network. The communication unit 301 includes, for example, a network interface board and a wireless communication circuit for wireless communication. The server 300 is connected to other external devices via the communication unit 301 so as to be able to communicate with them.

[0021] The storage unit 302 includes a main storage device and an auxiliary storage device. The main storage device is a memory in which the programs executed by the control unit 303 and data used by the control programs are expanded. The auxiliary storage device is a device in which the programs executed by the control unit 303 and data used by the control programs are stored. The storage unit 302 can store observation data (positioning coordinates of the present disclosure) of a satellite observation system (GNSS: Global Navigation Satellite System) described later.

[0022] The control unit 303 is a functional unit that controls the server 300. The control unit 303 can be realized by an arithmetic processing device such as a CPU. The control unit 303 further includes three functional units, an acquisition unit 3031, an error processing unit 3032, and a displacement calculation unit 3033. Each functional unit may be realized by the CPU executing a stored program. Note that the control unit 303 functions as a control unit according to the present disclosure by executing the processes of the acquisition unit 3031, the error processing unit 3032, and the displacement calculation unit 3033.

[0023] The server 300 may be operated from a predetermined user terminal 400 via the network 200. Here, the user terminal 400 may be any electronic device owned by a user who uses the information processing system 100, and may be, for example, a smartphone, a tablet terminal, a wearable terminal, a personal computer, or other terminal device.

[0024] The user terminal 400 has a communication unit 401, an input / output unit 402, and a storage unit 403 as functional units. The communication unit 401 is a communication interface for connecting the user terminal 400 to the network 200, and is configured to include, for example, a network interface board and a wireless communication circuit for wireless communication. The input / output unit 402 is a functional unit for displaying information transmitted from the outside via the communication unit 401, and for inputting information when transmitting the information to the outside via the communication unit 401. The storage unit 403 is configured to include a main storage device and an auxiliary storage device, similar to the storage unit 302 of the server 300.

[0025] The input / output unit 402 further includes a display unit 4021 and an operation input unit 4022. The display unit 4021 has a function of displaying various information, and is realized, for example, by an LCD (Liquid Crystal Display) display, an LED (Light Emitting Diode) display, an OLED (Organic Light Emitting Diode) display, etc. The operation input unit 4022 has a function of accepting operation input from a user, and is realized, specifically, by soft keys or hard keys of a touch panel or the like.

[0026] Moreover, the network 200 is, for example, an IP network. If the network 200 is an IP network, it may be wireless, wired, or a combination of wireless and wired, and for example, in the case of wireless communication, the user terminal 400 may access a wireless LAN access point (not shown) and communicate with the server 300 via a LAN or WAN. Furthermore, the network 200 is not limited to these examples, and may be, for example, a public switched telephone network, an optical fiber line, an ADSL line, a satellite communication network, or the like.

[0027] (Software configuration) Next, the software configuration of the information processing system according to the present disclosure that calculates the displacement of a specific object based on the GNSS observation results for the object will be described below.

[0028] 2 is a flowchart showing a process flow for calculating the displacement of a predetermined object based on the observation results of the object by the GNSS. This process is executed by the server 300. The object is, for example, a bridge.

[0029] In this flow, first, in S101, GNSS observation data is acquired. Here, the above observation data is positioning coordinates obtained by positioning an observation point installed on a bridge using a positioning satellite, and the acquisition unit 3031 of the server 300 can acquire the observation data by receiving radio waves from the positioning satellite using a well-known technology. Then, the acquisition unit 3031 stores the GNSS observation data in the storage unit 302 of the server 300 as an observation value yt at time t.

[0030] In this flow, next, the error processing unit 3032 of the server 300 performs error processing on the acquired observation value yt using a Kalman filter.

[0031] Here, GNSS observation data contains noise caused by various factors, and this noise adversely affects the measurement of bridge displacement, which is important for evaluating the soundness and load-bearing capacity of a bridge. Furthermore, with the conventional error processing of observation data using a Kalman filter, it was difficult to perform accurate error correction on GNSS observation data, which has a lot of error (variation).

[0032] Therefore, in the information processing system 100 of the present disclosure, when error processing is performed using a Kalman filter, an error-corrected coordinate is calculated based on the state of the observed value yt determined based on a predetermined pre-processing of the observed value yt. This error-corrected coordinate is the observed value yt that has been error-processed by the Kalman filter, and the displacement of the bridge is calculated based on the error-corrected coordinate.

[0033] Specifically, as a first process included in the above pre-processing, in S102, it is determined whether or not the observed value yt acquired in S101 is a missing value. If the observed value yt includes a missing value, the error processing unit 3032 of the server 300 can determine that the observed data at the time t is a missing value. If the determination in S102 is affirmative, the flow proceeds to S107, and if the determination in S102 is negative, the flow proceeds to S103.

[0034] If the result of the determination in S102 is negative, then in S103, it is determined whether the observation value yt acquired in S101 is a float solution. Here, the GNSS observation data includes a mixture of float solutions and fixed solutions. In RTK (Real Time Kinematic) surveying using GNSS, an integer bias is estimated to calculate the actual distance between the satellite and the receiving station, and a positioning solution calculated before the estimation of the integer bias is called a float solution, and a positioning solution calculated after the estimation of the integer bias is called a fixed solution. In RTK surveying, the positioning starts, and then the float solution is followed by the fix solution, but at the float solution stage, the positioning is not yet accurate, and the accuracy of the positioning solution is on the order of several centimeters to several meters. When performing interferometric positioning calculation in real time, the error processing unit 3032 of the server 300 first calculates the bias as a real value to calculate a float solution, and searches for the integer bias. Then, when the integer bias is estimated, a fix solution is calculated by positioning using the estimated integer bias. Then, the error processor 3032 can determine that the positioning solution calculated before estimating the integer bias is a float solution. Then, if the determination in the process of S103 is affirmative, the flow proceeds to the process of S106, and if the determination in the process of S103 is negative, the flow proceeds to the process of S104.

[0035] If a negative determination is made in S103, that is, if the observed value yt acquired in S101 is a fix solution, then in S104, the degree of anomaly of the observed value yt is calculated using the sliding window k nearest neighbor method.

[0036] Here, for the observation value yt acquired in S101, missing values ​​are removed by the process of S102, and furthermore, float solutions are removed by the process of S103. However, even if the observation value yt is a fix solution after the missing values ​​and float solutions are removed in this way, there is a possibility that the value is a misfix solution. Note that a misfix solution is GNSS observation data that may be included in a fix solution, even though it is a positioning solution with low accuracy like a float solution.

[0037] Therefore, in this embodiment, the sliding window k-nearest neighbor method is used to treat such misfix solutions as missing values ​​and process them accordingly.

[0038] The k-nearest neighbor method is a classification algorithm used in the field of machine learning, which classifies new input data by referring to data points around the input data. Specifically, for a given new data point, k nearest neighbors are found within a specific range. Using these nearest neighbors, the degree of anomaly in the input data is calculated and classified by comparing it with a threshold. The sliding window k-nearest neighbor method, which is an advanced version of the k-nearest neighbor method, slides the new data point over a fixed window size to accommodate time series data, and finds the k nearest neighbors around it.

[0039] In this embodiment, as a second process included in the above pre-processing, in S104, the error processor 3032 of the server 300 calculates the degree of anomaly of the observation value yt using the sliding window k-nearest neighbor method. At this time, the error processor 3032 sets the window size to data of a period that can be determined based on the satellite period and the displacement behavior due to temperature changes of bridge members. In addition, the threshold value and the neighborhood constant k are recursively set so that the rejection rate of the GNSS observation data by the sliding window k-nearest neighbor method is equivalent to the occurrence probability of a misfix solution (for example, so that the rejection rate is 10 to 20%). Then, the degree of anomaly is calculated based on the distance of the data group including the k nearest neighbor points.

[0040] In the second process, next, in S105, it is determined whether the degree of abnormality calculated in S104 is equal to or greater than a threshold value. If the result of the process in S105 is affirmative, the flow proceeds to S106, and if the result of the process in S105 is negative, the flow proceeds to S107.

[0041] If a positive judgment is made in S105, that is, if the degree of anomaly calculated in S104 is equal to or greater than the threshold value, or if a positive judgment is made in S103, that is, if the observed value yt acquired in S101 is a float solution, then in S106, a process is executed to regard the observed value yt as a missing value.

[0042] In this way, the error processing unit 3032 of the server 300 executes, as a first process included in the above pre-processing, a process of regarding the observed value yt acquired in S101 as a missing value when the observed value yt is a float solution, and executes, as a second process included in the above pre-processing, a process of regarding the observed value yt acquired in S101 as a missing value when the observed value yt is a misfix solution.

[0043] Then, after performing the process of regarding the observed value yt as a missing value as described above, or if a positive judgment is made in S102, i.e., if the observed value yt acquired in S101 is a missing value, or if a negative judgment is made in S105, i.e., if the observed value yt acquired in S101 is a fix solution containing no errors, then in S107, a process of estimating the parameters of the Kalman filter is performed.

[0044] In the process of S107, the error processor 3032 of the server 300 first calculates a one-period-ahead predictive distribution based on the state one time before. Here, if the mean in the state distribution one time before is mt-1 and the variance is Ct-1, the one-period-ahead predictive distribution is expressed by the following formulas 1 and 2.

number

number

[0045] Next, the error processor 3032 of the server 300 calculates the one-period-ahead prediction likelihood based on the above prediction distribution. Here, the one-period-ahead prediction likelihood is expressed by the following Equations 3 and 4.

number

number

[0046] According to a conventional Kalman filter, the above-mentioned observation error is assumed to be an accidental error that can be expressed by a normal distribution, and the following Kalman gain is calculated. In contrast, in this embodiment, the above-mentioned observation error is assumed to include multiple error patterns due to systematic errors caused by satellite arrangement, observation environment, etc., in addition to the accidental error, and the Kalman gain expressed by the following formula 5 is calculated.

number

[0047] In detail, the error processing unit 3032 of the server 300 according to this embodiment judges the current error pattern based on the current observation value yt and the previous observation value yt-1. Note that the storage unit 302 of the server 300 stores a plurality of error patterns and their pre-solutions that are determined in advance by pre-measurement, and the error processing unit 3032 can judge the current error pattern by, for example, comparing the fluctuation of the observation value, which is the difference between the current observation value yt and the previous observation value yt-1, with the pre-stored error pattern. The error pattern judged in real time at each time is stored in the storage unit 302 of the server 300.

[0048] If the current error pattern is the same as the previous error pattern, error processor 3032 uses the above-mentioned observation error as is to calculate the Kalman gain.

[0049] On the other hand, if the current error pattern is different from the previous error pattern, the error processor 3032 calculates the Kalman gain using a prior solution determined based on an error pattern that matches the current error pattern among a plurality of error patterns.

[0050] After estimating the parameters of the Kalman filter as described above, the error processing unit 3032 of the server 300 then determines in S108 whether the observed value yt is a missing value. If the result of the process in S108 is negative, the flow proceeds to S109, and if the result of the process in S108 is positive, the flow proceeds to S110.

[0051] If a negative determination is made in S108, then in S109, the error processor 3032 corrects the current state. Here, if the mean of the current state distribution is mt and the variance is Ct, the corrected current state distribution is expressed by the following Equations 6 and 7.

number

number

[0052] On the other hand, if the result of the S108 determination is affirmative, then in S110, the error processor 3032 uses the state one time before as the current state. If the mean of the state distribution at this time is mt and the variance is Ct, these are expressed by the following Equations 8 and 9.

number

number

[0053] The displacement calculation unit 3033 of the server 300 then calculates the displacement of the bridge based on the error-corrected coordinates in which the errors have been corrected by the Kalman filter as described above. This makes it possible to perform accurate error correction even for GNSS observation data with many errors (variations) in the observation results, and then monitor the soundness of the bridge in real time, thereby making it possible to suitably diagnose and predict the soundness, load-bearing capacity, etc. of the bridge.

[0054] As described above, according to this embodiment, when the displacement of a predetermined object is calculated based on the observation results of the object by the GNSS, the calculation accuracy can be improved as much as possible.

[0055] <Other Modifications> The above embodiment is merely an example, and the present disclosure may be modified as appropriate without departing from the spirit and scope of the present disclosure. For example, the processes and means described in the present disclosure may be freely combined and implemented as long as no technical contradiction occurs.

[0056] In addition, a process described as being performed by one device may be shared and executed by multiple devices, and a process described as being performed by different devices may be executed by one device. In an integrated circuit, the hardware configuration for realizing each function can be flexibly changed.

[0057] The present disclosure can also be realized by supplying a computer program implementing the functions described in the above embodiments to a computer, and having one or more processors of the computer read and execute the program. Such a computer program may be provided to the computer by a non-transitory computer-readable storage medium connectable to the system bus of the computer. Non-transitory computer-readable storage media include, for example, any type of disk, such as a magnetic disk (floppy disk, hard disk drive (HDD), etc.), an optical disk (CD-ROM, DVD disk, Blu-ray disk, etc.), a read-only memory (ROM), a random access memory (RAM), an EPROM, an EEPROM, a magnetic card, a flash memory, an optical card, and any type of medium suitable for storing electronic instructions. [Explanation of symbols]

[0058] 100 Information Processing Systems 300...Server 302...Storage section 303 Control section 3031··Acquisition Department 3032 Error processing section 3033 Displacement calculation section

Claims

1. 1. An information processing device that calculates a displacement of a predetermined object based on an observation point of the object, comprising: acquiring positioning coordinates of the observation point installed on the target object measured by a positioning satellite; performing error processing on the measured position coordinates using a Kalman filter; Calculating a displacement of the object based on error-corrected coordinates, which are the positioning coordinates that have been error-processed by the Kalman filter; A control unit that executes The control unit is When performing error processing using the Kalman filter, the error correction coordinates are calculated based on a state of the positioning coordinates determined based on a predetermined pre-processing of the positioning coordinates. Information processing device.

2. The control unit is As a first process included in the pre-processing, a process is executed to determine whether the measured coordinates are missing values, and when it is determined that the measured coordinates are not missing values, a process is executed to regard the measured coordinates as missing values ​​when the measured coordinates are a float solution. The information processing device according to claim 1 .

3. The control unit is When it is determined in the first process that the measured coordinates are not missing values, a second process included in the pre-processing calculates an anomaly degree of the measured coordinates using a sliding window k-nearest neighbor method when the measured coordinates are a fix solution, and regards the measured coordinates whose anomaly degree is equal to or greater than a predetermined threshold as missing values. The information processing device according to claim 2 .

4. The control unit is When performing error processing using the Kalman filter, a Kalman gain is calculated using a pre-defined solution based on a plurality of error patterns. The information processing device according to claim 2 .

5. The control unit is determining a current error pattern based on the current positioning coordinates and the previous positioning coordinates, and if the current error pattern is different from the previous error pattern, calculating the Kalman gain using the prior solution determined based on an error pattern that matches the current error pattern among the plurality of error patterns; The information processing device according to claim 4.

6. 1. An information processing method for calculating a displacement of a predetermined object based on an observation point of the object, comprising: The computer acquiring positioning coordinates of the observation point installed on the target object measured by a positioning satellite; performing error processing on the positioning coordinates using a Kalman filter; Calculating a displacement of the object based on error-corrected coordinates, which are the positioning coordinates that have been error-processed by the Kalman filter; Run The computer includes: When performing error processing using the Kalman filter, the error correction coordinates are calculated based on a state of the positioning coordinates determined based on a predetermined pre-processing of the positioning coordinates. Information processing methods.

7. An information processing program for calculating a displacement of a predetermined object based on an observation point of the object, comprising: On the computer, acquiring positioning coordinates of the observation point installed on the target object measured by a positioning satellite; performing error processing on the positioning coordinates using a Kalman filter; Calculating a displacement of the object based on error-corrected coordinates, which are the positioning coordinates that have been error-processed by the Kalman filter; Run the command, The computer includes: When the error processing is performed using the Kalman filter, the error correction coordinates are calculated based on a state of the positioning coordinates determined based on a predetermined pre-processing of the positioning coordinates. Information processing program.

Images