Error prediction device and error prediction method

Abstract

To provide a technique capable of allowing a worker to know in advance possible errors in measurements made by a surveying instrument.SOLUTION: An error prediction device 1 having a processor and a memory includes: an environmental data reception unit 11 that receives environmental data 91 of the survey site where a surveying instrument is installed; an error prediction unit 12 that predicts a prediction error in the survey result of the surveying instrument in the environment of the survey site by inputting the environmental data of the survey site to an error prediction model M; and a display data creation part 13 that creates display data for displaying a prediction error when the prediction error exceeds the allowable value. The error prediction model M is a trained model that is generated by machine learning using environmental data 91 showing the environment at the time of surveying and a set of error data in survey results as teacher data with respect to surveying instruments of the same model as the surveying instrument.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to an error prediction device and an error prediction method, and more particularly to an error prediction device and an error prediction method for predicting errors that depend on the environment in measurements made by a surveying instrument. [Background technology]

[0002] BACKGROUND ART It has been known that environmental factors such as temperature, humidity, and atmospheric pressure affect errors in measurement values ​​of surveying instruments such as total stations, laser scanners, electronic levels, and theodolites.

[0003] For example, Patent Document 1 discloses a surveying instrument that calculates an error according to the environmental temperature and corrects a distance measurement signal. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 60-86409 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the past, workers only became aware of errors after the measurement. Therefore, if they realized that the error exceeded the tolerance after the measurement, they had to analyze the cause, adjust the surveying instrument, and measure again, and there was a demand to know in advance whether the error exceeded the tolerance. In particular, in the case of scanners, it took a lot of time to perform one measurement. It takes Therefore, it was necessary to know the conditions beforehand before taking measurements, and when an error occurred, it could take time to identify the cause, so this was a particularly strong requirement. Also, if the error was within the allowable range, the surveying instrument would automatically correct the error, so the operator would not be able to recognize that it was the environmental conditions that were causing the error.

[0006] The present invention has been made in consideration of the above circumstances, and aims to provide a technique that enables an operator to recognize in advance errors that may occur in measurements using a surveying instrument. [Means for solving the problem]

[0007] In order to achieve the above object, one embodiment of the present invention provides an error prediction device that includes a processor and a memory, and is equipped with an environmental data receiving unit that receives environmental data of a surveying site where a surveying instrument is installed, an error prediction unit that inputs the environmental data of the surveying site into an error prediction model to predict a predicted error that will occur in the surveying results of the surveying instrument in the environment of the surveying site, and a display creation unit that creates display data for displaying the predicted error when the predicted error exceeds an allowable value, and is characterized in that the error prediction model is a trained model generated by machine learning using a set of environmental data that indicates the environment at the time of surveying and error data in the surveying results as training data for a surveying instrument of the same model as the surveying instrument.

[0008] Another aspect of the present invention provides an error prediction method, in which an arithmetic and control unit equipped with a processor and a memory acquires environmental data of a surveying site where a surveying instrument is installed, inputs the environmental data of the surveying site into an error prediction model to predict a predicted error that will occur in the surveying results of the surveying instrument in the environment of the surveying site, and generates a display to display the predicted error if the predicted error exceeds an allowable value, and the error prediction model is a trained model generated by machine learning using a set of environmental data indicating the environment at the time of surveying and errors in the surveying results as training data for a surveying instrument of the same model as the surveying instrument. [Effects of the Invention]

[0009] According to the error prediction device and the error prediction method of the above aspects, it becomes possible for an operator to recognize in advance errors that may occur in measurements using a surveying instrument. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is an external perspective view of a surveying instrument that is an error prediction device according to a first embodiment of the present invention. [Figure 2] FIG. 2 is a block diagram of the error prediction device. [Figure 3] 10 is a table showing examples of environmental sensors that configure the error prediction device. [Figure 4] FIG. 2 is a diagram illustrating a management system for a surveying instrument, which is the error prediction device. [Figure 5] FIG. 2 is a diagram showing an example of training data for generating an error prediction model used in the error prediction device. [Figure 6] FIG. 2 is a block diagram illustrating the configuration of an error prediction model generating device that generates the error prediction model. [Figure 7] FIG. 2 is a diagram illustrating an outline of error prediction by the error prediction device. [Figure 8] 10 is a flowchart of an error prediction process performed by the error prediction device. [Figure 9] 10 is a diagram showing an example of display data showing the prediction error of the error prediction device. FIG. [Figure 10] FIG. 10 is a configuration block diagram of a surveying instrument that is an error prediction device according to a second embodiment. [Figure 11] 10 is a flowchart of an error prediction process performed by the error prediction device. [Figure 12] FIG. 10 is a diagram showing an example of training data for generating error predictions used in the error prediction device. [Figure 13] FIG. 2 is a diagram showing an example of contribution data of environmental data in the error prediction device. [Figure 14] FIG. 10 is a diagram showing an example of display data showing the prediction error of the error prediction device. [Figure 15] FIG. 10 is a configuration block diagram of an information processing device that is an error prediction device according to a third embodiment. [Figure 16] 10 is a flowchart showing the process of error prediction by the survey error prediction method. [Figure 17] FIG. 10 is a configuration block diagram of a surveying instrument that is an error prediction device according to a fourth embodiment. [Figure 18] 10 is a flowchart of the error prediction process of the error prediction device. [Figure 19] FIG. 2 is a diagram showing an example of environmental forecast data used in the error prediction device. [Figure 20] FIG. 2 is a diagram showing an example of display data in the error prediction device. [Figure 21] FIG. 10 is a schematic diagram illustrating the appearance of an eyewear display device that is an error prediction device according to a fifth embodiment. [Figure 22] FIG. 2 is a block diagram of the error prediction device. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings, but the present invention is not limited thereto. In addition, the same components common to the embodiments and modifications will be designated by the same reference numerals, and duplicated descriptions will be omitted as appropriate.

[0012] I. First embodiment 1. Error prediction device Fig. 1 is a schematic diagram showing the appearance of an error prediction device (hereinafter simply referred to as "prediction device") according to a first embodiment of the present invention. Fig. 2 is a configuration block diagram of the prediction device. The prediction device is a surveying instrument 1, and in the illustrated example, it is a motor-driven total station installed at the surveying site via a tripod.

[0013] The surveying instrument 1 includes a control and calculation unit 10, a surveying unit 21, an environmental sensor 22, a display unit 23, an operation unit 24, a communication unit 25, and a storage unit 26. In addition, the surveying instrument 1 may include a tracking unit that automatically tracks a target, an image processing unit that processes captured images and videos, a GNSS receiving device, etc.

[0014] The surveying unit 21 includes a distance measuring unit 21a, an angle measuring unit 21b, and a rotation driving unit 21c. The distance measuring unit 21a includes a light emitting element, a distance measuring optical system, and a light receiving element (not shown) arranged inside the telescope 1c. The distance measuring unit 21a measures the distance to the target by emitting distance measuring light from the light emitting element through the distance measuring optical system to illuminate the target, and receiving the light reflected from the target with the light receiving element.

[0015] The angle measurement unit 21b measures the vertical rotation angle of the telescope 1c and the horizontal rotation angle of the support unit 1b relative to the base unit 1a using rotary encoders (not shown) arranged on each rotation axis, thereby measuring the horizontal angle and vertical angle to the target.

[0016] The rotation drive unit 21c is a motor, and is controlled by the control and calculation unit 10 to drive the rotation shafts of the telescope 1c and the base unit 1b.

[0017] The environmental sensor 22 is a sensor that detects the environment of the measurement site and acquires environmental data 91. The environment of the survey site may include the environment around the survey instrument 1 and the internal environment of the survey instrument 1. FIG. 3 shows an example of the environmental data 91 and the environmental sensor 22.

[0018] The ambient temperature sensor is a temperature sensor that measures the outside air temperature around the surveying instrument 1. The ambient humidity sensor is a humidity sensor that measures the humidity around the surveying instrument 1. The ambient air pressure sensor is a sensor that measures the air pressure around the surveying instrument 1. The internal temperature sensor is a temperature sensor that measures the air temperature inside the surveying instrument 1. The internal humidity sensor is a humidity sensor that measures the humidity inside the surveying instrument 1. The internal air pressure sensor is a sensor that measures the air pressure inside the surveying instrument 1.

[0019] The tilt sensor is a so-called optical tilt sensor that includes, for example, an electric bubble, a light source, a light receiving element, and control means, and is configured so that light from the light source passes through the electric bubble and is received by the light receiving element, and the control and calculation unit 10 calculates the tilt angle based on the received light signal, and is provided for at least two axes, the X axis of the surveying instrument 1 and the Y axis perpendicular to the X axis. The tilt sensor obtains tilt data, which is the levelness (tilt degree) of the surveying instrument 1, from its detected value, and obtains tilt stabilization time data, which indicates the stable state of the ground where the surveying instrument 1 is installed, from the stabilization time.

[0020] The visibility meter is, for example, a backscattering type visibility meter that includes a light-projecting unit equipped with an LED (Light Emitting Diode) that converges and emits red visible light (wavelength 620 nm), a light-receiving unit equipped with a light-receiving element such as a photodiode, and a narrow-band filter. The backscattering type visibility meter projects the emitted light through a narrow-band filter, detects backscattered light that is scattered by minute scatterers contained in the air through the narrow-band filter, and calculates visibility from the received light signal using a control and calculation unit 10 to obtain visibility data. Based on the visibility data, it is possible to consider the presence and extent of haze, fog, etc.

[0021] In this way, the environmental data 91 may be acquired by one environmental sensor 22, such as by acquiring ambient temperature data by an ambient temperature sensor. Alternatively, the environmental data 91 may be calculated from data acquired by two or more environmental sensors 22, such as the difference between indoor and outdoor humidity. Alternatively, two or more pieces of environmental data 91 may be acquired by one environmental sensor 22, such as a tilt sensor.

[0022] The environmental data and environmental sensors do not need to include all of the sensors and environmental data exemplified above, and can be used alone or in combination as appropriate depending on the environment to be considered. The environmental data is preferably two or more types selected from ambient temperature data, ambient humidity data, ambient air pressure data, tilt data, tilt stabilization time data, internal temperature data, internal humidity data, internal / external humidity difference data, wind speed data (not shown), light intensity data (not shown), etc. The number and types of sensors included in the environmental sensor 22 can be set as appropriate.

[0023] The display unit 23 is, for example, a liquid crystal display. The operation unit 24 has a power key, number keys, a decimal point key, plus / minus keys, an execution key, a scroll key, etc., and is configured to enable the operator to operate the surveying instrument 1 and input information to the surveying instrument 1.

[0024] The communication unit 25 is a communication control device such as a network adapter, a network interface card, or a LAN card, and connects the surveying instrument 1 to a communication network such as the Internet or a mobile phone communication network by wire or wirelessly. The control and calculation unit 10 is capable of inputting and outputting information to and from external devices via the communication unit 25 and the communication network.

[0025] The storage unit 26 is configured with a computer-readable storage medium such as a memory card or HDD (Hard Disk Drive). The storage unit 26 stores various programs for executing the functions of the surveying instrument 1. The storage unit 26 also stores measurement data of the surveying instrument 1, detection data of the environmental sensor 22, and various other information acquired by the surveying instrument 1. The storage unit 26 also stores an error prediction model M1 corresponding to the model of the surveying instrument 1. The error prediction model M1 will be described later.

[0026] The control and calculation unit 10 is a control and calculation unit that includes, for example, one or more processors such as a CPU (Central Processing Unit) and memories such as a RAM (Random Access Memory) and a ROM (Read Only Memory). The control and calculation unit 10 executes various processes by the processor reading necessary information (programs and data) from the storage unit 26 into the memory and executing it. The control and calculation unit 10 is connected to each hardware component that constitutes the surveying instrument 1.

[0027] The control and calculation unit 10 controls the surveying unit 21 to calculate the three-dimensional position coordinates of the object to be measured based on the measured values ​​obtained by measuring the distance and angle of the object to be measured. At this time, the control and calculation unit 10 corrects the measured values ​​based on environmental data 91 acquired simultaneously with the measurement or at a predetermined timing. The control and calculation unit 10 also acquires corresponding environmental data 91 based on the detection results of the environmental sensor 22. The control and calculation unit 10 also includes, as functional units, an environmental data accepting unit 11, an error predicting unit 12, and a display creating unit 13. The environmental data accepting unit 11 accepts the environmental data 91 detected by the environmental sensor 22.

[0028] The error prediction unit 12 inputs the environmental data 91 into the error prediction model M1, and predicts the predicted error that occurs in the measurement values ​​of the surveying instrument 1 under the environment of the surveying site. The display creation unit 13 creates display data 94 for displaying the prediction error.

[0029] The surveying instrument 1 is not limited to a motor-driven total station, and the surveying instrument may further include a rotating mirror in the vertical rotation drive unit of the surveying unit 21 that scans the ranging light in a 360° vertical direction, such as a 3D laser scanner that acquires 3D point cloud data, or an electronic level that includes an image sensor as the surveying unit 21 and measures height from the image pattern of a staff.

[0030] 2. Error prediction model Here, the error prediction model will be described. The error prediction model is a trained model generated by machine learning using a set of environmental data indicating the environment at the time of measurement and error data in the surveying results as training data for a surveying instrument of the same model as the surveying instrument 1. The error prediction model M1 is generated by an error prediction model generation device 3.

[0031] 2-1 Training data FIG. 4 shows a management system for the surveying instrument 1 for collecting training data. The surveying instrument 1 is communicably connected to a management server MS managed by the manager of the surveying instrument 1, such as the manufacturer, distributor (dealer), or management company of the surveying instrument 1. At predetermined intervals, in response to predetermined operations such as measurement or power-on, at the instruction of an operator, or at the request of the management server MS, the surveying instrument 1 can transmit various data related to the surveying instrument 1, such as measurement data acquired by the surveying instrument 1, error data obtained by analyzing the measurement values, environmental data during measurement, log data, etc., associated with the identification information of the surveying instrument 1 to the management server MS. The management server MS manages the data received from the surveying instrument 1 in a database associated with the identification information and model of the surveying instrument 1.

[0032] The management server MS stores not only the surveying instrument 1 but also a large number of surveying instruments S1 to S2, which may be the same model as the surveying instrument 1 or different models or types. n (These are collectively referred to as other surveying instruments S) are connected so that they can communicate. The management server MS collects and receives data from the other surveying instruments S in the same way as from the surveying instrument 1, and manages the data in a database in association with the identification information and model of each surveying instrument. Hereinafter, the data collected by the management server MS from the surveying instrument 1 and other surveying instruments in this way will be referred to as collected data.

[0033] Alternatively, the collected data may be collected by repeating test surveys in a factory or the like for the surveying instrument 1 and other surveying instruments S under various environments with variously changed environmental data 91, and associating the measurement data, error data, and environmental data at the time of measurement with identification information and model.

[0034] FIG. 5 is a diagram showing an example of training data. The training data is a set of environmental data at the time of measurement and error data at the time of measurement, which are collected from the collected data as described above for the same model. Here, an example is shown in which internal temperature data, internal humidity data, and internal air pressure data are used as environmental data. In addition, an example is shown in which errors related to distance and angle measurements, which are measured values, are used as error data. The error data is treated as a ratio to the measured values. It is preferable to normalize these environmental data before use.

[0035] 2-2 Error prediction model generator 3 6 is a configuration block diagram of the error prediction model generation device 3. The error prediction model generation device 3 is a computer, and includes a control unit 30, a communication unit 41, a main storage device 42, and an auxiliary storage device 43. The error prediction model generation device 3 may be a computer that constitutes the management server MS. Alternatively, it may be a separate computer that is configured to be able to communicate with the management server MS via a communication network N.

[0036] The communication unit 41 is a communication control device such as a network adapter, a network interface card, or a LAN card, and connects the error prediction model generation device 3 to the communication network N via a wired or wireless connection. The control unit 30 can send and receive various information to and from the management server MS and other external devices via the communication unit 41 and the communication network N.

[0037] The control unit 30 is a control and arithmetic unit configured with one or more processors such as CPUs (Central Processing Units), multi-core CPUs or GPUs (Graphics Processing Units), and memories such as RAM and ROM. The control unit 30 is connected to each hardware component constituting the error prediction model generation device via a bus.

[0038] The control unit 30 includes a teacher data generation unit 31, a model generation unit 32, and a model update unit 33 as functional units.

[0039] The teacher data generation unit 31 generates teacher data for each model from the collected data of the management server MS as a data set in which the environmental data at the time of measurement is used as an explanatory variable and the error data at the time of measurement is used as a target variable, and stores the data in the auxiliary storage device 43 as a teacher data DB (database) 43a.

[0040] The model generation unit 32 stores the training data DB 43a The model generator 32 reads out training data for the same model of surveying instrument from the database, performs machine learning on the environmental data at the time of measurement and the error data at the time of measurement as a set, and generates an error prediction model M. The generated error prediction model M is stored in the auxiliary storage device 43. The model generator 32 generates error prediction models M1, M2, ... M for each model. n (Here, n indicates the number of models. To indicate a specific model, use the symbol M x When the entire set is shown, it is denoted by the symbol M.) is generated, and each error prediction model is stored in the auxiliary storage device 43.

[0041] Machine learning is realized, for example, by a neural network that uses one or more layers of nonlinear units to predict an output for an input. Specifically, the techniques of DNN (Deep Neural Network), CNN (Convolutional Neural Network), and RNN (Recurrent Neural Network) can be used.

[0042] Furthermore, for machine learning, techniques such as SVR (Support Vector Regression), random forest, or Bayesian linear regression analysis may be used.

[0043] When new collected data is input from the management server MS, the model update unit 33 creates training data as a set of environmental data at the time of measurement included in the new collected data and error data at the time of measurement, and re-learns the error prediction model Mx of the model associated with the data using the same method as when the model was generated, thereby updating the error prediction model Mx.

[0044] As shown schematically in Figure 7, the error prediction model generation device 3 performs machine learning using environmental data 91 at the time of measurement and error data 92 in the measurement as a set of training data to generate a trained model, thereby making it possible to generate an error prediction model that can predict surveying errors with a surveying instrument of the relevant model in the environment indicated by the environmental data 91 input.

[0045] 3 Error prediction method 8 is a flowchart of the processing of the error prediction method by the error prediction device (surveying instrument 1) according to this embodiment. The error prediction is preferably performed at the start of the surveying work for the day, but may also be performed at the start of measurement at each instrument point. When an error prediction instruction is input by the operator, the process starts, and in step S01, the environmental data receiving unit 11 receives environmental data of the surveying site from the environmental sensor 22.

[0046] Next, in step 02, the error prediction unit 12 reads out the error prediction model M1 from the storage unit 26, inputs the environmental data 91 into the error prediction model M1, and predicts errors in measurement that will occur in the current environment. The error prediction model M1 does not necessarily have to be stored in the storage unit 26, but may be obtainable from an external device such as a management server via the communication unit 25 when necessary.

[0047] Next, in step S03, the error prediction unit 12 determines whether the predicted error exceeds a preset allowable value. The allowable value is a value that ensures measurement accuracy determined according to the model. If the tolerance is not exceeded (No), the process ends. If the tolerance is exceeded (Yes), the process proceeds to step S04, where the display creation unit 13 creates display data 94 for displaying the predicted error 93 on the display unit 23. Next, in step S05, the display unit 23 displays the display data 94, and the process ends. Note that in step S05, instead of or in addition to displaying the display data 94 on the display unit 23, the display data may be output to an external display device via a communication unit for display. In this case, the external display device includes at least a display unit such as a liquid crystal display, a storage unit for storing the display data 94, a communication unit capable of communicating with the surveying instrument 1, and a control unit for controlling the display.

[0048] Figure 9 (A)~(C) is an example of display data 94 for displaying a predicted error 93 on the display unit 23. FIGS. 9A and 9B are examples using the internal temperature data, internal humidity data, and internal air pressure data shown in FIG. 5 as environmental data, and FIG. 9C is an example using tilt stabilization time data as environmental data. As shown in FIGS. 9A to 9C, the display data 94 indicates a value 941 of the environmental data 91 and a value 942 of the predicted error 93, and also displays a message 943 alerting the operator that the predicted error will exceed the allowable value in measurement under that environment. The display data 94 may be configured so that, as shown in FIG. 9A, the window for the display data 94 is closed when "Confirm" is confirmed, for example by pressing the Enter key on the operation unit 24, and the operator is only notified that the predicted error will exceed the allowable value in measurement under that environment.

[0049] In this way, the operator can recognize that the measurement error will exceed the tolerance under that environment and make a decision regarding whether to stop the measurement or continue the measurement. Alternatively, as shown in Fig. 9(B), the display data 94 may display a selection button 944 presenting the following options: "Stop Measurement" to stop the measurement, "Correction Settings" to adjust the correction method depending on the magnitude of the error, and "Continue Measurement" to continue the measurement even though it is recognized that the error exceeds the tolerance.

[0050] 9(C), the display data 94 may display a message display 945 indicating the state indicated by the value of the environmental data 91, in addition to or instead of the value of the environmental data 91. For this purpose, messages corresponding to the numerical values ​​of the environmental data 91 may be stored in the storage unit 26 in the form of a table or the like. This configuration allows the worker to easily grasp the state indicated by the numerical value of the environmental data 91. When there are multiple types of environmental data 91, the display data 94 may include displays indicating the values ​​of the environmental data 91 and the state indicated by the values ​​of the environmental data 91 for all of the multiple types of environmental data 91.

[0051] In this way, the error prediction device according to this embodiment is configured to predict errors that will occur in the environment of the surveying site before actual measurement and to create display data 94 that displays the predicted errors 93. This makes it possible to recognize errors that will occur during measurement before measurement, and to eliminate unnecessary measurements such as discarding measurement values ​​after measurement.

[0052] This embodiment is configured to predict errors taking into account various environmental data 91. When calculating such errors, the surveying instrument of Patent Document 1 performs measurements under various conditions with varying temperatures and calculates a correction formula. Measurements for this purpose are usually performed with other environmental parameters fixed. However, factors that contribute to measurement errors depend not only on temperature but also on other environmental parameters. For example, the ambient temperature of the surveying instrument 1 affects the speed of light, resulting in distance measurement errors, and also causes thermal expansion of the tripod on which the surveying instrument 1 is mounted, resulting in angle measurement errors. In other words, a single environmental parameter can contribute to errors in various ways. While environmental factors may cause independent error variations, they may also interact with each other, resulting in a synergistic effect. An error calculation method that independently calculates each environmental parameter (type of environmental data) cannot address such error variations that result from synergistic effects.

[0053] In this embodiment, an error prediction model M is generated using multiple types of environmental data 91 and error data 92 as training data, and the error prediction model M is used to predict errors. This makes it possible to predict accurate errors according to the environment, taking into account the complex correlations between environmental parameters.

[0054] Furthermore, since the error prediction device according to this embodiment is configured as a surveying instrument 1, it is possible to predict errors and display the predicted error 93 using only the surveying instrument 1, and no other belongings are required other than the surveying instrument 1 used for the surveying work, so the burden on the worker is not increased.

[0055] II. Second embodiment 1. Error prediction device 10 is a configuration block diagram of a surveying instrument 1A, which is an error prediction device according to a second embodiment, which is a modification of the first embodiment. The surveying instrument 1A generally has the same hardware configuration as the surveying instrument 1, but the control and calculation unit 10A includes a contribution calculation unit 14, and also includes a display creation unit 13A instead of the display creation unit 13. In addition, the storage unit 26A includes a teacher data DB 43a used to generate the error prediction model M1.

[0056] The contribution calculation unit 14 calculates the contribution of each environmental data 91 to the prediction error 93 predicted by the error prediction unit 12 using the error prediction model M1. The contribution can be calculated by approximating the data set in the training data DB 43a used to create the error prediction model M1 using a linear regression method such as LIME (Local Interpretable Model-agnostic Explanations).

[0057] Furthermore, Shapley values ​​(SHAP (Shapley·Additive·exPlanation) Values) obtained using cooperative game theory may be calculated as values ​​indicating the degree of contribution of environmental data.

[0058] The contribution calculation unit 14 may also use, as an index indicating the contribution of the environmental data, the importance (Permutation Importance) obtained by rearranging explanatory variables as disclosed in Japanese Patent Application Laid-Open No. 2019-121162. Furthermore, as the contribution of the environmental data, a general index used when features are reduced using a recursive feature elimination method may also be used.

[0059] The display creating unit 13A generates display data 94A relating to the prediction error 93, which includes a display according to the contribution calculated by the contribution calculation unit 14 in addition to the prediction error 93.

[0060] 2. Error prediction method Fig. 11 is a flowchart of the processing of the error prediction method by the surveying instrument 1A according to this embodiment. Fig. 12 shows an example of training data used in this method. As in the first embodiment, the environmental data may include at least one of the environmental data exemplified in Fig. 3, but as shown in the figure, the environmental data 91 may include all of the environmental data exemplified in Fig. 3.

[0061] When an error prediction instruction is input by the operator, in steps S11 to S13, similar to steps S01 to S03, the environmental data receiving unit 11 receives environmental data from the environmental sensor 22, the error prediction unit 12 reads out the error prediction model M1 from the memory unit 26, inputs the environmental data 91 into the error prediction model M1, predicts the prediction error 93, and determines whether the prediction error 93 exceeds the allowable value.

[0062] If the tolerance is not exceeded (No), the process ends. If the tolerance is exceeded (Yes), the process proceeds to step S14, where the contribution calculation unit 14 calculates the contribution of the environmental data 91 to the prediction error 93, for the prediction error 93 predicted using the error prediction model M1.

[0063] 13 shows an example of the calculation result of the contribution degree calculation unit 14. The contribution degree calculation unit 14 calculates the contribution degree of each environmental data 91 to the prediction error 93.

[0064] Next, in step S15, the display creation unit 13A creates display data 94A for displaying information related to the prediction error. FIGS. 14A and 14B are examples of the display data 94A. As shown in FIG. 14A, the display data 94A includes a value 942 of the prediction error 93, a warning message 943, and an indication 946 of the type and value of environmental data 91 with a high contribution. In addition to or instead of the indication 946, the display data 94A may also include a message 947 indicating the state indicated by the environmental data 91.

[0065] 14(B), the environmental data 91 may further include a message 948 notifying a countermeasure for resolving the state indicated by the environmental data 91. For this purpose, data of the message for notifying a countermeasure for resolving the state indicated by the environmental data 91 may be stored in the storage unit 26A in association with the contribution distribution of the environmental data 91.

[0066] As shown in Figures 14(A) and 14(B), the environmental data 91 with the highest contribution may be displayed only for one environmental data 91 that is the factor that most affects the error. Alternatively, as shown in Figure 14(C), multiple environmental data 91 that are ranked higher (for example, within the top three) than the environmental data 91 that is the factor that affects the error may be displayed.

[0067] In this embodiment, the contribution calculation unit 14 is provided to identify environmental data 91 that have a high contribution to the error from among the factors that affect the error, and display 946 relating to the environmental data 91 with a high contribution to the error is included in the display data 94A together with the predicted error 93. This allows the operator to grasp the occurrence of a possible error and its cause in advance without having to perform a measurement or a detailed analysis. As a result, the operator can take measures to prevent an error from occurring before performing the measurement.

[0068] In particular, even in situations where it is difficult for an operator to identify the cause, such as an environmental condition where the stability of the installation surface is slightly unstable in the boundary area of ​​the measurable temperature range, it is possible to create data that shows which factor has a greater contribution and how to respond.By displaying this, the operator can easily understand the causes of errors that may occur under the current circumstances.

[0069] Furthermore, if the display data 94A is configured to include a message 948 that notifies the operator of measures to resolve the condition indicated by the environmental data 91, the operator can take measures to prevent errors from occurring without needing specialized knowledge.

[0070] III. Third embodiment 1. Error prediction device 15 is a configuration block diagram of an error prediction device according to the third embodiment. The error prediction device is configured as an information processing device 5 that is communicably connected to a surveying instrument 200 of the same model as the surveying instrument 1 via a communication network N.

[0071] The information processing device 5 includes a control and calculation unit 50, a communication unit 61, a main memory device 62, and an auxiliary memory device 63. The hardware configurations of the control and calculation unit 50, the communication unit 61, the main memory device 62, and the auxiliary memory device 63 are similar to those of the control unit 30, the communication unit 41, the main memory device 42, and the auxiliary memory device 43 of the error prediction model generation device 3, and therefore will not be described again.

[0072] The auxiliary storage device 43 stores error prediction models M (M1 to M n ) is stored. The information processing device 5 may be a computer that constitutes the management server MS. The control and calculation unit 50 includes, as functional units, an environmental data receiving unit 51, an error prediction unit 52, a display creation unit 53, and a display command unit .

[0073] The environmental data receiving unit 51 receives the environmental data 91 detected by the environmental sensor 22 of the surveying instrument 200 via the communication network N in association with the model information of the surveying instrument 200 . The error prediction unit 52 inputs the received environmental data 91 into an error prediction model (here, error prediction model M1) for the model corresponding to the model information associated with the environmental data 91, and predicts a predicted error 93 that will occur in the measurement values ​​of the surveying instrument 200 under the environment of the surveying site. The display creation unit 53 creates display data 94 for displaying the prediction error 93. The display command unit 54 commands the surveying instrument 200 via the communication network N to display the display data 94.

[0074] The surveying instrument 200 is the same model of surveying instrument as the surveying instrument 1, and has the same hardware configuration as the surveying instrument 1, except for the following points. The control calculation unit 210 does not include the environmental data receiving unit 11, the error prediction unit 12, or the display creation unit 13, but instead includes, as functional units, an environmental data transmission unit 15 and a display execution unit 16. In addition, the memory unit 226 does not include the error prediction model M1, but includes model information 226a that indicates the model of the surveying instrument 200.

[0075] The environmental data transmission unit 15 transmits the environmental data 91 acquired by the environmental sensor 22 to the information processing device 5 in association with the model information 226a in accordance with instructions from the worker. The display execution unit 16 displays the display data received from the information processing device 5 on the display unit 23 in accordance with a display command from the information processing device 5 .

[0076] 2. Error prediction method FIG. 16 is a flowchart of an error prediction method using the information processing device 5, which is an error prediction device according to the third embodiment. When an operator inputs an instruction to predict an error from the surveying instrument 200, processing begins, and in step S21, the environmental data transmission unit 15 transmits environmental data 91 of the surveying site detected by the environmental sensor 22 to the information processing device 5 in association with the model information 226a.

[0077] Next, in step S22, the environmental data receiving unit 51 receives the environmental data 91. Next, in step S23, the error prediction unit 52 reads out from the memory unit 226 the error prediction model M1 corresponding to the model information 226a associated with the environmental data 91, inputs the environmental data 91 into the error prediction model M1, and predicts errors in measurement that will occur in the environment of the surveying site.

[0078] Next, in step S24, the error prediction unit 52 determines whether the prediction error 93 exceeds a preset allowable value. If the tolerance is not exceeded (No), the process ends. If the tolerance is exceeded (Yes), the process proceeds to step S25, where the display creation unit 53 creates display data 94 for displaying the predicted error 93 on the display unit 23. Next, in step S26, the display command unit 54 sends a command to the surveying instrument 200 to display the display data 94.

[0079] Next, in step S27, the display execution unit 16 of the surveying instrument 200 displays the display data 94 on the display unit 23 in accordance with the display command, and then ends the process.

[0080] In this way, even if the error prediction unit 12 and the error prediction model M are not provided in the surveying instrument, but in an information processing device 5 connected via a communication network N, and the surveying instrument 200 makes the information processing device 5 on the communication network N make predictions based on the environmental data 91 acquired at the surveying site, the same effect as that of the error prediction device of the first embodiment can be achieved.

[0081] In general, it is easy to configure the information processing device 5 as a control and calculation unit with higher performance than the surveying instrument 200, and therefore it is possible to reduce the load of calculation processing in the surveying instrument 200.

[0082] IV Fourth Embodiment 1. Error prediction device 17 is a configuration block diagram of a surveying instrument 1B, which is an error prediction device according to the fourth embodiment. The surveying instrument 1B has roughly the same hardware configuration as the surveying instrument 1, but differs in that it further includes a position information acquisition unit 27, a control and calculation unit 10B includes an error resolution prediction unit 17, and a display creation unit 13B instead of the display creation unit 13. The surveying instrument 1B is also connected to a weather server WS via a communication network N.

[0083] The weather server WS is, for example, a server managed by a weather information provider, and it is possible to obtain environmental forecast data 95 related to weather, such as the weather, temperature, humidity, and air pressure for each region, at a predetermined time interval, such as every 30 minutes, from the weather server WS.

[0084] The location information acquisition unit 27 is, for example, a GNSS receiver, and can acquire location information of the surveying instrument 1 B. Alternatively, the location information may be acquired by the operator inputting the address of the current location of the surveying instrument 1 B via the operation unit 24.

[0085] The error resolution prediction unit 17 acquires environmental forecast data 95 for the area surrounding the current position of the surveying instrument 1B from the weather server WS, and based on the environmental forecast data 95, determines the time when the predicted error will not exceed the allowable value, i.e., will be within the allowable value, and includes the predicted error 93 in the display data 94B that displays it.

[0086] 2. Error prediction method FIG. 18 is a flowchart of an error prediction method using a surveying instrument 1B, which is an error prediction device according to the fifth embodiment.

[0087] The processing of steps S31 to S33 is the same as the processing of steps S01 to S03, and therefore the description will be omitted. If the predicted error exceeds the allowable value in step S33 (Yes), the processing proceeds to step S34, where the error resolution prediction unit 17 acquires position information of the surveying instrument 1B.

[0088] Next, in step S35, the error resolution prediction unit 17 acquires, from the weather server WS, environmental forecast data 95 for a predetermined time period (e.g., six hours) for the current location of the surveying instrument 1B, i.e., the surrounding area of ​​the surveying site. FIG. 19 shows an example of the acquired environmental forecast data 95. The environmental forecast data 95 is, for example, forecast values ​​related to weather, such as temperature, humidity, and air pressure, at 30-minute intervals. For example, if the current time is 12:15, six hours' worth of data from 12:30 to 18:30 is acquired. Here, the temperature, humidity, and air pressure data for the surrounding area of ​​the surveying site are considered to correspond to the ambient temperature, ambient humidity, and ambient air pressure of the surveying instrument 1B. Therefore, the error resolution prediction unit 17 acquires environmental forecast data 95 corresponding to the environmental data 91. That is, if the environmental data 91 includes ambient temperature data, ambient humidity data, and ambient air pressure data, the temperature data, humidity data, and air pressure data are acquired as the environmental forecast data 95.

[0089] Next, in step S36, the error resolution prediction unit 17 inputs the environmental forecast data 95 acquired, which is the environmental forecast data 95 after a predetermined time (for example, 30 minutes), into the error prediction model M1, and predicts the prediction error 93 under the environment of the environmental forecast data 95 after 30 minutes.

[0090] Next, in step S37, the error resolution prediction unit 17 determines whether the prediction error 93 exceeds the allowable value, and if it does (Yes), returns to step S36, inputs the environmental forecast data 95 for another 30 minutes into the error prediction model M1, and predicts the prediction error under the environment of the environmental forecast data 95 for another 30 minutes into the environment.

[0091] If the prediction error falls within the allowable range (No) in step S37, the process proceeds to step S38, where the display creation unit 13B creates display data 94B including the prediction error 93 and an indication 949 of the time when the error will be resolved. Fig. 20 shows an example of the display data 94B. Then, in step S39, the display data 94B is displayed on the display unit 23, and the process ends.

[0092] In step S37, if the prediction error exceeds the allowable value even after all the data obtained from the weather server WS has been evaluated, the display data 94B may display a message such as, for example, "Proper measurements will not be possible within the next six hours."

[0093] In this way, in this embodiment, the environmental forecast data 95 for the area surrounding the current position of the surveying instrument 1B is acquired from the weather server WS, and the time when the predicted error 93 will not exceed the tolerance, i.e., will be within the tolerance, is determined and displayed based on the environmental forecast data 95. As a result, if an operator is unable to perform a proper survey in the current environment, he or she can know how long he or she will have to wait before being able to perform a proper survey, thereby reducing the burden of having to wait unnecessarily for the environment to recover to one where a proper survey is possible.

[0094] V Fifth embodiment Fig. 21(A) is an external perspective view of an eyewear display device (hereinafter referred to as "eyewear device") 7, which is an error prediction device according to the fifth embodiment, and Fig. 21(B) is a diagram showing how surveying work is carried out at a surveying site using the eyewear device 7 near a surveying instrument S of the same model as the surveying instrument 1. The eyewear device 7 is a wearable device that is worn on the head of the worker.

[0095] 22 is a configuration block diagram of the eyewear device 7. The eyewear device 7 includes a control and calculation unit 70, a display unit 81, an environment sensor 82, a communication unit 83, a relative position detection sensor 84, a relative direction detection sensor 85, a memory unit 86, and an operation switch 87.

[0096] The display unit 81 is a goggle-lens type transmission display that covers both eyes of the worker when worn by the worker. As an example, the display unit 81 is an optical see-through display using a half mirror, and is configured to display at least a virtual image of the display data 94 created by the control and calculation unit 70 superimposed on the work site scenery. Alternatively, the display unit 81 may be a video see-through display equipped with a camera (not shown) that captures the front view of the eyewear device 7 in real time, and may display an image in which the display data 94 created by the control and calculation unit 70 is superimposed on the front view image acquired by the camera. Furthermore, the projection method may be a virtual image projection method or a retinal projection method. The display unit 81 displays the display data 94 created by the display creation unit 73.

[0097] The environmental sensors 82 include sensors from the environmental sensors 22 that detect the surrounding environment of the surveying instrument, such as an ambient temperature sensor, an ambient humidity sensor, an ambient air pressure sensor, a visibility meter, an anemometer, etc. Therefore, the environmental data 91 acquired by the environmental sensors 82 is surrounding environment data. The communication unit 83 is a communication control device similar to the communication unit 25, and enables transmission and reception of information to and from the surveying instrument S.

[0098] The relative position detection sensor 84 performs wireless positioning using a GPS antenna, a WiFi (registered trademark) access point, an ultrasonic oscillator, etc. installed at the observation site, and detects the position of the eyewear device 7 within the observation site.

[0099] The relative direction detection sensor 85 is a combination of a triaxial acceleration sensor or gyro sensor and an inclination sensor. The relative direction detection sensor 85 detects the inclination of the eyewear device 7 with the up / down direction defined as the Z-axis, the left / right direction defined as the Y-axis, and the front / back direction defined as the X-axis.

[0100] The eyewear device 7 is used at the surveying site to support the surveying work of the surveying instrument S by acquiring the position and direction using the relative position detection sensor 84 and the relative direction detection sensor 85, and converting the internal coordinate system of the surveying instrument S installed at the same surveying site using the reference point and reference direction set at the measurement site, making it possible to manage it in the coordinate system of the eyewear device 7.

[0101] The storage unit 86 is, for example, a memory card. The storage unit 86 stores a program for the control and calculation unit 70 to execute functions. The storage unit 86 also includes an error prediction model M1 that matches the model of the surveying instrument S.

[0102] The operation switch 87 is, for example, a push button provided on the outer surface of the display unit 81, as shown in Figure 21, and enables the power of the eyewear device 7 to be turned on / off and the operator to make selections, input instructions, etc. according to the display on the display unit 81.

[0103] The control and calculation unit 70 is, for example, an arithmetic and control unit in which at least a processor such as a CPU and memory (RAM, ROM, etc.) are implemented in an integrated circuit. The control and calculation unit 70 includes, as functional units, an environmental data reception unit 71, an error prediction unit 72, and a display creation unit 73 that implement the same functions as the environmental data reception unit 11, the error prediction unit 12, and the display creation unit 13 of the surveying instrument 1.

[0104] With the above configuration, it is possible to display the predicted error of the surveying instrument S that is used simultaneously, using the eyewear device 7, which is a display device used to support surveying work near the surveying instrument S. This makes it possible to achieve the same effect as in embodiment 1, at least with respect to the surrounding environment of the surveying instrument. In particular, by using the eyewear device 7, the operator can check the error prediction result even if he or she is not near the surveying instrument S. Furthermore, the eyewear device 7 may acquire internal environment data acquired by the surveying instrument S via the communication unit 83, and the environmental data receiving unit 71 may receive the internal environment data.

[0105] While preferred embodiments of the present invention have been described above, the above embodiments are merely examples of the present invention, and these can be combined based on the knowledge of those skilled in the art, and such combinations are also included in the scope of the present invention. Specifically, modifications to the error prediction device (surveying instrument 1) of the first embodiment in the second and fourth embodiments may be applied to the error prediction device (information processing device 5) and the fifth error prediction device (eyewear device 7) of the third embodiment. [Explanation of symbols]

[0106] 1,1A:Surveying equipment (error prediction device) 5: Information processing device (error prediction device) 7: Eyewear display device (error prediction device) 11, 51, 71: Environmental Data Reception Department 12, 52, 72: Error prediction section 13, 53, 73: Display creation section 14: Contribution calculation section 22,82: Environmental sensors 23,81:Display section 25, 61, 83: Communications Department 27: Location information acquisition section 91: Environmental Data 92: Error data 93: Prediction error 94, 94A, 94B: Display data 95: Environmental forecast data M: Error prediction model N: Communication network WS: Weather Server

Claims

1. a control and calculation unit having a processor and a memory; The control calculation unit an environmental data receiving unit that receives environmental data of a surveying site where the surveying instrument is installed; an error prediction unit that inputs the environmental data of the surveying site into an error prediction model and predicts a predicted error that will occur in the surveying results of the surveying instrument under the environment of the surveying site; a display creation unit that creates display data for displaying the prediction error when the prediction error exceeds a tolerance value; The error prediction device is characterized in that the error prediction model is a trained model generated by machine learning using a set of environmental data indicating the environment at the time of surveying and error data in the surveying results as training data for a surveying instrument of the same model as the surveying instrument.

2. 2. The error prediction device according to claim 1, wherein the surveying instrument is configured to include an environmental sensor that acquires the environmental data and inputs it into the environmental data receiving unit, and a display unit that displays the display data created by the display creation unit.

3. an information processing device configured to be able to communicate with the surveying instrument via a communication network; 2. The error prediction device according to claim 1, wherein the surveying instrument comprises an environmental sensor that acquires the environmental data and inputs it into the environmental data receiving unit, and a display unit that displays the display data created by the display creation unit.

4. The error prediction device described in Claim 1 is characterized in that it is an eyewear display device that is equipped with a display unit that displays the display data created by the display creation unit, a communication unit that can communicate with the surveying instrument, and an environmental sensor that acquires the environmental data and inputs it into the environmental data receiving unit, and is capable of displaying the display data so that it can be observed superimposed on the site scenery.

5. the environmental data includes a plurality of environmental data; the control calculation unit includes a contribution calculation unit that calculates a contribution of environmental data to the prediction error; 5. The error prediction device according to claim 1, wherein the display data includes information about the environmental data that has a high degree of contribution to the prediction error.

6. a position information acquisition unit that acquires a current position of the surveying instrument; configured to be able to communicate with a weather server via a communication network; An error prediction device according to any one of claims 1 to 5, characterized in that it obtains environmental forecast data for the area surrounding the current position of the surveying instrument from the weather server, determines the time at which the value of the predicted error based on the environmental forecast data will be equal to or less than the allowable value, and includes the time in the display data.

7. a control and calculation unit having a processor and a memory, Obtain environmental data from the surveying site where the surveying equipment is installed, inputting the environmental data of the surveying site into an error prediction model to predict a predicted error that will occur in the surveying results of the surveying instrument under the environment of the surveying site; generating an indication of the prediction error if the prediction error exceeds a tolerance; The error prediction method is characterized in that the error prediction model is a trained model generated by machine learning using the environmental data indicating the environment at the time of surveying and a set of errors in the surveying results as training data for a surveying instrument of the same model as the surveying instrument.

Images