WO2015012382A1

WO2015012382A1 - Buried-pipeline measurement device and buried-pipeline measurement method

Info

Publication number: WO2015012382A1
Application number: PCT/JP2014/069656
Authority: WO
Inventors: 宮崎　幸; 上道　司; 俊男山根; 健次小阪; 勝久番上; 正人番上; 諭田所; 和則大野
Original assignee: 積水化学工業株式会社; 株式会社アスコ; 国立大学法人東北大学; 特定非営利活動法人国際レスキューシステム研究機構
Priority date: 2013-07-26
Filing date: 2014-07-25
Publication date: 2015-01-29
Also published as: JP6202559B2; JP2015025753A

Abstract

　In order to make it possible to estimate the travel route of an instrumented vehicle and estimate the position of a buried pipeline even when the buried pipeline is devoid of characteristic landmarks, this buried-pipeline measurement device includes: an instrumented vehicle provided with an odometry means for measuring a travel distance, an IMU for measuring the orientation of the instrumented vehicle, and three laser measurement devices (6-8) for scanning in three mutually intersecting planes and obtaining internal-surface-shape information of the pipeline; a position specifying means for specifying a measurement start position, a measurement end position, and an orientation; a sequential position estimation calculation means for sequentially estimating and calculating sequential position information of the instrumented vehicle using the travel distance obtained by the odometry means and the orientation information obtained by the IMU; and a sequential position correction calculation means for performing a correction calculation on the sequential position information of the instrumented vehicle on the basis of the internal-surface-shape information of the pipeline obtained by the three laser measurement devices (6-8); an overall correction of the correction-calculated sequential position information being performed by a three-dimensional SLAM algorithm and overall position information in the travel route being specified after the measurement end position is reached.

Description

Buried pipe measuring device and buried pipe measuring method

The present invention relates to a technique for measuring the geographical coordinates of the inner surface of a buried pipe such as a sewage pipe, and more particularly to measurement of the buried position of the buried pipe and the technique for measuring the inner shape of the buried pipe. It is.

In recent years, with the aging of infrastructure facilities such as sewage pipes, various investigations have become necessary for the replacement of aging pipes.
For example, the investigation and diagnosis work includes the identification of the position where the aging pipe is buried, the quantification of the degree of deterioration of the aging pipe, and the accuracy measurement of the inner surface shape required for rehabilitation.
Among these, there is a case where there is not enough information in the official management ledger to specify the buried position, and the actual situation is that excavation and manual surveying are performed.
In addition, there is a rehabilitation method for laying another pipe in the existing pipe as a measure to restore the old pipe, but before rehabilitation, an accurate survey of the inner surface shape is required. The diameter required for construction is determined using a dummy pipe insertion survey method, which is inserted over the entire pipe path and inferred the rough shape.

JP 2013-113702 JP 2009-121945

When excavation work and manual surveying are performed to identify the location of buried pipes such as sewage pipes, blocking public roads and enormous surveying costs are required. In the future, the aging social infrastructure will increase more and more, and solving these cost issues will be an urgent need to proceed with the development of the aging infrastructure stock without delay.
In surveying the inner shape of the buried pipe, the dummy pipe insertion survey method does not know the details of the unevenness of the pipe, meandering, and steps, and may not be constructed as planned.

As a method of measuring the three-dimensional position of a survey object with a moving object, the vehicle is equipped with an IMU (Inertial Measurement Unit), odometry (odometer), GPS receiver, and GPS information. There is a method (Japanese Patent Laid-Open No. 2013-113702) for performing surveying while successively correcting the self-position based on the above (see Patent Document 1).
In addition, in order to correct the accumulated error of IMU and the error caused by the variation in GPS reception in the same surveying as above, Bayesian filter processing is performed on the obtained data (Kalman filtering) and the surveying method is performed while increasing the accuracy of the self-location. (Japanese Patent Laid-Open No. 2009-121945) (see Patent Document 2).

Each of these prior arts is a method of identifying self-position by successively correcting the error by GPS which can be received at intervals of about 1 second while measuring continuous self-position movement by odometry and IMU.
However, when these technical algorithms are applied to the geographical coordinate positioning of buried pipes, sequential GPS correction cannot be performed.

On the other hand, as an algorithm for minimizing IMU and odometry errors, in recent years, an attempt has been made to use a Graph-SLAM algorithm which is a kind of three-dimensional SLAM (Simultaneous Localization And Mapping) algorithm. The three-dimensional SLAM is an algorithm for simultaneously estimating the position / posture of a measuring device and a surrounding map in a three-dimensional space, and the Graph-SLAM algorithm is one of three-dimensional SLAM solutions.
Graph in the Graph-SLAM algorithm means a graph of a figure. In the Graph-SLAM algorithm, every time the surroundings are measured with a measuring device (cart / sensor), the measuring device is a node (bundling) and the link is made between them. Then, a graph is created in which the link connection is an estimation error regarding the position. Each time a landmark is observed, a subgraph is added to the previous graph, with the measurement device and the landmark as a node (cohesion), the observation between them as a link, and the strength of the link connection as an observation error. . This is a method of constructing a map of the position / posture of the measuring device and its surroundings by estimating the shape of the most applicable graph when the positions of some nodes in the completed graph are known. (See Non-Patent Document 1)

However, it is difficult to set a characteristic landmark in the buried pipeline, and even if it can be installed, it is not realistic to install it over the entire pipeline.
Therefore, even if there are no characteristic landmarks in the buried pipeline, if the coordinates of the measurement start position and measurement end position can be specified at least, the travel circuit of the measurement vehicle in the buried pipeline can be The present invention has been made for the purpose of providing a technique that enables accurate estimation and further enables the geographical coordinate positioning of an embedded pipe.

The buried pipe measuring device according to claim 1 of the present invention is
Travel distance measurement means for measuring travel distance and outputting travel distance information, attitude measurement means or inertia measurement means for measuring the attitude of a measuring vehicle and outputting posture information, and measuring the inner surface shape of a pipe to measure the inner surface A measuring vehicle capable of traveling in a pipeline with an inner surface shape measuring means for outputting shape information;
Start position specifying means for specifying the position and orientation of the measurement vehicle at the measurement start position and outputting measurement start position information;
An end position specifying means for specifying the position and orientation of the measurement vehicle at the measurement end position and outputting the measurement end position information;
While traveling in the pipeline, using the travel distance information by the travel distance measurement means and the attitude information by the attitude measurement means or the inertia measurement means, the sequential position information of the measurement vehicle based on the measurement start position is obtained. Sequential position estimation calculation means for performing sequential estimation calculation;
Based on the inner surface shape information of the pipeline by the inner surface shape measuring means, sequential position correction calculating means for sequentially correcting and calculating the sequential position information of the measuring vehicle while traveling in the pipeline;
After reaching the measurement end position, a three-dimensional SLAM that performs position estimation and map creation in parallel on the basis of the measurement start position information and the measurement end position information based on the sequential position information that has been subjected to the successive correction calculation. And a total correction calculation means for performing total correction calculation processing by a (Simultaneous Localization And Mapping) algorithm and specifying overall position estimation information of a route traveled by the measurement vehicle.

In claim 2, further,
Based on the overall position estimation information of the route traveled by the measuring vehicle and the inner surface shape information of the pipe line by the inner surface shape measuring unit obtained by the all correction calculation unit,
The present invention is characterized by comprising a pipeline geographical coordinate estimating means for estimating the position of the inner surface of the pipeline and estimating the geographic coordinates of the pipeline.
In claim 3,
As the inner surface shape measuring means, a laser measuring device for scanning the inner surface of the pipe line is used.
In claim 4,
The inner surface shape measuring means is configured to obtain information on the inner surface shape of the pipeline by scanning in three non-parallel planes.

In claim 5,
The inner surface shape measuring means is configured to scan at least a cross-sectional shape of the pipe line by scanning in a plane perpendicular to the pipe axis.
In claim 6,
The inner surface shape measuring means scans a plane specified by a tube axis direction component and a horizontal direction component orthogonal to the tube axis, and measures at least the cross-sectional shape of the pipe in the tube axis direction. It is configured.

In claim 7,
The inner surface shape measuring means is configured to scan a vertical plane including a component in the tube axis direction and measure at least a longitudinal sectional shape in the tube axis direction of the conduit.
In claim 8,
An elevating mechanism is provided for elevating and lowering the gantry on which the inner surface shape measuring means is disposed.

In claim 9,
The sequential position estimation calculation unit uses the travel distance information by the travel distance measurement unit and the posture information by the posture measurement unit or the inertial measurement unit at the predetermined time for each predetermined time interval. The sequential position information of the measurement vehicle with respect to the vehicle is sequentially updated.
In claim 10,
The sequential position correction calculation means includes estimated inner surface shape information based on the sequential position information of the measurement vehicle by the sequential position estimation calculation means and the inner surface shape information by the inner surface shape measurement means at predetermined time intervals. Based on this, the sequential position information is subjected to sequential correction calculation processing.

In claim 11,
The total correction calculation process by the three-dimensional SLAM algorithm is as follows:
In the three-dimensional space, the sequential position information that has been subjected to the sequential correction calculation is subjected to all correction calculation processing based on the measurement start position information and the measurement end position information, so that position estimation in the three-dimensional space and map creation are performed in parallel. It is characterized by being executed.

In claim 12,
As the three-dimensional SLAM algorithm,
With the specified measurement start position and the specified measurement end position as landmarks,
The sequential position information of the measurement vehicle between the measurement start position and the specified measurement end position is a link and a node between them,
Create a graph with measurement error of link strength,
It is characterized by using a Graph-SLAM algorithm that executes the map creation in parallel with the position / posture of the measurement vehicle by estimating the most applicable shape of the graph.

The buried pipe measuring method according to claim 13 is:
Travel distance measurement means for measuring travel distance and outputting travel distance information, attitude measurement means or inertia measurement means for measuring the attitude of a measuring vehicle and outputting posture information, and measuring the inner surface shape of a pipe to measure the inner surface A method for measuring a buried pipeline using a measuring vehicle capable of traveling in a pipeline with an inner surface shape measuring means for outputting shape information,
A first step of obtaining measurement start position information by specifying the position and orientation of the measurement vehicle at the measurement start position;
While traveling in the pipeline, using the travel distance information by the travel distance measurement means and the attitude information by the attitude measurement means or the inertia measurement means, the sequential position information of the measurement vehicle based on the measurement start position is obtained. A second step of performing successive estimation operations;
A third step of successively correcting and calculating the sequential position information of the measuring vehicle while traveling in the pipeline based on the inner surface shape information of the pipeline by the inner surface shape measuring means;
A fourth step of obtaining the measurement end position information by specifying the position and orientation of the measurement vehicle at the measurement end position after reaching the measurement end position;
After reaching the measurement end position, a three-dimensional SLAM that performs position estimation and map creation in parallel on the basis of the measurement start position information and the measurement end position information based on the sequential position information that has been subjected to the successive correction calculation. (Simultaneous Localization And Mapping) includes a fifth step of performing all correction calculation processing and specifying the overall position estimation information of the route traveled by the measurement vehicle.

According to the present invention, the measurement start position coordinates of the pipe to be measured are specified, the travel distance measuring means, the posture measuring means or the inertia measuring means, and the inner surface shape measuring means for measuring the inner surface shape of the pipe. Move the equipped measuring vehicle in the pipeline,
Using the travel distance by the travel distance measurement means and the posture information by the posture measurement means or the inertia measurement means, the sequential position information of the measurement vehicle with respect to the measurement start position is sequentially obtained while traveling in the pipeline. Based on the inner surface shape information of the pipeline by the inner surface shape measuring means, the sequential position information of the measurement vehicle while traveling in the pipeline is corrected and calculated.
And after reaching the measurement end position,
Specify the measurement end position coordinates,
Since the correction-calculated sequential position information is fully corrected by the three-dimensional SLAM algorithm based on the measurement start position information and the measurement end position information, the entire position information of the route traveled by the measurement vehicle is specified. ,
It is possible to accurately estimate the travel route in the buried pipeline that cannot use GPS directly.

According to claim 2, in addition to the above configuration,
Based on the overall position estimation information of the route traveled by the measuring vehicle and the inner surface shape information of the pipe line by the inner surface shape measuring unit obtained by the all correction calculation unit,
Since it includes a pipeline geographic coordinate estimation means for estimating the position of the inner surface of the pipeline and estimating the geographic coordinates of the pipeline,
It is possible to accurately estimate the position of the pipeline based on the travel route.
According to the third aspect of the present invention, since the inner surface shape measuring means uses a laser measuring device that scans the inner surface of the pipeline, it is possible to configure a measuring device for the buried pipeline at a small size and at a low cost.
According to the fourth aspect of the present invention, the inner surface shape measuring means scans the three non-parallel planes to obtain the inner surface shape information of the pipe line, so that the inner surface shape information of the pipe line can be obtained in more detail. It is possible to more accurately grasp the situation such as unevenness of the pipeline, meandering and the like.

According to the fifth aspect, since the inner surface shape measuring means scans in a plane perpendicular to the tube axis, it is easy to measure the cross-sectional shape of the pipe line perpendicular to the tube axis.
According to the sixth aspect, the inner surface shape measuring means scans the plane specified by the tube axis direction component and the horizontal direction component orthogonal to the tube axis, so the shape of the side surface portion of the inner surface of the pipe line Easy to measure.

According to the seventh aspect, since the inner surface shape measuring means scans the vertical plane including the component in the tube axis direction, it is easy to measure the shapes of the ceiling and the floor on the inner surface of the pipe.
According to the eighth aspect of the present invention, since the elevating mechanism for elevating the gantry on which the inner surface shape measuring means is disposed is provided, the height of the inner surface shape measuring means coincides with the tube axis even in different pipes. It is easy to arrange and make it easy to grasp the shape of the inner surface of the pipe.

According to the ninth aspect, the successive position estimation calculation means obtains the travel distance information by the travel distance measurement means and the posture information by the posture measurement means or the inertia measurement means at the predetermined time for each predetermined time interval. Since the sequential position information of the measurement vehicle with respect to the measurement start position is sequentially updated, the sequential position of the measurement vehicle in the actual buried pipeline can be estimated.
According to the tenth aspect, the successive inner surface shape measuring means and the inner surface shape measuring means based on the successive position information of the measuring vehicle by the successive position estimation computing means by the successive position correction computing means. Since the sequential position information is sequentially corrected and calculated based on the inner surface shape information according to the above, the sequential position of the measuring vehicle in the actual buried pipeline can be estimated more accurately.

In claim 11,
Since all correction calculation processing by the SLAM algorithm in the three-dimensional space is performed, the actual position of the buried pipeline can be estimated more accurately.

In claim 12,
As the three-dimensional SLAM algorithm, since the Graph-SLAM algorithm using the specified measurement start position and the specified measurement end position as landmarks is used, it is possible to estimate the position of the actual buried pipeline more accurately. Can do.

The buried pipe measuring method according to claim 13 is:
While traveling in the pipeline, using the travel distance information by the travel distance measurement means and the attitude information by the attitude measurement means or the inertia measurement means, the sequential position information of the measurement vehicle based on the measurement start position is obtained. A second step of performing successive estimation operations;
A third step of successively correcting and calculating the sequential position information of the measuring vehicle while traveling in the pipeline based on the inner surface shape information of the pipeline by the inner surface shape measuring means;
A fourth step of obtaining the measurement end position information by specifying the position and orientation of the measurement vehicle at the measurement end position after reaching the measurement end position;
After reaching the measurement end position, a three-dimensional SLAM that performs position estimation and map creation in parallel on the basis of the measurement start position information and the measurement end position information based on the sequential position information that has been subjected to the successive correction calculation. (Simultaneous Localization And Mapping) includes a fifth step of performing all correction calculation processing and specifying the overall position estimation information of the route traveled by the measurement vehicle.
It is possible to accurately estimate the travel route in the buried pipeline where GPS cannot be directly used, and it is possible to accurately estimate the position of the pipeline based on the travel route.

It is a schematic diagram which shows the one part cross section of the pipe line with which measurement object was embedded in this invention. It is a side view of the measuring vehicle in this invention. It is a front view of the measurement vehicle. It is drawing of the mount frame of the said measuring vehicle, (A) is a top view, (B) is a side view. It is explanatory drawing for demonstrating the scanning plane of each laser measuring device. It is a flowchart of the measurement calculation process in this invention. It is explanatory drawing for demonstrating the component of Roll, Pitch, and Yaw of the said measurement vehicle. It is explanatory drawing for demonstrating Graph-SLAM in this invention. It is side surface sectional drawing of the manhole for demonstrating the 1st specific method for pinpointing the coordinate position of the measurement vehicle in this invention. It is a plane sectional view of a manhole for explaining the 2nd concrete method for specifying the coordinate position of a measurement car in the present invention. It is side surface sectional drawing of the manhole for demonstrating the 3rd specific method for pinpointing the coordinate position of the measurement vehicle in this invention. It is side surface sectional drawing of the manhole for demonstrating the 4th specific method for pinpointing the coordinate position of the measurement vehicle in this invention. It is side surface sectional drawing of the pipe line for demonstrating the technique which isolate | separates meandering and unevenness of a pipe | tube in this invention.

FIG. 1 is a schematic diagram showing an outline of a cross section of a buried pipeline to be measured,
m1 is a manhole at the measurement start position, m2 is a manhole at the measurement end position, and 1 is a measurement vehicle arranged in the pipe P to be measured. The measurement vehicle 1 moves while measuring the inside of the pipeline P from the manhole m1 at the measurement start position to the manhole m2 at the measurement end position. When the inner diameter of the pipe P is sufficiently large, the operator can push it and move it. However, when the inner diameter of the pipe P is small, a rope or the like is connected to the hook of the measuring wheel 1 to complete the measurement. It is moved by pulling from the manhole m2 at the position.

Next, the structure of the measuring wheel 1 will be described with reference to FIGS.
The measuring vehicle 1 includes, for example, a vehicle body 2 having two front wheels and two rear wheels, and a horizontal base 3 disposed on the upper portion of the vehicle body 2. Note that FIG. 2 shows a state in which the measuring vehicle 1 is arranged in a horizontal pipeline.
The rear part of the vehicle body 2 is provided with an odometry 4 as a travel distance measuring means for measuring travel distance and outputting travel distance information.
On the upper surface of the gantry 3 is an IMU 5 as an inertia measurement unit that measures the posture of the measuring vehicle 1 and outputs posture information;
First, second, and third three

laser measuring devices

6, 7, and 8 are fixed as inner surface shape measuring means for measuring the inner surface shape of the pipe and outputting inner surface shape information.
The three

laser measuring devices

6, 7, and 8 are configured to measure the inner surface shape information of the pipeline by scanning in three planes that are not parallel to each other (specifically, orthogonal to each other). .
It should be noted that other posture measuring means can be used instead of the IMU 5 as the inertial measuring means.

As shown in FIG. 3, the odometry 4 includes a mechanism that supports the device in an oblique manner while swinging the support boom 41 so as to be able to contact the inner surface of the tube at a position slightly swung from the tube bottom position just below.
Since there is a high probability that sewage flows or deposits exist near the bottom of the tube, the mechanism that supports it diagonally avoids such sewage and deposits and makes contact with the inner surface of the tube, making it more accurate. Distance measurement is possible.
The IMU 5 includes a three-axis gyro and a three-direction accelerometer, and requires a three-dimensional angular velocity and acceleration. By integrating the three-dimensional angular velocity information, the direction (posture) of the measurement vehicle can be obtained, and the direction of gravity can be obtained based on the acceleration information.
By accumulating the travel distance by the odometry 4 and the travel direction by the IMU 5, it is possible to sequentially obtain the self-position information of the measurement vehicle based on the measurement start position, and further obtain a travel locus from the measurement start position. . This process is referred to as sequential position estimation calculation (A). Details of the sequential position estimation calculation (A) will be described later.
The odometry 4 and the IMU 5 constitute a self-position measuring means for the measuring vehicle.

As shown in FIG. 4, the three

laser measuring devices

6, 7, and 8 and the IMU 5 are fixed to the gantry 3 disposed on the measuring vehicle 1 together with the camera 9 for taking an in-tube image.
The gantry 3 can be adjusted horizontally so that the scanning axis of the first laser measuring device 6 and the tube axis substantially coincide with each other by moving up and down horizontally by a pantograph mechanism 31 by a handle operation. (See FIGS. 2 and 3.)

Next, the arrangement of the

laser measuring devices

6, 7, and 8 will be described with reference to FIGS.
The first laser measuring device 6 scans the plane perpendicular to the pipe axis direction (plane B, XZ plane in FIG. 5) and measures the sequential position of each point on the inner surface of the pipe in the cross section of the pipe. Is configured to do. In addition, the said tube axis direction is a scanning-axis direction of the laser beam in the 1st laser measuring device 6 in an Example.
Measurement data obtained from the first laser measurement device 6 is referred to as cross-section measurement sensor data.

The second laser measuring device 7 scans a plane (plane A, XY plane in FIG. 5) defined by a tube axis direction component and a horizontal direction component orthogonal to the tube axis direction component, A sequential position of each point on the inner surface of the pipe in a cross section in the pipe axis direction of the pipe is measured.
Measurement data obtained from the second laser measurement device 7 is referred to as horizontal measurement sensor data.

The third laser measuring device 8 scans the vertical plane (plane C, YZ plane in FIG. 5) including the component in the tube axis direction, and each of the inner surface of the tube in the longitudinal section in the tube axis direction of the tube. It is comprised so that the sequential position of a point may be measured.
Measurement data obtained from the third laser measurement device 8 is referred to as vertical measurement sensor data.

As shown in FIG. 4A, the gantry 3 has a notch 61 for ensuring a wide scanning range of the first laser measuring device 6 and a wide scanning range of the third laser measuring device 8. A notch 81 for securing is provided.
As shown in FIG. 5, the scanning planes of the three

laser measuring devices

6, 7, and 8 are three planes that are orthogonal to each other. However, if the planes are not parallel to each other and intersect, the calculation is complicated. Although it does, it does not need to be orthogonal.

The first to third

laser measuring devices

6, 7, 8 measure the sequential position of each point on the inner surface of the pipe with reference to a specific reference point of the measuring vehicle 1, thereby 3 Measure the dimensional shape.

Next, a procedure for measuring the pipe line P illustrated in FIG. 1 using the measurement vehicle 1 will be described.
First, the initial position and orientation (direction) of the measuring vehicle 1 arranged at the bottom of the manhole m1 that is the measurement start point are measured. This is determined by measuring using a position specifying means such as a GPS device or a total station installed on the ground surface.
In such an initial position determination operation, the measuring vehicle 1 is present at the bottom of the manhole or underground such as in a pipe, so it is difficult to directly determine the initial position with a GPS device using GPS radio waves. It is also necessary to determine the posture (direction, direction) of the measurement vehicle in the initial state. Various specific procedures are conceivable as in Examples 1 to 4 described later.

After determining the initial position and posture of the measurement vehicle 1 by any procedure, the measurement vehicle 1 is moved and measurement of the pipeline P is started.
While moving in the pipeline, by accumulating the travel distance by the odometry 4 and the travel direction by the IMU 5, the self-position information of the measurement vehicle 1 based on the measurement start position is calculated as a sequential position estimation calculation (A ).
Further, self-position information obtained by sequential position estimation calculation (A) using cross-sectional measurement sensor data, horizontal measurement sensor data, and vertical measurement sensor data obtained by the three

laser measurement devices

6, 7, and 8. Are sequentially corrected by a sequential position correction calculation (B) described later to obtain corrected self-position information.

As described above, the self-position information based on the measurement start position is sequentially measured in the sequential position estimation calculation (A) while moving the measurement vehicle 1 from the measurement start position in the pipeline embedded therein, and further, Then, it goes to the measurement end position while sequentially correcting by the sequential position correction calculation (B).
When reaching the bottom of the manhole m2, which is the measurement end position, the end position and end posture (direction) of the measurement vehicle 1 are measured using a position specifying means such as a GPS device or a total station installed on the ground.
Based on the end position and orientation of the measuring vehicle 1 obtained in this way, all the self-position information and orientation information obtained by the sequential position correction calculation (B) are corrected using the Graph-SLAM algorithm. . This correction calculation processing is referred to as full correction Graph-SLAM (C).

Graph in the Graph-SLAM algorithm means a graphic graph, and the Graph-SLAM algorithm is one of three-dimensional SLAM solutions.
In the Graph-SLAM algorithm, every time the surroundings are measured with a measurement device (cart / sensor), a graph is created with the measurement device as a node (bundling), a link between them, and a link connection as an estimation error regarding the position. . Also, every time a landmark is observed, the graph and the measurement device and the landmark are nodes (units), and the observation between them is a link.
Add a subgraph whose link error is the observation error. This is a method of constructing a map of the position / posture of the measuring device and its surroundings by estimating the shape of the most applicable graph when the positions of some nodes in the completed graph are known.
In the case of the present invention, the landmark corresponds to the position coordinate (measurement start position information) of the carriage at the position of the manhole m1 and the position coordinate (measurement end position information) of the carriage at the position of the manhole m2. The link and the node correspond to the sequential position information of the measuring device between the manholes m1 and m2.

Hereinafter, the sequential position estimation calculation (A), the sequential position correction calculation (B), and the total correction calculation (C) will be described in detail with reference to FIG.
In step S1 shown in FIG.
The initial position (X0, Y0, Z0) and the initial posture (R0, P0, Ya0) of the measuring vehicle 1 are obtained by using position specifying means such as a GPS device installed in the manhole m1 at the measurement start position.
X0, Y0, and Z0 indicate the X coordinate component, Y coordinate component, and Z coordinate component of the position information, respectively, and R0, P0, and Ya0 indicate the Roll component, Pitch component, and Yaw component of the posture information, respectively.

In step S2, sequential position estimation calculation (A) is performed based on the travel distance information (x, y, z) by odometry 4 and the attitude information (R, P, Ya) by IMU 5, and in step S3, the next Self-position information (Xt, Yt, Zt) and attitude information (Rt, Pt, Yat) are obtained at the position of.
In step S4, the self-position information (Xt, Yt, Zt) and posture information (Rt, Pt, Yat) are sequentially used by using cross-sectional measurement sensor data, horizontal measurement sensor data, and vertical measurement sensor data. Correction calculation (B) is performed, and in step S5, corrected self-position information (X′t, Y′t, Z′t) and posture information (R′t, P′t, Ya′t) are obtained.
The process from step S2 to step S5 is repeated at a predetermined time interval, and the process proceeds to the measurement end position.

When it reaches the manhole m2 at the measurement end position,
In step S6, the end position (Xe, Ye, Ze) and the end position (Re, Pe, Yae) of the measuring vehicle 1 are obtained using position specifying means such as a GPS device installed in the manhole m2 at the measurement end position.
Thereafter, in step S7, all the corrected self-position information (X′t, Y′t, Z′t) and posture information (R′t, P ′) obtained in steps S1 to S5 are obtained. (t, Ya't) is corrected based on the end position (Xe, Ye, Ze) and the end position (Re, Pe, Yae) of the measuring vehicle 1.
This correction is the total correction Graph-SLAM (C).

In step S8, all the corrected self-position information (X′t, Y′t, Z′t) and postures from the measurement start position to the measurement end position by the above-described all correction Graph-SLAM (C). All self-position information (X, Y, Z) and posture information (R, P, Ya) from the measurement start position to the measurement end position by correcting the information (R't, P't, Ya't) Confirm.

In step S9, the self-position information (overall position estimation information) of the measurement vehicle 1 from the measurement start position to the measurement end position is determined over the entire route as described above.
Based on the overall position estimation information of the route traveled by the measuring vehicle and the inner surface shape information of the pipe line by the inner surface shape measuring means, it is possible to estimate the position of the tube inner surface at the sequential position of the measuring vehicle 1, The geographical coordinates of the pipeline from the measurement start position to the measurement end position can be determined.

Furthermore, in step S10, unevenness information in the pipe and meandering information of the pipe line can be obtained as follows.
For separation of meandering and unevenness, a first laser measuring device 6 for measuring a tube cross section and a third laser measuring device 8 for measuring a vertical cross section parallel to the tube axis are used. In FIG. 13, the measurement status of the tube cross section by the first laser measurement device 6 is indicated by an ellipse, and the measurement status of the vertical cross section by the third laser measurement device 8 is indicated by an arrow line.
In the pipe, the ceiling and the floor are normally parallel as shown in the area A of FIG. 13, and therefore the inclination (posture) of the measuring wheel 1 and the third laser measuring device 8 as shown in the area B of FIG. 13. If the inclination of the ceiling in the pipe measured by (1) does not match, it can be determined that it is uneven. In this case, the inclination of the ceiling and floor does not match.
Moreover, when the inclination of the measuring wheel 1 and the inclination of the ceiling and the floor coincide with each other as in the region C of FIG. 13, it can be determined that the pipe is meandering and not uneven.

In addition, the step S1 is a first step described in the claims,
The steps S2 and S3 are the second steps described in the claims,
The steps S4 and S5 are the third step described in the claims,
The step S6 is a fourth step described in the claims,
Steps S7 and S8 correspond to the fifth step recited in the claims.
The step S9 corresponds to the processing in the pipeline geographical coordinate estimation means described in the claims.

As described above, if the geographical coordinates in the global coordinate system can be obtained at the measurement start position and the measurement end position using a position specifying means such as a GPS device or a total station, the buried pipe from the measurement start position to the measurement end position All the positions can be obtained.
Furthermore, since meandering and unevenness can be separated in the middle of the pipeline, the geographical coordinates can be obtained sufficiently accurately even in buried pipelines such as sewage pipes that have actually been used for a long period of time. Therefore, it is possible to accurately determine the location of existing buried pipes such as sewage pipes without performing open-cut work and manual surveys, which can reduce the measurement costs, and future aging social infrastructure Stock development can be promoted without delay.

Note that some or all of the sequential position estimation calculation (A), the sequential position correction calculation (B), and the total correction calculation (C) are performed by a personal computer (portable type) mounted on the measuring vehicle 1. It may be processed by a small computer. In addition, a computer installed outside by transmitting a part or all of measurement information by the odometry 4, the IMU 5, and the three

laser measuring devices

6, 7, 8 to the outside at any time via a wired or wireless communication means. May be used to handle accumulation and computation. As a communication means for that purpose, a wireless communication means using radio waves or infrared rays, or a wired communication means such as a protected communication cable or optical fiber can be used. The protected communication cable or optical fiber may be laid along a lobe that pulls the measuring vehicle 1. In the above, the personal computer mounted on the measuring vehicle 1 or the computer installed outside is the sequential position estimating means, sequential position correcting means, all correction calculating means, and pipeline geographical coordinate estimating means described in the claims. It corresponds to.

Next, details of the sequential position estimation calculation (A) will be described.
Position x _t of the measuring vehicle at time t is obtained by the following equation.

x is the position of the measuring vehicle, θ is the posture of the measuring vehicle, and u is the minute change amount.

Δx _t : Measurement result of odometry 4 at time t θ _r : Rotation angle around roll axis θ _p : Rotation angle around pitch axis θ _y : Rotation angle around yaw axis

The angle update formula is shown below.

The position update formula is shown below.

R is a rotation matrix around each axis as shown below.

Next, details of the sequential position correction calculation (B) will be described.
The sequential position correction calculation (B) is performed by the cross-section measurement sensor data obtained by the first laser measurement device 6, the horizontal measurement sensor data obtained by the second laser measurement device 7, and the third laser measurement device 8. The correction is performed using the obtained vertical measurement sensor data, and it is considered that the error is large only by the position estimation by the odometry 4. Therefore, the position is sequentially corrected by using the three

laser measuring devices

6, 7, and 8. It is.

The position by the odometry 4 at time t is x _t , and the measured value of the tube shape by the laser measuring device is z _t . h (x _t ) is an estimated value of the in-pipe shape at time t based on minute position movement measured by odometry and IMU from the position at the previous time.
Since z _t and h (x _t ) do not match due to an error, the position x _t is corrected by adding the difference multiplied by a constant K.

K is a Kalman gain, which is expressed as a ratio of covariance matrices Σ and Ο (position data error magnitude).

z: Value measured by laser measuring device
h (x): Predicted value of measurement H: Jacobian matrix Σ: Covariance matrix of motion model Ο: Covariance matrix of measurement

Next, details of the total correction Graph-SLAM (C) will be described.
Graph-SLAM is used to correct the measurement end position at the manhole m2 position.
The x that minimizes J _Graph-SLAM shown by the following equation is the most probable estimation result.
Since g (u _t , x _tt ) is a function that estimates the current position from the previous position and the minute movement amount, x _t -g (u _t , x _tt ) is the difference between the estimated positions. [X _t -g (u _t , x _tt )] ^T R ^-1 [x _t -g (u _t , x _tt )] Is a constraint of the motion model.
h (m _c , x _t ) is a measurement function when the position x of the measurement vehicle is measured from the manhole position m. Σ [z _t -h (m _c , x _t )] ^T Q ^-1 [z _t -h] obtained by multiplying this and the actual measured value z by the covariance matrix Q (indicating the magnitude of the error) (m _c , x _t )] is a constraint condition of the measurement model. (See Figure 8.)
The sum of these constraints is J _Graph-SLAM shown below.

In measurement using the IMU, the vector having the largest tilt error is Yaw.
Roll and pitch can be corrected by detecting the direction of gravity with IMU.
Yaw needs to detect geomagnetism for error correction, but it cannot use Yaw's automatic error correction because geomagnetism cannot be used in buried pipes. However, according to the present invention, Yaw's automatic error correction can be performed by using the correction by the three laser measuring devices in combination and performing the sequential correction calculation (B) in the pipeline. Furthermore, by correcting all the geographical coordinates by the total correction Graph-SLAM (C) based on the position information by GPS or the like at the measurement end position, it becomes possible to specify the geographical coordinates of the pipeline more accurately.

Below, the Example of the position measurement method in the manhole m1, m2 of a measurement start position and a measurement end position is described.
From two different coordinate positions (Xm0, Ym0, Zm0) and (Xm1, Ym1, Zm1) of the measuring vehicle 1 measured by a global positioning system such as GPS or a total station, one coordinate position and direction (X0, Y0, Z0, Ya0) is obtained by calculation, and the orientation (R0, P0) other than the direction is obtained by the IMU 5, and one coordinate position and orientation X0, Y0, Z0, R0, P0, Identify Ya0).

First, a first specific method for specifying the coordinate position of the measurement vehicle 1 at the measurement start position and the measurement end position will be described with reference to FIG.
As shown in FIG. 9, the total station TS is disposed in the upper opening of the manhole m1 at the measurement start position, and the measurement vehicle 1 is disposed on the bottom surface of the manhole m1 directly below the total station TS. The bottom of the manhole m1 is connected to the pipe P to be measured.
The grid plate 11 is installed in the measuring wheel 1, the laser centripetal device LC is attached to the total station TS, and the ground position of the total station TS is adjusted so that the centripetal laser is irradiated to the grid plate 11.

The ground coordinates are obtained by measurement at the total station TS at that position. Also, the position on the grid plate 11 where the centripetal laser is irradiated is recorded.
The distance from the total station TS to the grid plate 11 is measured using a distance measuring means such as a measure.
The above method is also performed at another position on the grid plate 11, and two different coordinate positions of the measuring wheel 1 are specified. The posture of the measuring vehicle 1 is specified from the two coordinate positions.
As described above, the position and orientation of the measurement vehicle 1 at the measurement start position are determined.
The same operation is performed in the manhole m2 at the measurement end position.
In the above, the total station TS, the laser centripetal device LC, the distance measuring means, and the grid plate 11 correspond to the start position specifying means or the end position specifying means described in the claims.

In the above case, the measuring vehicle 1 can be seen directly from the ground, but the manhole m1 may be just a space for connection to the pipe as shown in FIG.
In that case, the following second specific method can be used.
A cube 12 is installed directly under the manhole.
From the ground, the position of the cube 12 is measured using a total station or the like in the same manner as described above.
Further, from the measuring wheel 1, the position shape of the cube 12 is measured using the

laser measuring devices

7, 8 and 9.
Thereby, the position and posture of the measurement vehicle 1 at the measurement start position are determined.
The same operation is performed in the manhole m2 at the measurement end position.
In the above, the total station, the cube 12, and the

laser measuring devices

7, 8, and 9 correspond to the start position specifying means or the end position specifying means described in the claims.

In the third specific method, as shown in FIG. 11, the LED lamps 13 are installed in four places on the measuring vehicle 1, and an appliance with a camera CM that can be installed instead of a manhole cover is installed.
The position of the camera CM is measured on the ground using a total station or the like.
The position and orientation of the measuring vehicle 1 are specified by photogrammetry.
The same operation is performed in the manhole m2 at the measurement end position.
In the above, the total station, the camera CM, and the LED lamp 13 correspond to the start position specifying means or the end position specifying means described in the claims.

As shown in FIG. 12, the fourth specific method is to install the grid plate 14 on the measuring wheel 1 and specify the grid plate 14 with the handy measuring instruments H1, H2, and H3 from three locations on the ground. Measure the distance to the location.
The coordinate positions of the handy measuring instruments H1, H2, and H3 are measured on the ground using a total station or the like.
Thereby, the position and orientation of the measurement vehicle 1 are specified.
The same operation is performed in the manhole m2 at the measurement end position.
Although there are various methods for specifying the position of the measuring vehicle 1, the above four specific methods are exemplified.
The manhole m1 at the measurement start position and the manhole 2 at the measurement end position are not limited to the same method, and different methods may be used.
In the above, the total station, the handy measuring instruments H1, H2, H3, and the grid plate 14 correspond to the start position specifying means or the end position specifying means described in the claims.

As the travel distance measuring means, various distance measuring means can be used as long as the travel distance is measured instead of the odometry using the wheels.
As the attitude measurement means or the inertia measurement means, a technique such as an attitude / direction reference apparatus (AHRS) using geomagnetism can be used instead of the inertia measurement apparatus (IMU).
An inertial measurement device (IMU) is preferred because geomagnetism may not be available in the buried pipeline.
The inner surface shape measuring unit is not limited to the laser measuring device, and various distance measuring devices that can measure the distance from the inner surface measuring unit to the inner wall surface of the tube in a non-contact manner can be used.
For example, a distance measuring device using a sufficiently high-precision millimeter wave radar can be used.

Moreover, although the example which mounted the three

laser measuring devices

6, 7, and 8 was demonstrated as an inner surface shape measurement means, it is not limited to three, Inner surface shape information is used using one or more laser measuring devices. It is possible to obtain
If the number of laser measuring devices is increased, more accurate measurement can be performed.
If the inner diameter of the tube is known to some extent, one laser measuring device can be used.
It is also possible to use a three-dimensional measuring apparatus using a light cutting method that obtains a three-dimensional position of an object by projecting a thin sheet of light based on the principle of triangulation.
Although it is expensive at the present time, it is also possible to configure the inner surface shape measuring means of the tube with a single three-dimensional scanner device using a three-dimensional scanner device.

m1 Manhole at the measurement start position m2 Manhole at the measurement end position Pipe 1 to be measured 1 Measurement vehicle 2 Car body 3 Mount 31 Pantograph mechanism 4 Odometry, mileage measurement means 5 IMU, inertia measurement means 6 First laser measurement device, inner surface Shape measuring means 61 Incision 7 Second laser measuring device, inner surface shape measuring means 8 Third laser measuring device, inner surface shape measuring means 81 Notch 9 Camera 11 Square plate 12 Cube 13 LED lamp 14 Square plate TS Total station, start position specification Means, end position specifying means CM camera H1, H2, H3 Handy measuring instrument

Claims

Travel distance measurement means for measuring travel distance and outputting travel distance information, attitude measurement means or inertia measurement means for measuring the attitude of a measuring vehicle and outputting posture information, and measuring the inner surface shape of a pipe to measure the inner surface A measuring vehicle capable of traveling in a pipeline with an inner surface shape measuring means for outputting shape information;
Start position specifying means for specifying the position and orientation of the measurement vehicle at the measurement start position and outputting measurement start position information;
An end position specifying means for specifying the position and orientation of the measurement vehicle at the measurement end position and outputting the measurement end position information;
While traveling in the pipeline, using the travel distance information by the travel distance measurement means and the attitude information by the attitude measurement means or the inertia measurement means, the sequential position information of the measurement vehicle based on the measurement start position is obtained. Sequential position estimation calculation means for performing sequential estimation calculation;
Based on the inner surface shape information of the pipeline by the inner surface shape measuring means, sequential position correction calculating means for sequentially correcting and calculating the sequential position information of the measuring vehicle while traveling in the pipeline;
After reaching the measurement end position, a three-dimensional SLAM that performs position estimation and map creation in parallel on the basis of the measurement start position information and the measurement end position information based on the sequential position information that has been subjected to the successive correction calculation. A buried pipe measuring device comprising: all correction calculation means for performing total correction calculation processing by a (Simultaneous Localization And Mapping) algorithm and specifying overall position estimation information of a route traveled by the measurement vehicle.
Based on the overall position estimation information of the route traveled by the measurement vehicle obtained by the all correction calculation means, and the inner surface shape information of the pipeline by the inner surface shape measurement means,
The buried pipe measuring device according to claim 1, further comprising pipe geographical coordinate estimating means for estimating a position of an inner surface of the pipe and estimating geographical coordinates of the pipe.
The buried pipe measuring device according to any one of claims 1 and 2, wherein the inner surface shape measuring means uses a laser measuring device that scans the inner surface of the pipe.
The inner surface shape measuring means includes
The buried pipe measuring device according to any one of claims 1 to 3, wherein the inner pipe shape information is obtained by scanning in three non-parallel planes. .
The inner surface shape measuring means includes
The buried pipeline measurement according to any one of claims 1 to 4, wherein the measurement is at least a cross-sectional shape of the pipeline by scanning in a plane perpendicular to the tube axis. apparatus.
The inner surface shape measuring means includes
It is configured to scan a plane specified by a tube axis direction component and a horizontal direction component orthogonal to the tube axis, and measure at least the cross-sectional shape of the pipe in the tube axis direction. The buried pipe measuring device according to any one of claims 1 to 5.
The inner surface shape measuring means includes
7. The apparatus according to claim 1, wherein a vertical plane including a component in the tube axis direction is scanned to measure at least a longitudinal sectional shape of the tube in the tube axis direction. The buried pipe measuring device described.
The buried pipe measuring device according to any one of claims 1 to 7, further comprising an elevating mechanism for elevating and lowering a gantry on which the inner surface shape measuring means is disposed.
The sequential position estimation calculation means includes
At each predetermined time interval, using the travel distance information by the travel distance measurement means and the posture information by the attitude measurement means or the inertia measurement means at the predetermined time, the measurement vehicle based on the measurement start position is used. The buried pipe measuring device according to any one of claims 1 to 8, wherein the position information is sequentially updated.
The sequential position correction calculation means includes:
Based on the estimated inner surface shape information based on the sequential position information of the measuring vehicle by the sequential position estimation calculation means and the inner surface shape information by the inner surface shape measurement means, at each predetermined time interval, The buried pipe measuring device according to any one of claims 1 to 9, characterized in that a sequential correction calculation process is performed.
The total correction calculation process by the three-dimensional SLAM algorithm is as follows:
In the three-dimensional space, the sequential position information that has been subjected to the sequential correction calculation is subjected to all correction calculation processing based on the measurement start position information and the measurement end position information, so that position estimation in the three-dimensional space and map creation are performed in parallel. The buried pipe measuring device according to any one of claims 1 to 10, wherein the measuring device is executed.
As the three-dimensional SLAM algorithm,
With the specified measurement start position and the specified measurement end position as landmarks,
The sequential position information of the measurement vehicle between the measurement start position and the specified measurement end position,
With links and nodes between them,
Create a graph with measurement error of link strength,
12. The embedded pipeline according to claim 11, wherein a graph-SLAM algorithm that executes map creation in parallel with a position / posture of the measurement vehicle is used by estimating a shape most applicable to the graph. Measuring device.
Travel distance measurement means for measuring travel distance and outputting travel distance information, attitude measurement means or inertia measurement means for measuring the attitude of a measuring vehicle and outputting posture information, and measuring the inner surface shape of a pipe to measure the inner surface A method for measuring a buried pipeline using a measuring vehicle capable of traveling in a pipeline with an inner surface shape measuring means for outputting shape information,
A first step of obtaining measurement start position information by specifying the position and orientation of the measurement vehicle at the measurement start position;
While traveling in the pipeline, using the travel distance information by the travel distance measurement means and the attitude information by the attitude measurement means or the inertia measurement means, the sequential position information of the measurement vehicle based on the measurement start position is obtained. A second step of performing successive estimation operations;
A third step of successively correcting and calculating the sequential position information of the measuring vehicle while traveling in the pipeline based on the inner surface shape information of the pipeline by the inner surface shape measuring means;
A fourth step of obtaining the measurement end position information by specifying the position and orientation of the measurement vehicle at the measurement end position after reaching the measurement end position;
After reaching the measurement end position, a three-dimensional SLAM that performs position estimation and map creation in parallel on the basis of the measurement start position information and the measurement end position information based on the sequential position information that has been subjected to the successive correction calculation. And a fifth step of identifying all position estimation information of a route traveled by the measurement vehicle by performing a total correction calculation process using a (Simultaneous Localization And Mapping) algorithm. .