WO2022024259A1

WO2022024259A1 - Data correction method and data correction device

Info

Publication number: WO2022024259A1
Application number: PCT/JP2020/029047
Authority: WO
Inventors: 秀幸坪井; 和人後藤; 秀紀俊長; 直樹北; 武鬼沢
Original assignee: 日本電信電話株式会社
Priority date: 2020-07-29
Filing date: 2020-07-29
Publication date: 2022-02-03
Also published as: JPWO2022024259A1; JP7339588B2

Abstract

In the present invention: horizontal plane position data that is more accurate than horizontal plane position data indicating the position of a moving body on a horizontal plane obtained at each observation time is acquired, or vertical plane position data indicating the position of the moving body on a vertical plane is acquired; movement trajectory data indicating the movement trajectory of the moving body is generated on the basis of the acquired highly accurate horizontal plane position data or the acquired vertical plane position data; the movement trajectory data is analyzed and the position and the movement state of the moving body for each measurement time is estimated; measurement condition data indicating the measurement condition for a laser radar device for each of the measurement times is generated on the basis of the position and the movement state of the moving body for each of the measurement times; and collected data is corrected on the basis of the measurement condition data.

Description

Data correction method and data correction device

The present invention relates to a data correction method and a data correction device for correcting data related to measurement conditions when measuring point cloud data.

Figure 16 shows a use case (for example, non-use case) proposed by mmWave Networks in TIP (Telecom InfraProjcet) (main members: Facebook, Deutsche Telecom, Intel, NOKIA, etc.), which is a consortium that promotes open specifications of communication network equipment in general. It is a diagram which has been partially modified and schematicized with reference to Patent Documents 1 to 3). mmWave Networks is one of the TIP project groups, aiming to build networks faster and cheaper than laying optical fibers using unlicensed millimeter-wave radio.

In buildings such as

buildings

800, 801 and

houses

810, 811, 812 shown in FIG. 16, terminal station devices (hereinafter referred to as "terminal stations") 840 to 844 and electric poles 821 to 826 installed on the respective wall surfaces of the building. The installed base station devices (hereinafter referred to as "base stations") 830 to 834 are devices called mmWave DN (Distribution Node).

The base stations 830 to 834 are connected to the communication device provided in the station building (Fiber PoP (Point of Presence)) 850, 851 by

optical fibers

900, 901. This communication device is connected to the communication network of the provider. Between the terminal stations 840 to 844 and the base stations 830 to 834 (hereinafter, also referred to as "between both stations"), mmWaveLink, that is, millimeter wave radio is performed. In FIG. 16, the millimeter-wave radio link is shown by a dashed line.

Base stations 830 to 834 are installed on utility poles 821 to 826, terminal stations 840 to 844 are installed on the wall of a building, and base stations 830 to 834 and terminal stations 840 to communicate in a form in which both stations are communicated by millimeter wave radio. Selecting a candidate position for installing the 844 is called station design (hereinafter also referred to as "station design").

FIG. 17 shows the schematic configuration of the MMS (Mobile Mapping System) that collects the point cloud data used for the station placement design of the millimeter wave station, the types of data collected by the MMS, and the station placement support tool that uses the point cloud data. It is a figure which showed the displayed screen 500. The MMS is, for example, a vehicle 700 including a laser radar 702, an antenna 703 that receives radio waves from a GPS (Global Positioning System), and a measuring device 701. While the vehicle 700 is traveling, the laser radar 702 irradiates the surroundings with laser light at each measurement time. The laser radar 702 measures the time until the reflected light returns from the measurement target such as a building existing in the surroundings, and calculates the distance to the measurement target based on the measured time. The measuring device 701 generates point cloud data by recording the distance to the object to be measured at each measurement time and the irradiation direction of the laser beam.

The measuring device 701 receives radio waves from a plurality of GPS satellites 600-1 and 600-2 at each observation time. The measuring device 701 detects the position of the vehicle 700 for each observation time based on the data superimposed on the received radio wave. The measuring device 701 generates the traveling locus data 711 from the history of the position of the vehicle 700 for each detected observation time.

Since the observation time interval is longer than the measurement time interval, it is unlikely that the measurement time and the observation time will match. Therefore, first, the measuring device 701 estimates the position, speed, and traveling direction of the vehicle 700 for each measurement time from the generated travel locus data 711. Next, the measuring device 701 calculates the irradiation position of the laser radar 702 in the three-dimensional space from the estimated position of the vehicle 700. Then, the measuring device 701 calculates the direction in which the laser radar 702 is facing from the traveling direction of the vehicle 700. Based on the detected irradiation position of the laser radar 702, the direction in which the laser radar 702 is facing, and the point cloud data, the coordinates of each point included in the point cloud data in the three-dimensional space can be specified. ..

When the measuring device 701 determines that the vehicle 700 is traveling at a constant speed and is traveling straight based on the estimated speed and traveling direction of the vehicle 700 at each measurement time, the point group data at the measurement time is determined. A reliability index "high" indicating that the degree of reliability is high is given to the data, and the data is classified as point group data (type A) 712. On the other hand, when the measuring device 701 determines that the vehicle 700 is not traveling at a constant speed or is not traveling straight based on the estimated speed and traveling direction of the vehicle 700 at each measurement time, the measurement is performed. A reliability index "low" indicating that the degree of reliability is low is given to the point group data at the time, and the point group data (type B) 713 is classified.

When the vehicle 700 is not traveling at a constant speed, for example, it is accelerating or decelerating, and when the vehicle 700 is not traveling straight, for example, it is traveling in a curved manner including a right / left turn. If there is. The reason why the reliability index "low" is given is that the position, speed, and estimation accuracy of the traveling direction of the vehicle 700 are lowered when the vehicle is accelerating, decelerating, or traveling in a curved manner including right / left turn. .. If the point cloud data is not reliable and cannot be used instead of the reliability index "low", the reliability index "not" may be given.

Here, the reliability index is an index showing the priority of use when using the point cloud data in the station support tool. For example, the point cloud data to which "none" is given is not used, and when the point cloud data to which "low" is given is used, "high" indicating the same position is given. It is used when other point cloud data does not exist, and processing is performed so that the point cloud data to which "high" is given is preferentially used.

The station station support tool uses the point cloud data to detect the presence or absence of a line-of-sight between the position where the base station 501 shown on the screen 500 is installed and the position where each of the terminal stations 511 to 514 is installed. For example, the station support tool sets a line-of-sight detection line between the position of the base station 501 and the position where each of the terminal stations 511 to 514 is installed, and point cloud data indicating the shape of the building is displayed on the line-of-sight detection line. The presence or absence of a line of sight is detected depending on whether or not it exists.

The station support tool displays map data on the screen 500. For example, the area surrounded by the solid line indicated by the

reference numerals

521 and 522 is a section in which a building or the like shown in the map data exists, and the blank space between the sections is a road. The station station support tool displays the travel locus 540 of the vehicle 700 on the map data based on the travel locus data 711. Along the travel locus 540, for example, a region having a certain width along the solid arrow indicated by reference numerals 531,532 (hereinafter referred to as “region A”) is a place where the point cloud data (type A) 712 exists. .. Along the travel locus 540, for example, a region having a certain width along the dotted line indicated by reference numeral 533, 534 (hereinafter referred to as “region B”) is a place where the point cloud data (type B) 713 exists.

The screen 500 shows a portion of the pattern of the plurality of areas A and a portion of the pattern of the plurality of areas B. In the pattern portion of the region A, the locus indicated by the traveling locus data 711 is a straight line. On the other hand, it can be seen that the locus indicated by the traveling locus data 711 is bent at the portion of the pattern in the region B.

As described above, the reason for assigning the confidence index "low" or "none" to the point cloud data classified as the point cloud data (type B) 713 shown in FIG. 17 is acceleration, deceleration, or right. This is because the position, speed, and estimation accuracy of the traveling direction of the vehicle 700 are lowered when the vehicle is traveling in a curved manner including a left turn. Specific examples of this reason are shown in FIGS. 18 and 19.

FIG. 18 is a view of the road 550 and the sidewalk 570, which are in the vicinity of the road in one direction of the straight two-lane road on one side, from the sky. The line indicated by reference numeral 552 is a center line indicating a boundary with an oncoming lane.

The double-lined arrow indicated by the reference numeral 541 is a locus on which the vehicle 700, which is an MMS, has traveled, and is hereinafter referred to as a travel locus 541. The vehicle 700 is traveling in the right direction indicated by the arrow of the traveling locus 541. On the road 550, the vehicle 750 is stopped near the sidewalk 570. In the vehicle 700, in order to avoid the stopped vehicle 750 while traveling in the traveling lane 561, the overtaking lane 562 temporarily crosses the line 551 that separates the traveling lane 561 and the overtaking lane 562. A driving operation is performed in which the vehicle travels and then returns to the traveling lane 561.

Since such a driving operation is performed, the traveling locus 541 of the vehicle 700 is in a bent state at two places. The plurality of dotted lines shown along the traveling locus 541 are laser irradiation lines indicating the irradiation direction of the laser light emitted by the laser radar 702 mounted on the vehicle 700 at each measurement time. When the vehicle 700 is traveling in a straight line at a constant speed in the traveling lane 561 and the overtaking lane 562, that is, when the vehicle is traveling straight at a constant speed, the laser irradiation line is perpendicular to the traveling direction. It becomes a regular arrangement parallel to each other at equal intervals.

On the other hand, in the section indicated by the reference numerals 401, 402 in which the vehicle 700 moves from the traveling lane 561 to the overtaking lane 562 and returns from the overtaking lane 562 to the traveling lane 561, the laser irradiation lines are not parallel but disturbed. Will occur, and if the speed is not constant, it will not be evenly spaced.

FIG. 19 is a diagram showing a travel locus 542 in which a vehicle 700, which is an MMS, travels at an intersection where roads with one lane on each side intersect. Similar to FIG. 18, in FIG. 19, the laser irradiation line for each measurement time is shown by a dotted line. The lines indicated by reference numerals 580-1 and 580-2 are center lines, and the lines indicated by reference numerals 581-1 to 581-4 are lines indicating the boundary between the roadway and the roadside zone. In the vehicle 700, as indicated by the arrow of the traveling locus 542, the vehicle travels upward from the lower part of FIG. 19, approaches the center line 580-1 side before turning right at the intersection, and then turns right at the intersection and goes straight to the right. The driving operation is being performed.

Due to such a driving operation, the traveling locus 542 of the vehicle 700 is not parallel but turbulent in the section 403 approaching the center line 580-1 side and in the section 404 turning right at the intersection. If it occurs and becomes irregular, and if it is not constant velocity, it will not be evenly spaced. After making a right turn, the vehicle 700 is traveling straight at a constant speed, so that the laser irradiation lines are perpendicular to the traveling direction and are arranged in a regular arrangement parallel to each other at equal intervals.

As shown in FIGS. 18 and 19, the laser irradiation lines are not parallel when the traveling lane is changed, or when the vehicle does not go straight, such as when turning right or left at an intersection, and when the speed is not constant, the laser irradiation lines are not evenly spaced. When the laser irradiation line is irregularly disturbed in this way, it is difficult or impossible to estimate the shape and size of the space in which the individual points included in the point cloud data exist. Therefore, for the point cloud data when the laser irradiation line is irregularly disturbed in this way, the reliability level is lowered, so that the reliability index "low" or the reliability index "none" is given.

Such a phenomenon that the laser irradiation line is disturbed is not limited to the horizontal plane shown in FIGS. 18 and 19, and for example, the vehicle 700 climbs a slope from a flat ground, returns from a slope to a flat ground, or has irregularities. It also occurs when driving on the road.

In view of the above circumstances, an object of the present invention is to provide a technique capable of making available point cloud data whose reliability is lowered due to disturbance of a laser irradiation line.

One aspect of the present invention is the point group when the laser radar device mounted on the moving body measures the distance to the object to be measured by irradiating the laser light at each measurement time and generates point group data. The collected data generated in association with the data and indicating the measurement conditions of the laser radar device estimated from the horizontal plane position data indicating the position of the moving object on the horizontal plane obtained at each observation time is corrected. This is a data correction method for acquiring horizontal plane position data with higher accuracy than the horizontal plane position data obtained at each observation time, or vertical plane position data indicating the position of the moving body on the vertical plane. The movement trajectory data generation data generation step and the movement For each measurement time, based on the movement locus analysis step that analyzes the locus data and estimates the position and movement state of the moving body at each measurement time, and the position and movement state of the moving body at each measurement time. It is a data correction method including a measurement condition generation step for generating measurement condition data indicating the measurement conditions of the laser radar device, and a correction processing step for correcting the collected data based on the measurement condition data.

One aspect of the present invention is the point group when the laser radar device mounted on the moving body measures the distance to the object to be measured by irradiating the laser light at each measurement time and generates point group data. The collected data generated in association with the data and indicating the measurement conditions of the laser radar device estimated from the horizontal plane position data indicating the position of the moving object on the horizontal plane obtained at each observation time is corrected. This is a data correction device that acquires horizontal plane position data with higher accuracy than the horizontal plane position data obtained at each observation time, or vertical plane position data indicating the position of the moving body on the vertical plane. The movement locus data generation unit that generates the movement locus data indicating the movement locus of the moving body based on the acquired high-precision horizontal plane position data or the vertical plane position data, and the movement locus data. The movement locus analysis unit that estimates the position and the moving state of the moving body at each measurement time, and the movement locus analysis unit that estimates the position and the moving state of the moving body at each measurement time, and the said at each measurement time based on the position and the moving state of the moving body at each measurement time. It is a data correction device including a measurement condition generation unit that generates measurement condition data indicating measurement conditions of a laser radar device, and a correction processing unit that corrects the collected data based on the measurement condition data.

According to the present invention, it is possible to make available point cloud data whose reliability is lowered due to the disturbance of the laser irradiation line.

It is a block diagram which shows the structure of the point cloud data collection system in 1st Embodiment. It is a figure which shows the data which the storage part of 1st Embodiment stores. It is a figure which showed the method of estimating the position and the traveling direction of a vehicle for each measurement time from the traveling locus data by the traveling locus analysis unit of 1st Embodiment. It is a figure which showed the method of obtaining the position and the traveling direction of a vehicle by applying the traveling state data to the position of the vehicle estimated in FIG. It is a figure for demonstrating the estimation accuracy of a laser irradiation direction when only GPS observation data is used. It is a figure for demonstrating the estimation accuracy of a laser irradiation direction when GPS observation data and quasi-zenith satellite observation data are used. It is a flowchart which shows the flow of the process by the collection data generation part of 1st Embodiment. It is a flowchart which shows the flow of the process which corrects the collected data in 1st Embodiment. It is a block diagram which shows the structure of the point cloud data collection system of 2nd Embodiment. It is a figure which shows the data which the storage part of the 2nd Embodiment stores. It is a figure which shows the change of the traveling locus of a vehicle when traveling on the uneven road. It is a figure which shows the change of the laser irradiation direction when the vehicle of the 2nd Embodiment travels on the road shown in FIG. It is a flowchart which shows the flow of the process which corrects the collected data in 2nd Embodiment. It is a block diagram which shows the structure of the point cloud data collection system of 3rd Embodiment. It is a figure which shows the change of the laser irradiation direction when the vehicle of the 3rd Embodiment travels on the road shown in FIG. It is a figure which shows an example of the use case proposed by TIP. It is a figure which shows the outline of the station design using the point cloud data. It is a figure which shows the change of the laser irradiation direction when the MMS travels in a bent manner. It is a figure which shows the change of the laser irradiation direction when MMS turns right.

(First Embodiment)
Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a point cloud data collection system α according to the first embodiment. The point cloud data collection system α includes a vehicle 1 equipped with a point cloud data collection device 2, a plurality of GPS satellites 10-1, 10-2, and a quasi-zenith satellite 11. The vehicle 1 is, for example, a moving body such as an automobile, and corresponds to the above-mentioned MMS.

In FIG. 1, two GPS satellites 10-1 and 10-2 are shown as an example, but in reality, more than two GPS satellites are in operation. In FIG. 1, one quasi-zenith satellite 11 is shown as an example, but in the future, a plurality of quasi-zenith satellites 11 are scheduled to be operated. Since the line-of-sight state changes depending on the position where the vehicle 1 is traveling, the number of GPS satellites 10-1, 10-2 and the quasi-zenith satellite 11 that can receive radio waves while the vehicle 1 is traveling changes.

In the first embodiment, in addition to the data superimposed on the radio waves received from the GPS satellites 10-1 and 10-2, the data superimposed on the radio waves received from the quasi-zenith satellite 11 is used. Here, the outline of the quasi-zenith satellite 11 will be described. In recent years, by using the QZSS (Quasi-Zenith Satellite System), which is mainly composed of the Quasi-Zenith Satellite 11 (hereinafter also referred to as "QZS" (Quasi-Zenith Satellite)), the accuracy is higher than when GPS is used. Attempts have been made to detect the position.

For example, page 7 of Non-Patent Document 4 shows an example of an experimental system configuration for a GPS complementary function effect. In this experimental system, a receiver corresponding to the quasi-zenith satellite 11 is mounted on the MMS, which is a mobile positioning device, and the reference MMS travels along with the reception logs of the quasi-zenith satellite 11 and GPS satellites 10-1 and 10-2. Collecting trajectories. In Non-Patent Document 4, the "complementary" function was realized by performing the positioning calculation by adding the information from the quasi-zenith satellite 11 to the GPS satellites 10-1 and 10-2 by the experimental system, from the quasi-zenith satellite 11. The verification of the improvement effect of the mobile body positioning performance in the urban area by the addition of the information of is shown.

On page 16 of Non-Patent Document 5, an example is shown in which a positioning terminal is mounted on an MMS, real-time mobile positioning is performed on a forest road, and a centimeter-class positioning reinforcement for use verification is conducted. Results with a FIX rate of 79% and a positioning accuracy of 2.4 cm were obtained, indicating that the results are good.

According to the "accuracy evaluation in urban areas" shown in FIG. 6 on page 6 of Non-Patent Document 6, at the point where the error of the traveling locus was (1) 1.28 m in the positioning by GPS alone, (2) QZS + GPS. It is shown that the positioning by is improved to 0.21 m. According to "Improvement of positioning performance by QZS" in Table 6 on page 6 of Non-Patent Document 6, the degree of improvement of positioning rate by adding QZS is 1.7 times and 3D (Three) when compared with positioning by GPS alone. -Dimensions) accuracy has been shown to increase from 169 cm to 45 cm.

The accuracy of positioning using only GPS satellites 10-1 and 10-2 is at most several meters to several tens of meters. On the other hand, when the GPS satellites 10-1 and 10-2 and the quasi-zenith satellite 11 are used, an accuracy of 2.4 cm can be obtained in a good environment for traveling on a forest road shown in Non-Patent Document 5. Even in a bad environment of driving on a road in a valley of buildings in an urban area shown in Non-Patent Document 6, the accuracy of positioning is about ten and several because the accuracy of 0.21 m is obtained. It will improve from centimeters to several centimeters.

Returning to FIG. 1, the point cloud data collecting device 2 includes a laser radar device 21, a satellite radio wave receiving antenna 22, an information receiving unit 23, a traveling state measuring unit 24, a storage unit 25, a collected data generation unit 26, and a data correction unit 3. To prepare for.

The laser radar device 21 can irradiate the laser beam by rotating the irradiation hole of the laser beam 360 ° around the rotation axis, for example, and the surface formed by the irradiated laser beam and the rotation axis are perpendicular to each other. It is configured to intersect with. Hereinafter, the direction in which the rotation axis of the laser radar device 21 faces is referred to as the direction of the laser radar device 21. The laser radar device 21 is fixedly mounted on the top of the vehicle 1 so that the surface formed by the irradiated laser beam is perpendicular to the traveling direction of the vehicle 1. Therefore, as shown in FIG. 1, the direction of the laser radar device 21 is the direction indicated by the arrow of reference numeral 4, and is 180 ° opposite to the traveling direction of the vehicle 1.

The laser radar device 21 irradiates the surroundings with laser light at each measurement time. The laser radar device 21 measures the time until the reflected light returns from the measurement target such as a building existing in the surroundings, and calculates the distance to the measurement target based on the measured time. The interval of the measurement time is a constant time, for example, about 0.005 seconds. The laser radar device 21 rotates at a rotation speed of, for example, about 200 [Hz] during the interval of the measurement time. The laser radar device 21 generates point cloud data 43 in association with the distance to the object to be measured at each measurement time and the irradiation direction of the laser beam, and writes the generated point cloud data 43 in the storage unit 25. Here, when the irradiation hole of the laser beam rotates clockwise, for example, in the direction of the laser radar device 21, the direction in which the irradiation angle of the laser beam "0 °" is directed toward the sky, that is, the direction directly above. Then, the irradiation angle of the laser beam increases clockwise, and the direction from the front surface to the back surface of FIG. 1, that is, the left direction with respect to the traveling direction of the vehicle 1 becomes "90 °". The direction toward the ground, that is, the direction directly below is "180 °", and the direction from the back surface to the front surface of FIG. 1, that is, the direction to the right of the traveling direction of the vehicle 1 is "270 °". The current high-precision laser radar device 21 can measure measurement positions of about 1 million points (1 million points / second) per second. For the sake of simplicity, in the present embodiment, the laser radar device 21 has 720,000 points per second (720,000 / sec (≈200 [Hz] × 360 [°] ÷ 0.1 [°]]. )) It shall be possible to measure the measurement position.

The satellite radio wave receiving antenna 22 receives radio waves from GPS satellites 10-1 and 10-2 and the quasi-zenith satellite 11. The information receiving unit 23 corresponds to a receiver or a positioning terminal of the GPS satellites 10-1 and 10-2 and the quasi-zenith satellite 11. The information receiving unit 23 detects the data superimposed on the radio waves received from the GPS satellites 10-1 and 10-2 by the satellite radio wave receiving antenna 22 at each observation time, and the vehicle 1 at the observation time from the detected data. The position of the GPS observation data 40 is calculated, and the calculated position of the vehicle 1 is associated with the observation time and written in the storage unit 25 as GPS observation data 40. The information receiving unit 23 detects the data superimposed on the radio waves received from the quasi-zenith satellite 11 by the satellite radio wave receiving antenna 22 at each observation time, and calculates the position of the vehicle 1 at the observation time from the detected data. , The calculated position of the vehicle 1 and the observation time are associated with each other and written in the storage unit 25 as quasi-zenith satellite observation data 41. The interval of the observation time for receiving radio waves from the GPS satellites 10-1 and 10-2 and the quasi-zenith satellite 11 is a fixed time, for example, 1 second.

The traveling state measuring unit 24 is connected to, for example, a vehicle speed sensor that detects the speed from the rotation speed of the tire of the vehicle 1 and a steering sensor that detects the steering angle of the steering of the vehicle 1. The vehicle speed sensor and the steering sensor are provided inside the vehicle 1. The traveling state measuring unit 24 measures the speed of the vehicle 1 at each measurement time based on the data obtained from the vehicle speed sensor. The traveling state measuring unit 24 measures the steering angle of the steering at each measurement time based on the data obtained from the steering sensor. The traveling state measuring unit 24 calculates the traveling direction of the vehicle 1 on the horizontal plane for each measurement time from the steering angle of the steering.

The traveling state measuring unit 24 includes a traveling state data 42-1 indicating the speed of the vehicle 1 for each measurement time and a traveling direction data 42-2 indicating the traveling direction of the vehicle 1 on the horizontal plane for each measurement time. Generate data 42. The traveling state measuring unit 24 writes the generated traveling state data 42 in the storage unit 25.

As shown in FIG. 2, the storage unit 25 stores the GPS observation data 40, the quasi-zenith satellite observation data 41, the traveling state data 42, and the point cloud data 43 described above. The storage unit 25 stores the collected data 44 generated by the collected data generation unit 26 based on the data stored in the storage unit 25.

The collected data generation unit 26 determines the position of the vehicle 1 on the horizontal plane at each measurement time based on the GPS observation data 40 stored by the storage unit 25, the vehicle speed data 42-1 and the traveling direction data 42-2. , Estimate the speed, the traveling direction on the horizontal plane, and the measurement position interval. Here, the position of the vehicle 1 on the horizontal plane is, for example, a position indicated by two-dimensional coordinates indicated by latitude and longitude. The measurement position interval at a certain measurement time is the distance between the position of the vehicle 1 at the certain measurement time and the position of the vehicle 1 at the measurement time immediately before the certain measurement time.

The collected data generation unit 26 has three dimensions of the laser radar device 21 based on the position of the vehicle 1 on the horizontal plane at each estimated measurement time and the positional relationship of the laser radar device 21 fixedly installed in the vehicle 1. Calculate the position in space.

The collected data generation unit 26 sets the direction 180 ° opposite to the traveling direction of the vehicle 1 on the horizontal plane at each estimated measurement time as the direction on the horizontal plane of the laser radar device 21 at each measurement time. As described above, the laser radar device 21 is mounted by fixing the surface formed by the irradiated laser beam to the top of the vehicle 1 so as to be perpendicular to the traveling direction of the vehicle 1. Therefore, since the rotation axis of the laser radar device 21 is always constant with respect to the horizontal plane, the collected data generation unit 26 always sets the direction of the laser radar device 21 on the vertical plane to "0 °".

The collected data generation unit 26 includes each of the measurement times, the position of the laser radar device 21 corresponding to each of the measurement times, the direction of the laser radar device 21 corresponding to each of the measurement times on the horizontal plane and the vertical plane, and the measurement time. The collected data 44 is generated by associating the measurement position interval corresponding to each of the above, the speed of the vehicle 1 corresponding to each of the measurement times, and the traveling direction of the vehicle 1 corresponding to each of the measurement times. The traveling direction of the vehicle 1 is a direction specified by a horizontal component and a vertical component. However, in the first embodiment, since only the traveling direction of the vehicle 1 on the horizontal plane is obtained, the vertical component is obtained. Will not be included.

The collected data generation unit 26 writes the generated collected data 44 to the storage unit 25. The collected data generation unit 26 assigns a reliability index to each of the point cloud data 43 for each measurement time stored in the storage unit 25.

The data correction unit 3 includes a travel locus data generation unit 31, a travel locus analysis unit 32, a measurement condition generation unit 33, a road traffic information acquisition unit 34, a travel locus data normality determination unit 35, and a correction processing unit 36. The travel locus data generation unit 31 generates travel locus data indicating the travel locus of the vehicle 1 based on the GPS observation data 40 stored in the storage unit 25 and the quasi-zenith satellite observation data 41. The travel locus data generation unit 31 is an aspect of the travel locus data generation unit.

The travel locus analysis unit 32 is based on the travel locus data generated by the travel locus data generation unit 31, the position and speed of the vehicle 1 on the horizontal plane at each measurement time, the travel direction on the horizontal plane, and the measurement position. Estimate the interval. The traveling locus analysis unit 32 includes estimated position data indicating the estimated position of the vehicle 1 on the horizontal plane, estimated measurement position interval data indicating the estimated measurement position interval, estimated speed of the vehicle 1, and the estimated vehicle 1 for each measurement time. Generates estimated running state data including the running direction on the horizontal plane. The traveling locus analysis unit 32 is an aspect of the traveling locus analysis unit.

Here, with reference to FIG. 3, a method for estimating the position and speed of the vehicle 1 on the horizontal plane, the traveling direction on the horizontal plane, and the measurement position interval for each measurement time by the traveling locus analysis unit 32. explain. In FIG. 3, the curve of the double line arrow indicated by reference numeral 5 is shown by the travel locus data generated by the travel locus data generation unit 31 based on the GPS observation data 40 and the quasi-zenith satellite observation data 41. It is a running track of.

The positions of the circles indicated by the symbols PJ _-1 , _PJ , PJ _{+ 1} , and PJ _{+ 2} are the positions of the vehicle 1 at the observation time indicated by the GPS observation data 40 and the quasi-zenith satellite observation data 41. The travel locus data generation unit 31 obtains a curve 5 passing through PJ _-1 , _PJ , PJ _{+ 1} , and PJ _{+ 2} as a line that approximates the travel locus of the vehicle 1, and obtains data indicating the obtained curve 5 as travel locus data. Generate as.

Here, it is assumed that there are N measurement times from _PJ to PJ _{+ 1} . Note that FIG. 3 is an example of N = 5. At each of the N measurement times, the positions of the N irradiating laser beams are designated as _{pj (1)} , _{pj (2)} , ..., _{pj (i)} , ..., _{Pj (N)} . do. The final p _{j (N)} corresponds to P _{J + 1} .

At this time, the positions of the remaining N-1 p _{j (1)} , p _{j (2)} , ..., p _{j (i)} , ..., P _{j (N-1)} are between P _J and P _{J + 1} . It is not always evenly spaced. Therefore, the traveling locus analysis unit 32 considers the front-back distance between P _J and P _{J + 1} , that is, the length between P _J-1 and P _J , and the length between P _{J + 1} and P _{J + 2} . , _PJ to PJ _{+ 1} are divided into N pieces, and N-1 pieces of _{pj (1)} , _{pj (2)} , ..., _{pj (i)} , ..., _{pj (N-1)} Find the position.

In the case of the example shown in FIG. 3, the length between PJ _-1 and _PJ _{+ 1} is shorter than the length between _PJ -1 and PJ _{+ 2} and the length between PJ _{+ 1} and PJ + 2. Therefore, in the traveling locus analysis unit 32, the distance between both ends is larger than the distance between the central parts, depending on the length between P _J-1 and P _J and the length between P _{J + 1} and P _{J + 2} . The positions of N-1 p _{j (1)} , p _{j (2)} , ..., p _{j (i)} , ..., p _{j (N-1)} are obtained by dividing them so as to be long. The traveling locus analysis unit 32 sets the respective positions of p _{j (1)} , p _{j (2)} , ..., p _{j (i)} , ..., p _{j (N-1)} , and P _{J + 1} from P _J to P _{J + 1.} Estimated position data for each of N measurement times is used, and the length between each is generated as estimated measurement position interval data for each of N measurement times.

The traveling locus analysis unit 32 generates a vector connecting P _J and p _{j (1)} , and obtains the direction of the generated vector as the traveling direction of the vehicle 1 in p _{j (1)} . The traveling locus analysis unit 32 divides the length of the generated vector by the difference between the measurement time in _PJ and the measurement time in _{pj (1)} , that is, the measurement time interval, in pj ₍₁₎ . Obtained as the speed of vehicle 1. The traveling locus analysis unit 32 generates similar vectors for each of p _{j (2)} , ..., p _{j (i)} , ..., p _{j (N-1)} , and P _{J + 1} , and p _{j (2).} , ..., p _{j (i)} , ..., p _{j (N-1)} , P _{J + 1} The traveling direction of the vehicle 1 and the speed of the vehicle 1 at each position are obtained. The traveling locus analysis unit 32 determines the combination of the traveling direction and the speed at each position of p _{j (2)} , ..., p _{j (i)} , ..., p _{j (N-1)} , and P _J _{+ 1} . To the estimated running state data for each of N measurement times between PJ _{+ 1} .

For example, assuming that the speed of the vehicle 1 is 60 [km / hour], the speed per second is 1666.7 [cm / second]. As described above, the interval between the measurement times of the laser radar device 21 is, for example, about 0.005 seconds. During 0.005 seconds, vehicle 1 moves about 8.3 [cm] (≈1666.7 × 0.005). As described above, the interval between the observation times of the GPS observation data 40 and the quasi-zenith satellite observation data 41 is 1 second, so that there are 200 measurement times in 1 second. As described above, the accuracy of positioning when both the GPS observation data 40 and the quasi-zenith satellite observation data 41 are used is about a dozen centimeters to a few centimeters. Therefore, if the movement changes by about 8.3 [cm] at intervals of 0.005 seconds, the position of the vehicle 1 at each measurement time can be accurately estimated with high probability. On the other hand, the accuracy of positioning when only the GPS observation data 40 is used is at most several meters to several tens of meters. An error of several meters to several tens of meters is a non-negligible error compared to the observation time interval, that is, 1666.7 [cm] in which the vehicle 1 moves in one second, and the position of the vehicle 1 can be estimated accurately. It is assumed that it cannot be done.

The travel locus analysis unit 32 compares the estimated travel state data at a certain measurement time with the travel state data 42 stored in the storage unit 25 corresponding to the measurement time, and the difference in speed is within a predetermined range. Moreover, when the difference in the traveling direction is within a predetermined range, the traveling state data 42 is used to perform a process of improving the accuracy of the estimated position data, the estimated measurement position interval data, and the estimated traveling state data.

Here, the predetermined range for the difference in speed and the predetermined range for the difference in the traveling direction are, for example, the speed and the traveling direction of the vehicle 1 indicated by the traveling state data 42, and the estimated traveling state data. Within a range not significantly different from the speed and traveling direction of the vehicle 1 shown, the traveling state data 42 can be used for processing for improving the accuracy of the estimated position data, the estimated measurement position interval data, and the estimated running state data. Inside.

FIG. 4 is a diagram showing a vector obtained by correcting the vector for specifying the position, speed, and traveling direction of the vehicle 1 shown in FIG. 3 by the vehicle speed data 42-1 and the traveling direction data 42-2 for each measurement time. .. As shown in FIG. 4, according to the vehicle speed data 42-1 and the traveling direction data 42-2 for each measurement time, each vector is delicately reflected in the left-right direction, reflecting the delicate accelerator operation, brake operation, and steering wheel operation. Change. The traveling locus analysis unit 32 generates highly accurate estimated position data and estimated measurement position by generating estimated position data, estimated measurement position interval data, and estimated traveling state data again from each of the vectors shown in FIG. Interval data and estimated running state data will be obtained.

The measurement condition generation unit 33 is the measurement condition of the laser radar device 21 at the measurement time based on the estimated position data for each measurement time generated by the travel locus analysis unit 32, the estimated measurement position interval data, and the estimated travel state data. Generate measurement condition data indicating.

The road traffic information acquisition unit 34 is, for example, a car navigation device, and acquires road traffic information related to the road on which the vehicle 1 is traveling. When the travel locus data normality determination unit 35 receives a normality determination instruction signal including the measurement time and the estimated position data of the vehicle 1 at the measurement time, the road traffic information acquired from the road traffic information acquisition unit 34 and the normality. Based on the estimated position data of the vehicle 1 at the measurement time included in the determination instruction signal, the vehicle 1 is the GPS satellite 10-1, 10 in the tunnel or under the elevated at the measurement time included in the normality determination instruction signal. -It is determined whether or not the travel locus data is normal based on whether or not the vehicle exists at a position where radio waves cannot be normally received from -2 and the quasi-zenith satellite 11.

The correction processing unit 36 corrects the collected data 44 stored in the storage unit 25 based on the measurement condition data generated by the measurement condition generation unit 33. The correction processing unit 36 corrects the reliability index given to the point cloud data 43 stored in the storage unit 25.

(Error included in the collected data)
Next, with reference to FIG. 5, the error included in the collected data 44 generated by the collected data generation unit 26 will be described. FIG. 5 is an enlarged view of a part of the figure shown in FIG. 18, and the same components as those in FIG. 18 are designated by the same reference numerals. In FIG. 5, it is assumed that the vehicle 1 is traveling instead of the vehicle 700, and the traveling locus 541 indicates the traveling locus of the vehicle 1. The traveling locus 541 is generated by the collected data generation unit 26 drawing a line passing through the position of the vehicle 1 for each observation time obtained from the GPS observation data 40, and drawing an approximate curve for the bent portion. Generated.

Actually, radio waves from GPS satellites 10-1 and 10-2 cannot always be received. Therefore, as described above, the collected data generation unit 26 uses the vehicle speed data 42-1 and the traveling direction data 42-2 in addition to the GPS observation data 40 on the horizontal plane of the vehicle 1 at each measurement time. Estimate the position, speed, running direction on the horizontal plane, and measurement position interval.

However, in FIG. 5, in order to explain the error included in the collected data 44 in an easy-to-understand manner, the collected data generation unit 26 uses only the GPS observation data 40 to position the vehicle 1 on the horizontal plane and on the horizontal plane for each measurement time. An example of estimating the traveling direction of is shown. Further, in FIG. 5, the plurality of dotted lines shown along the traveling locus 541 indicate the irradiation direction of the laser beam actually irradiated by the laser radar device 21 mounted on the vehicle 1 at each measurement time. It is a line, and the corresponding measurement times t0 to t20 are attached to each of the laser irradiation lines.

When the vehicle 1 is traveling at a constant speed and traveling straight, the laser irradiation lines shown by the dotted lines are at equal intervals, so that the collected data generation unit 26 uses the GPS observation data based on the difference between the observation time and the measurement time. From 40, the position of the vehicle 1 at the measurement time can be accurately estimated. The

positions

71, 72, 73 of the vehicle 1 (hereinafter, the position 71 of the vehicle 1 is referred to as the vehicle position 71) are the

laser irradiation lines

81, 82, 83 and the traveling locus 541, which are the positions where the laser light is actually irradiated. This is an example in which the collected data generation unit 26 can accurately estimate the position of the vehicle 1 because it coincides with the intersection with the above.

The accuracy of positioning specified by the GPS observation data 40, that is, the error of the specified position is several meters to several tens of meters as described above. On the other hand, when the vehicle 1 is traveling at a speed of 60 [km / hour] and the measurement time interval is 0.005 seconds as described above, the vehicle 1 is during the measurement time interval. Move about 8.3 [cm] to. The moving distance of about 8.3 [cm] is considerably shorter than the distance of several meters to several tens of meters. In consideration of this difference in distance, the collected data generation unit 26 generates a vector having a length of about several hundred to one thousand times the latest measurement position interval, starting from the estimated position of the vehicle 1. However, FIG. 5 shows an example of generating a vector having a length of three times instead of several hundred to one thousand times so that it can be illustrated and explained.

When the collected data generation unit 26 estimates the position of the vehicle 1 at the measurement time t5 corresponding to the laser irradiation line 85, for example, the vehicle position 71 at the measurement time t1 four times before the estimation and the vehicle position 71 three times before. The speed of the vehicle 1 at the vehicle position 72 is calculated based on the vehicle position 72 at the measurement time t2 of. The unit time of the speed calculated by the collected data generation unit 26 is the interval of the measurement time. The collected data generation unit 26 generates a vector V85 obtained by triple the calculated speed along the travel locus 541 starting from the vehicle position 72. The collected data generation unit 26 estimates the position which is the terminal position of the generated vector V85 as the position of the vehicle 1 at the measurement time t5, that is, the vehicle position 75. Further, the collected data generation unit 26 sets the traveling direction of the vehicle 1 at the measurement time t5 as the direction of the vector V85.

Since the vehicle 1 starts decelerating from the vicinity of the measurement time t4 corresponding to the laser irradiation line 84, the vehicle position 75 estimated by the collected data generation unit 26 as the position of the vehicle 1 at the measurement time t5 is the actual position. The position is different from the intersection of the laser irradiation line 85 and the traveling locus 541, which is the position where the laser beam is irradiated at the measurement time t5, and there is a difference indicated by the bidirectional arrow. Assuming a right triangle with the vector V85 as one of the two sides of the right angle, the other side of the right triangle, including the right angle, indicates the laser irradiation direction corresponding to the estimated vehicle position 75. It becomes. In FIG. 5, the laser irradiation line corresponding to the estimated position of the vehicle 1 is shown by a alternate long and short dash line.

The collected data generation unit 26 repeatedly generates a vector such as the above vector V85 starting from the position of the latest observation time, and estimates the position and the traveling direction of the vehicle 1 for each measurement time. Using this method, when the vehicle 1 is traveling at a constant speed and traveling straight, the collected data generation unit 26 accurately estimates the position of the vehicle 1 as in the vehicle positions 71, 72, 73. be able to. On the other hand, when the vehicle 1 starts decelerating, accurate estimation cannot be performed, and the vehicle positions 74, 75, 76 estimated at the measurement times t4, t5, t6 and the vehicle 1 actually irradiated with the laser beam. There will be a deviation from the position of. However, while the vehicle 1 is decelerating but traveling straight, the laser irradiation line corresponding to the estimated position of the vehicle 1 indicated by the alternate long and short dash line is the

laser irradiation line

84, 85, which indicates the actual laser irradiation direction. It will be parallel to 86.

When the vehicle 1 moves from the traveling lane 561 to the overtaking lane 562 in order to avoid the vehicle 750, the traveling locus 541 bends. Here, tentatively, the collected data generation unit 26 estimates the vehicle position 77, which is an accurate position as the position of the vehicle 1 at the measurement time t11 corresponding to the laser irradiation line 87, and also corresponds to the laser irradiation line 88. It is assumed that the vehicle position 78, which is an accurate position, is estimated as the position of the vehicle 1 at the measurement time t12.

When the collected data generation unit 26 estimates the position of the vehicle 1 at the measurement time t15 corresponding to the laser irradiation line 89, the vehicle position 77 at the measurement time t11 four times before and the vehicle position three before the measurement time t11 according to the above procedure. The speed of the vehicle 1 at the vehicle position 78 is calculated based on the vehicle position 78 at the measurement time t12. The collected data generation unit 26 generates a vector V89 in which the magnitude of the calculated speed is tripled along the traveling locus 541. The collected data generation unit 26 estimates the vehicle position 79, which is the terminal position of the generated vector V89, as the position of the vehicle 1 at the measurement time t15. Assuming a right triangle in which the vector V89 is one of two sides sandwiching a right angle, the other side including the right angle of the right triangle indicates the laser irradiation direction corresponding to the estimated vehicle position 79. It becomes. It can be seen that there is a difference between the direction of the laser irradiation line indicated by the alternate long and short dash line passing through the vehicle position 79 and the direction of the laser irradiation line 89 at the actual measurement time t15, which is indicated by the bidirectional arrow.

That is, when the vehicle 1 is not traveling at a constant speed, the collected data 44 generated by the collected data generation unit 26 includes a position error according to the accuracy of the GPS observation data 40. If the vehicle 1 is not traveling straight, an error in the traveling direction according to the accuracy of the GPS observation data 40 will be added. Therefore, when the vehicle 1 is not traveling at a constant speed and is not traveling straight, an error is included in both the position and the traveling direction of the vehicle 1.

(Accuracy when using quasi-zenith satellite observation data)
FIG. 6 shows the position and speed of the vehicle 1 on the horizontal plane for each measurement time estimated by the travel locus analysis unit 32 based on the travel locus data generated by the travel locus data generation unit 31, and the travel on the horizontal plane. It is a figure which showed the direction.

FIG. 6 is an enlarged view of a part of the figure shown in FIG. 18, similarly to FIG. 5, and the same components as those in FIG. 18 are designated by the same reference numerals. Also in FIG. 6, it is assumed that the vehicle 1 is traveling instead of the vehicle 700, and the traveling locus 541 indicates the traveling locus of the vehicle 1. The plurality of dotted lines shown in FIG. 6 are laser irradiation lines indicating the irradiation direction of the laser light irradiated by the laser radar device 21 at each measurement time, and have the same positions and inclinations as those in FIG.

The travel locus data generation unit 31 generates travel locus data based on the GPS observation data 40 and the quasi-zenith satellite observation data 41. As described above, by using the GPS observation data 40 and the quasi-zenith satellite observation data 41, the positioning error becomes about a dozen centimeters to a few centimeters. Therefore, each data of the position, speed, and traveling direction of the vehicle 1 on the horizontal plane at each measurement time estimated by the traveling locus analysis unit 32 is the data estimated by the collected data generation unit 26 corresponding to each. Will be more accurate than.

Therefore, the position of the vehicle 1 for each measurement time estimated by the travel locus analysis unit 32 is indicated by a circle on the travel locus 541 in FIG. 6, and the laser radar device 21 irradiates the laser beam at the actual measurement time. It will match the position. Therefore, even when the vehicle 1 is traveling straight but not traveling at a constant speed, the traveling locus analysis unit 32 sets the position of the vehicle 1 at five consecutive measurement times as the vehicle positions 90 and 91, for example. , 92, 93, 94. A right triangle is assumed in which a vector connecting adjacent objects at vehicle positions 90, 91, 92, 93, and 94 is generated, and the vector generated as described above is one of two sides sandwiching a right angle. Then, the other side of the right triangle sandwiching the right angle coincides with the laser irradiation line.

When the vehicle 1 avoids the vehicle 750 and the travel locus 541 bends, the travel locus analysis unit 32 sets, for example, the position of the vehicle 1 at five consecutive measurement times as the vehicle positions 95, 96, 97, 98, 99. presume. A right triangle with the vector generated as described above as one of the two sides sandwiching the right angle after generating a vector connecting the adjacent ones at the vehicle positions 95, 96, 97, 98, 99. Assuming that, the other side of the right triangle sandwiching the right angle coincides with the inclined laser irradiation line.

Therefore, it is found that the position and traveling direction of vehicle 1 can be estimated with high accuracy by using the quasi-zenith satellite observation data 41 for estimating the position of vehicle 1, and further, the vehicle speed data 42-1 and the traveling direction can be estimated. By using the data 42-2, the accuracy of estimation can be improved.

(Processing by the collected data generator)
FIG. 7 is a flowchart showing the flow of the generation process of the collected data 44 by the collected data generation unit 26. The collected data generation unit 26 detects that the point cloud data 43 corresponding to the new measurement time has been written to the storage unit 25 by the laser radar device 21 (step S1).

Based on the GPS observation data 40, the collected data generation unit 26 measures the position, speed, traveling direction and measurement of the vehicle 1 on the horizontal plane at the detected measurement time by the method described with reference to FIG. Estimate the position interval. The collected data generation unit 26 applies the traveling state data 42 at the detected measurement time when estimating the position and speed of the vehicle 1 on the horizontal plane, the traveling direction on the horizontal plane, and the measurement position interval at the detected measurement time. It also makes corrections. The collected data generation unit 26 is a tertiary of the laser radar device 21 at the detected measurement time based on the estimated position of the vehicle 1 on the horizontal plane and the positional relationship of the laser radar device 21 fixedly installed in the vehicle 1. Calculate the position in the original space.

The collected data generation unit 26 sets the direction 180 ° opposite to the estimated traveling direction of the vehicle 1 on the horizontal plane as the direction on the horizontal plane of the laser radar device 21 at the detected measurement time, and the laser at the detected measurement time. The direction of the radar device 21 on the vertical plane is set to "0 °".

The collected data generation unit 26 has the detected measurement time, the position of the laser radar device 21 corresponding to the measurement time, the direction of the laser radar device 21 on the horizontal plane and on the vertical surface, the measurement position interval, the speed of the vehicle 1, and the speed of the vehicle 1. Collected data 44 is generated in association with the traveling direction of the vehicle 1, and the generated collected data 44 is written in the storage unit 25 (step S2).

In the collected data generation unit 26, the vehicle 1 is traveling at a constant speed depending on whether or not the traveling state data 42 at the detected measurement time and the traveling state data 42 at the measurement time immediately before the measurement time match. , And it is determined whether or not the vehicle is going straight (step S3). That is, the collected data generation unit 26 matches the speed of the vehicle 1 indicated by the vehicle speed data 42-1 at the detected measurement time with the speed of the vehicle 1 indicated by the vehicle speed data 42-1 at the measurement time immediately before the measurement time. By determining whether or not to do so, it is determined whether or not the vehicle 1 is traveling at a constant speed. Further, the collected data generation unit 26 has the traveling direction of the vehicle 1 indicated by the traveling direction data 42-2 at the detected measurement time and the traveling of the vehicle 1 indicated by the traveling direction data 42-2 at the measurement time immediately before the measurement time. By determining whether or not the directions match, it is determined whether or not the vehicle 1 is traveling straight. When the speeds of the two vehicles 1 are the same and the traveling directions of the two vehicles 1 are the same, the collected data generation unit 26 is traveling at a constant speed and traveling straight. (Step S3, Yes), and a confidence index "high" is given to the point group data 43 corresponding to the detected measurement time (step S4).

On the other hand, when the speeds of the two vehicles 1 do not match or the traveling directions of the two vehicles 1 do not match, the collected data generation unit 26 indicates that the vehicle 1 is not at a constant speed or goes straight. It is determined that the data is not present (step S3, No), and the confidence index “low” is given to the point group data 43 corresponding to the detected measurement time (step S5). The collected data generation unit 26 replaces the reliability index “low” with “not” when there is a situation for improving the reliability of the design result in the station design performed using the point cloud data 43, for example. May be given. The collected data generation unit 26 assigns a reliability index "low" when the speed is not constant but is traveling straight, and assigns a reliability index "not" when the speed is not constant and the vehicle is not traveling straight. You may.

The collected data generation unit 26 generates the collected data 44 based on whether or not, for example, an instruction signal for terminating the measurement of the point cloud data 43 is received by the operation input of the user of the point cloud data collecting device 2. It is determined whether or not to end (step S6). When it is determined that the collected data generation unit 26 has received the instruction signal to end the measurement of the point cloud data 43 (step S6, Yes), the process ends.

On the other hand, when the collected data generation unit 26 determines that the instruction signal for ending the measurement of the point cloud data 43 has not been received (step S6, No), the processing from step S1 is continued.

(Processing when using quasi-zenith satellite observation data)
FIG. 8 is a flowchart showing a flow of processing for correcting the collected data 44 by the data correction unit 3.
The traveling locus data generation unit 31 detects that the collection data 44 at the new measurement time has been recorded by the collection data generation unit 26 in the storage unit 25 (step Sa1). The travel locus data generation unit 31 includes GPS observation data 40 at four observation times before and after including the measurement time of the newly recorded collected data 44 (hereinafter, also referred to as “measurement time to be processed”), and quasi-zenith satellite observation. The data 41 and the data 41 are read from the storage unit 25. The travel locus data generation unit 31 generates travel locus data based on the read GPS observation data 40 and the quasi-zenith satellite observation data 41 (step Sa2). The travel locus data generation unit 31 outputs the generated travel locus data and the measurement time of the processing target to the travel locus analysis unit 32.

The travel locus analysis unit 32 captures the measurement time of the processing target output by the travel locus data generation unit 31 and the travel locus data. The traveling locus analysis unit 32 measures the position and speed of the vehicle 1 on the horizontal plane and the traveling direction on the horizontal plane at the measurement time of the processing target based on the traveling locus data by the method shown in FIG. Estimate the position interval. The travel locus analysis unit 32 includes estimated position data indicating the estimated position of the vehicle 1 on the horizontal plane, estimated measurement position interval data indicating the estimated measurement position interval, and estimated speed of the vehicle 1 and traveling on the horizontal plane. Estimated running state data including directions are generated (step Sa3).

The traveling locus analysis unit 32 reads the traveling state data 42 at the measurement time to be processed from the storage unit 25. The travel locus analysis unit 32 compares the read travel state data 42 with the generated estimated travel state data, and the difference in speed is within a predetermined range, and the difference in travel direction is within a predetermined range. It is determined whether or not it is (step Sa4). That is, the travel locus analysis unit 32 is within a range in which the difference between the speed of the vehicle 1 indicated by the vehicle speed data 42-1 included in the travel condition data 42 and the speed of the vehicle 1 included in the estimated travel condition data is predetermined. Difference between the traveling direction of the vehicle 1 indicated by the traveling direction data 42-2 included in the traveling state data 42 and the traveling direction of the vehicle 1 included in the estimated traveling state data on the horizontal plane. Is within a predetermined range.

When the travel locus analysis unit 32 determines that the difference between the two speeds is not within the predetermined range or the difference between the two travel directions is not within the predetermined range (steps Sa4, No), the processing target is The normality determination instruction signal including the measurement time and the generated estimated position data is output to the travel locus data normality determination unit 35.

When the travel locus data normality determination unit 35 receives the normality determination instruction signal from the travel locus analysis unit 32, the travel locus data normality determination unit 35 informs the road traffic information acquisition unit 34 of the road traffic information at the measurement time of the processing target included in the normality determination instruction signal. Request and get. Based on the acquired road traffic information and the position indicated by the estimated position data included in the normality determination instruction signal, the travel locus data normality determination unit 35 sets the vehicle 1 in a tunnel, under an elevated vehicle, or the like at the measurement time to be processed. It is determined whether or not the travel locus data is normal based on whether or not the GPS satellites 10-1 and 10-2 and the quasi-zenith satellite 11 are present at positions where radio waves cannot be normally received (step Sa5).

The travel locus data normality determination unit 35 is the travel locus data when the vehicle 1 is present at a position where radio waves cannot be normally received from the GPS satellites 10-1 and 10-2 and the quasi-zenith satellite 11 at the measurement time to be processed. Determines that it is not normal (steps Sa5 and No), outputs an abnormality to the outside (steps Sa6), and outputs a processing end instruction signal to the travel locus data generation unit 31. The travel locus data generation unit 31 ends the processing when it receives the processing end instruction signal.

On the other hand, the travel locus data normality determination unit 35 travels when the vehicle 1 is present at a position where radio waves can be normally received from the GPS satellites 10-1 and 10-2 and the quasi-zenith satellite 11 at the measurement time to be processed. It is determined that the locus data is normal (steps Sa5, Yes), and a correction instruction signal for which a reliability index "low" including the measurement time of the processing target is given is output to the correction processing unit 36. When the correction processing unit 36 receives the correction instruction signal from the travel locus data normality determination unit 35, the correction processing unit 36 refers to the point cloud data 43 of the storage unit 25, and the point cloud data corresponding to the measurement time of the processing target included in the correction instruction signal. The confidence index of 43 is corrected to "low" (step Sa7). After that, the process proceeds to step Sa13.

Here, in step Sa5, when the travel locus data normality determination unit 35 determines that the travel locus data is normal, the reason for assigning the reliability index "low" to the point cloud data 43 at the measurement time to be processed. There is a difference exceeding a predetermined range between the running state data 42 and the estimated running state data, and there is a possibility that the vehicle 1 slips or the like and the running state data 42 is not normal. Therefore, processing is performed. This is because the point cloud data 43 at the measurement time of the target may not be collected normally.

On the other hand, in step Sa4, when the travel locus analysis unit 32 determines that the difference between the two speeds is within the predetermined range and the difference between the two travel directions is within the predetermined range (step Sa4, Yes), by the method shown in FIG. 4, the running state data 42 at the measurement time to be processed is applied to the estimated position data, the estimated measurement position interval data, and the estimated running state data, and the estimated position data and the estimated running state data are obtained again. The estimated measurement position interval data and the estimated running state data are generated (step Sa8).

The traveling locus analysis unit 32 reads out the collected data 44 corresponding to the measurement time immediately before the measurement time of the processing target from the storage unit 25. The travel locus analysis unit 32 has a measurement position interval included in the read collected data 44, a travel direction of the vehicle 1, each of which is a measurement position interval indicated by the estimated measurement position interval data, and a travel direction of the vehicle 1 indicated by the estimated travel state data. It is determined whether or not each of the above is matched (step Sa9).

When the travel locus analysis unit 32 determines that the distance between the two measurement positions matches and the travel directions of the two vehicles 1 match (steps Sa9, Yes), the vehicle 1 has a constant velocity at the measurement time of the processing target. It means that you are driving at and going straight. Therefore, since it is not necessary to correct the collected data 44 at the measurement time of the processing target, the traveling locus analysis unit 32 proceeds to the process of step Sa13 without proceeding to the process of correcting the collected data 44.

On the other hand, when the travel locus analysis unit 32 determines that the distance between the two measurement positions does not match or the travel directions of the two vehicles 1 do not match (steps Sa9, No), the estimated position data generated again and the estimated position data are used. The estimated measurement position interval data, the estimated running state data, and the measurement time of the processing target are output to the measurement condition generation unit 33.

The measurement condition generation unit 33 is based on the estimated position data output by the traveling locus analysis unit 32 and the positional relationship of the laser radar device 21 fixedly installed in the vehicle 1, and the laser radar device 21 at the measurement time to be processed. Calculate the position of the above in three-dimensional space. The measurement condition generation unit 33 sets the direction 180 ° opposite to the horizontal component direction of the vehicle 1 included in the estimated travel state data output by the travel locus analysis unit 32, and the laser radar device 21 at the measurement time to be processed. The horizontal component in the direction of is set to "0 °" and the vertical component in the direction of the laser radar device 21 is set to "0 °".

The measurement condition generation unit 33 uses the calculated position of the laser radar device 21, the direction of the laser radar device 21 on the horizontal plane and the vertical surface, the measurement position interval indicated by the estimated measurement position interval data, and the estimated running state data. Data including the speed and traveling direction of the included vehicle 1 and the measurement time to be processed is generated as measurement condition data (step Sa10). The measurement condition generation unit 33 outputs the generated measurement condition data to the correction processing unit 36.

The correction processing unit 36 takes in the measurement condition data output by the measurement condition generation unit 33. The correction processing unit 36 selects the collection data 44 corresponding to the measurement time of the processing target included in the measurement condition data captured from the collection data 44 stored in the storage unit 25. The correction processing unit 36 rewrites the data included in the selected collected data 44 with the data included in the measurement condition data, and corrects the collected data 44 (step Sa11). The correction processing unit 36 refers to the point cloud data 43 of the storage unit 25, and corrects the reliability index of the point cloud data 43 corresponding to the measurement time of the processing target included in the measurement condition data to “high” (step). Sa12).

The correction processing unit 36 ends the correction of the collected data 44 based on whether or not, for example, an instruction signal for ending the measurement of the point cloud data 43 is received by the operation input of the user of the point cloud data collecting device 2. It is determined whether or not to do so (step Sa13). When it is determined that the correction processing unit 36 has received the instruction signal to end the measurement of the point cloud data 43 (step Sa13, Yes), the correction processing unit 36 outputs the processing end instruction signal to the travel locus data generation unit 31. The travel locus data generation unit 31 ends the processing when it receives the processing end instruction signal.

On the other hand, when the correction processing unit 36 determines that the instruction signal for ending the measurement of the point cloud data 43 has not been received (step Sa13, No), the correction processing unit 36 outputs the processing continuation instruction signal to the travel locus data generation unit 31. When the travel locus data generation unit 31 receives the processing continuation instruction signal, the processing of step Sa1 is restarted.

In the configuration of the first embodiment described above, the travel locus data generation unit 31 determines the position of the vehicle 1 on the horizontal plane with higher accuracy than the data indicating the position of the vehicle 1 on the horizontal plane obtained at each observation time. The data shown is acquired, and based on the acquired data indicating the position of the vehicle 1 on the horizontal plane, the travel locus data indicating the travel locus of the vehicle 1 is generated. The traveling locus analysis unit 32 analyzes the traveling locus data and estimates the position and traveling state of the vehicle 1 at each measurement time. The measurement condition generation unit 33 generates measurement condition data indicating the measurement conditions of the laser radar device 21 for each measurement time based on the position and running state of the vehicle 1 for each measurement time. The correction processing unit 36 corrects the collected data 44 based on the measurement condition data.

In the collected data 44 generated by the collected data generation unit 26, the reliability index is "low" with respect to the point cloud data 43 collected when the vehicle 1 accelerates or decelerates, turns left or right, or travels in a curved manner. Or, "None" was given. Therefore, in the station design for determining the line-of-sight and the shielding rate between the base station and the terminal station using the point cloud data 43, the point cloud data to which the reliability index "low" or "none" is given. 43 could not be used effectively. On the other hand, in the first embodiment, the quasi-zenith satellite observation data 41 is used in addition to the GPS observation data 40. As a result, even when the vehicle 1 accelerates or decelerates, turns left or right, or travels in a bent manner, the position of the vehicle 1 and the traveling direction on the horizontal plane can be estimated with high accuracy. .. Thereby, the position of the laser radar device 21 included in the collected data 44 and the direction of the laser radar device 21 on the horizontal plane can be accurately corrected. Therefore, the reliability of the point cloud data 43 collected when the vehicle 1 accelerates or decelerates, turns left or right, or travels in a bent manner can be increased, so that the vehicle 1 can turn left or right or travel in a bent manner. In this case, the confidence index "high" can be given to the point cloud data 43 collected. Therefore, by estimating the position and running state of the vehicle at the place where the laser irradiation line is disturbed with higher accuracy than before, it is possible to use the point cloud data whose reliability is lowered due to the disturbance of the laser irradiation line. For example, the point cloud data 43 that can be used when designing a station can be increased.

In the first embodiment described above, the traveling state measuring unit 24 acquires the steering angle of the steering for each measurement time from the steering sensor and calculates the traveling direction of the vehicle 1 on the horizontal plane. Connects to a three-dimensional gyro sensor (3D gyro sensor), a compass, etc., acquires information indicating the direction from the three-dimensional gyro sensor, the compass, etc., and determines the traveling direction of the vehicle 1 on the horizontal plane. It may be calculated.

(Second embodiment)
FIG. 9 is a block diagram showing a configuration of the point cloud data collection system β of the second embodiment. In the point cloud data collection system β of the second embodiment, the same configurations as the point cloud data collection system α of the first embodiment are designated by the same reference numerals, and different configurations will be described below. The point cloud data collection system β includes a vehicle 1a equipped with a point cloud data collection device 2a, a plurality of GPS satellites 10-1, 10-2 and a quasi-zenith satellite 11, and a location information service provider server device 50. ing. The vehicle 1a is, for example, an automobile and corresponds to the above-mentioned MMS.

The point cloud data collecting device 2a of the second embodiment estimates the traveling direction of the vehicle 1a on the vertical surface in addition to the traveling direction of the vehicle 1a on the horizontal plane. Therefore, the point cloud data collecting device 2a is referred to as, for example, the pan-global positioning navigation satellite system (hereinafter referred to as “GNSS” (Global Navigation Satellite System)) shown in Reference 1-1 below as a method for detecting the altitude of the vehicle 1a. .. ) Is used.

[Reference 1-1: "Technology to realize computerized construction, Pan-Earth Positioning Navigation Satellite System (GNSS)", Ministry of Land, Infrastructure, Transport and Tourism, Kyushu Regional Development Bureau, [Search on July 12, 2nd year of Reiwa], Internet (URL) : Http://www.qsr.mlit.go.jp/ict/technology/jitsugen_3.html)]

GNSS is a positioning system that uses positioning satellites. Here, the positioning satellites include, for example, GLONASS (Global Navigation Satellite System) in Russia, GALIEO in Europe, and the like, in addition to the GPS satellites 10-1 and 10-2 and the quasi-zenith satellite 11 described above. In principle, the position of the satellite radio wave receiver is specified by each distance from the three positioning satellites whose positions are known in space to the satellite radio wave receiver. As shown in Reference 1-1, there are the following types of positioning methods using GNSS.

For example, in the RTK (Real Time Kinematic) -GNSS positioning method, a mobile station equipped with a satellite radio receiver is installed at the observation point to be measured, and a reference station whose position is known is installed in addition to the mobile station for positioning. The accuracy of positioning by the RTK-GNSS positioning method is about 2 to 3 cm in the horizontal direction and about 3 to 4 cm in the vertical direction.

On the other hand, there is a network type RTK-GNSS positioning method shown in Reference 1-1. The network type RTK-GNSS positioning method includes a VRS (Virtual Reference Station) method and an FKP (Flachen Korrektur Parameter) method.

In the network type RTK-GNSS positioning method, satellite radio receivers installed at multiple electronic reference points installed nationwide receive data from the positioning satellite. The Geospatial Information Authority of Japan's analysis equipment receives and analyzes data from each of the satellite radio receivers installed at multiple electronic reference points. The processing device of the location information service provider receives the analysis result from the analysis device of the Geographical Survey Institute, and generates the correction data 45 indicating the phase difference obtained from the reference point of the Geographical Survey Institute. The person performing the positioning installs a mobile station equipped with a satellite radio wave receiver at the observation point, and performs positioning based on the data from the positioning satellite received at the observation point and the correction data 45 received through the communication line. The network type RTK-GNSS positioning method is slightly less accurate than the RTK-GNSS positioning method, but unlike the RTK-GNSS positioning method, it does not require the installation of a reference station. The RTK-GNSS positioning method and the network type RTK-GNSS positioning method have better accuracy as the number of satellites increases, and the accuracy is better when the sky is open than the valley.

In the second embodiment, an example of using the network type RTK-GNSS positioning method is shown, but if a reference station can be installed, the RTK-GNSS positioning method may be used.

Returning to FIG. 9, the location information service provider server device 50 is a device corresponding to the processing device of the location information service provider that receives the analysis result from the above-mentioned analysis device of the Geographical Survey Institute, and generates correction data 45. do. The location information service provider server device 50 transmits the generated correction data 45 to the point cloud data collection device 2a of the vehicle 1a via, for example, a wireless communication network.

The point cloud data collecting device 2a includes a laser radar device 21, a satellite radio wave receiving antenna 22, a radio wave receiving antenna 27, an information receiving unit 23a, a traveling state measuring unit 24a, a storage unit 25a, a collected data generation unit 26a, and data correction. A unit 3a is provided.

The radio wave receiving antenna 27 receives the radio wave on which the correction data 45 is superimposed from the location information service provider server device 50 via the wireless communication network. The information receiving unit 23a has the following configuration in addition to the configuration provided in the information receiving unit 23. The information receiving unit 23a is connected to the radio wave receiving antenna 27, detects the correction data 45 superimposed on the radio wave received by the radio wave receiving antenna 27, and stores the detected correction data 45 in the storage unit 25a. Write.

The data correction unit 3a includes a travel locus data generation unit 31a, a travel locus analysis unit 32a, a measurement condition generation unit 33a, a road traffic information acquisition unit 34, a travel locus data normality determination unit 35, and a correction processing unit 36.

The traveling state measuring unit 24a is connected to, for example, a vehicle speed sensor for detecting the speed of the vehicle 1a, a steering sensor for detecting the steering angle of the steering, and a horizontal level. The traveling state measuring unit 24a measures the speed of the vehicle 1a at each measurement time based on the data obtained from the vehicle speed sensor. The traveling state measuring unit 24a measures the steering angle of the steering at each measurement time based on the data obtained from the steering sensor. The traveling state measuring unit 24a measures the inclination of the vehicle 1a at each measurement time based on the data obtained from the horizontal level.

The traveling state measuring unit 24a calculates the traveling direction of the vehicle 1a on the horizontal plane from the steering angle of the steering at each measurement time, and calculates the traveling direction on the vertical surface from the inclination of the vehicle 1a at each measurement time. The traveling state measuring unit 24a is specified by the vehicle speed data 42-1 indicating the speed of the vehicle 1a at each measurement time, the traveling direction of the vehicle 1a on the horizontal plane at each measurement time, and the traveling direction on the vertical surface. The traveling state data 42a including the traveling direction data 42a-2 indicating the traveling direction is generated. The traveling state measuring unit 24a writes the generated traveling state data 42a in the storage unit 25a.

The collected data generation unit 26a has the same configuration as the collected data generation unit 26 of the first embodiment except for the following points. The collected data generation unit 26 of the first embodiment referred to the traveling direction data 42-2, but the collected data generation unit 26a of the second embodiment refers to the traveling direction data 42a-2. Only the direction of the horizontal component of the traveling direction included in the traveling direction data 42a-2 is referred to.

As shown in FIG. 10, the storage unit 25a stores the GPS observation data 40, the quasi-zenith satellite observation data 41, the correction data 45, the traveling state data 42a, the point group data 43, and the collected data 44.

The travel locus data generation unit 31a generates travel locus data indicating the travel locus of the vehicle 1a based on the GPS observation data 40 stored in the storage unit 25a, the quasi-zenith satellite observation data 41, and the correction data 45. The correction data 45 is data generated from GPS observation data 40, quasi-zenith satellite observation data 41, and the like, and the correction data 45 includes the observation time. Therefore, the travel locus data generation unit 31a generates travel locus data using the correction data 45 corresponding to the observation time of the GPS observation data 40 and the quasi-zenith satellite observation data 41. The travel locus data generated by the travel locus data generation unit 31a includes components in the vertical direction, unlike the travel locus data generated by the travel locus data generation unit 31 of the first embodiment.

Based on the travel locus data generated by the travel locus data generation unit 31a, the travel locus analysis unit 32a extends the method shown in FIG. 3 in the vertical direction in the first embodiment to the vehicle for each measurement time. The position of 1a, the speed, the traveling direction, and the measurement position interval are estimated. Here, since the position of the vehicle 1a at each measurement time includes a horizontal component and a vertical component, it is a position indicated by three-dimensional coordinates indicated by, for example, latitude, longitude, and altitude. The traveling direction of the vehicle 1a at each measurement time is a direction specified by a horizontal component and a vertical component.

The travel locus analysis unit 32a includes estimated position data indicating the position of the vehicle 1a for each estimated measurement time, estimated measurement position interval data indicating the estimated measurement position interval, and speed and travel of the vehicle 1a for each estimated measurement time. Generate estimated driving condition data including direction.

The traveling locus analysis unit 32a compares the estimated traveling state data at a certain measurement time with the traveling state data 42a stored in the storage unit 25a corresponding to the measurement time, and the difference in speed is within a predetermined range. Moreover, when the difference in the traveling direction is within a predetermined range, the traveling state data 42a is used to perform a process of improving the accuracy of the estimated position data, the estimated measurement position interval data, and the estimated traveling state data.

Here, the predetermined range for the difference in speed and the predetermined range for the difference in the traveling direction are, for example, the speed and the traveling direction of the vehicle 1a indicated by the traveling state data 42a, and the estimated traveling state data. Within a range not significantly different from the speed and traveling direction of the vehicle 1a shown, the traveling state data 42a can be used for processing for improving the accuracy of the estimated position data, the estimated measurement position interval data, and the estimated running state data. Inside.

The measurement condition generation unit 33a is the measurement condition of the laser radar device 21 at the measurement time based on the estimated position data for each measurement time generated by the travel locus analysis unit 32a, the estimated measurement position interval data, and the estimated travel state data. Generate measurement condition data indicating.

(Change in laser irradiation direction when the position of the vehicle changes in the vertical direction)
With reference to FIGS. 11 and 12, changes in the laser irradiation direction when the position of the vehicle 1a fluctuates in the vertical direction will be described. FIG. 11 is a view of the vehicle 1a traveling on the uneven road 300 as viewed from the side surface. The double-lined arrow indicated by reference numeral 200 indicates the locus on which the vehicle 1a has traveled, and is hereinafter referred to as the travel locus 200. The vehicle 1a is traveling to the right as indicated by the arrow of the traveling locus 200. The alternate long and short dash line indicated by reference numeral 210 indicates the average altitude of the vehicle 1a, and is hereinafter referred to as the vehicle average altitude 210.

The traveling locus 200 changes up and down depending on the unevenness of the road 300. By comparing with the vehicle average altitude 210, it is possible to know whether the position of the vehicle 1a is higher or lower than the average altitude. For example, it can be seen that in the section indicated by reference numeral 310, the vehicle 1a is located at a position higher than the average altitude, and in the section indicated by reference numeral 320, the vehicle 1a is located at a position lower than the average altitude. Compared with the change in the traveling direction on the horizontal plane such as turning left and right and bending described in the first embodiment, the change in the traveling direction on the vertical surface is small. For example, on a 5% uphill and downhill, the change on the vertical surface is 5m for a 100m run, but it is generally said to be a steep slope.

FIG. 12 is a diagram showing changes in the laser irradiation direction of the laser radar device 21 of the vehicle 1a traveling on the road 300 shown in FIG. As described above, even if the road is generally called a steep slope, the change on the vertical surface is small. Therefore, in FIG. 12, the change in the traveling direction on the vertical surface is largely shown. The change in the unevenness of the road 300 is exaggerated.

A plurality of dotted lines shown along the traveling locus 200 are laser irradiation lines indicating the irradiation direction of the laser light emitted by the laser radar device 21 mounted on the vehicle 1a at each measurement time. As can be seen from FIG. 12, when the vehicle 1a is traveling at a constant speed and the altitude of the vehicle 1a does not change, the laser irradiation lines are perpendicular to the traveling direction and are arranged in parallel at equal intervals. become. On the other hand, when the altitude of the vehicle 1a changes up and down, the angle of the laser irradiation line changes according to the inclination of the vehicle 1a, so that it is not parallel and turbulent. ..

Therefore, even if the position of each point of the point cloud data 43 is specified based on the collected data 44 without considering the vertical component of the traveling direction of the vehicle 1a, the direction of the laser radar device 21 is not accurate. The position of each point in the point cloud data 43 is also not accurate. Therefore, it is appropriate that the point cloud data 43 collected while the altitude of the vehicle 1a is changing has a reliability index of "low" or "absent", as in the case of turning left or right or bending. be.

In the second embodiment, the travel locus data generation unit 31a generates travel locus data based on the GPS observation data 40, the quasi-zenith satellite observation data 41, and the correction data 45. As described above, by using the GPS observation data 40, the quasi-zenith satellite observation data 41, and the correction data 45, the altitude of the vehicle 1a can be estimated with a high accuracy of about 3 to 4 cm.

Therefore, the position of the vehicle 1a at each measurement time estimated by the travel locus analysis unit 32a is, for example, as shown by a circle on the travel locus 200 in FIG. 12, the laser radar device 21 emits laser light at the actual measurement time. It will match the irradiated position. For example, suppose that when the vehicle 1a is traveling straight at a constant speed, the traveling locus analysis unit 32a estimates the positions of the vehicle 1a at four consecutive measurement times as the vehicle positions 60, 61, 62, 63. A vector connecting the adjacent portions at the vehicle positions 60, 61, 62, and 63 is generated, and the generated vector as described in the first embodiment is used as one of the two sides sandwiching the right angle. Assuming a right triangle, the other side of the right triangle that sandwiches the right angle coincides with the laser irradiation line.

It is assumed that when the altitude of the vehicle 1a changes, the traveling locus analysis unit 32a estimates the positions of the vehicle 1a at four consecutive measurement times as the vehicle positions 64, 65, 66, 67. Also in this case, after generating a vector connecting the adjacent ones at the vehicle positions 64, 65, 66, 67, the vector generated as described above is set as one of the two sides sandwiching the right angle. Assuming a right triangle, the other side of the right triangle that sandwiches the right angle coincides with the tilted laser irradiation line.

Therefore, by using the GPS observation data 40 and the quasi-zenith satellite observation data 41 to estimate the position of the vehicle 1a, it is possible to estimate the position and the traveling direction of the vehicle 1a in consideration of the vertical direction with high accuracy. Further, by using the vehicle speed data 42-1 and the traveling direction data 42a-2, it is possible to estimate with higher accuracy.

(Processing of the second embodiment)
The collected data generation unit 26a performs the same processing as the collected data generation unit 26 of the first embodiment to generate the collected data 44. FIG. 13 is a flowchart showing a flow of processing for correcting the collected data 44 by the data correction unit 3a.

The travel locus data generation unit 31a detects that the collection data 44 at the new measurement time has been recorded by the collection data generation unit 26a in the storage unit 25a (step Sb1). The travel locus data generation unit 31a includes GPS observation data 40 at four observation times before and after including the measurement time of the newly recorded collected data 44 (hereinafter, also referred to as “measurement time to be processed”), and quasi-zenith satellite observation. The data 41 and the correction data 45 are read from the storage unit 25a. The travel locus data generation unit 31a generates travel locus data including a vertical component based on the read GPS observation data 40, the quasi-zenith satellite observation data 41, and the correction data 45 (step Sb2). The travel locus data generation unit 31a outputs the generated travel locus data and the measurement time of the processing target to the travel locus analysis unit 32a.

The travel locus analysis unit 32a captures the measurement time of the processing target output by the travel locus data generation unit 31a and the travel locus data. The travel locus analysis unit 32a extends the method shown in FIG. 3 in the vertical direction, and based on the travel locus data, the position, speed, and horizontal of the vehicle 1a in the three-dimensional space at the measurement time of the processing target. The traveling direction including the component and the vertical component and the measurement position interval are estimated. The travel locus analysis unit 32a obtains estimated position data indicating the estimated position of the vehicle 1a, estimated measurement position interval data indicating the estimated measurement position interval, and estimated travel state data including the estimated speed and travel direction of the vehicle 1a. Generate (step Sb3).

The traveling locus analysis unit 32a reads out the traveling state data 42a at the measurement time to be processed from the storage unit 25a. The travel locus analysis unit 32a compares the read travel state data 42a with the generated estimated travel state data, and the difference in speed is within a predetermined range, and the difference in travel direction is within a predetermined range. (Step Sb4). That is, the traveling locus analysis unit 32a is within a range in which the difference between the speed of the vehicle 1a indicated by the vehicle speed data 42-1 included in the traveling state data 42a and the speed of the vehicle 1a included in the estimated traveling state data is predetermined. In addition to determining whether or not there is, the difference between the traveling direction of the vehicle 1a indicated by the traveling direction data 42a-2 included in the traveling state data 42a and the traveling direction of the vehicle 1a included in the estimated traveling state data is predetermined. Determine if it is within range.

When the travel locus analysis unit 32a determines that the difference between the two speeds is not within the predetermined range or the difference between the two travel directions is not within the predetermined range (steps Sb4, No), the processing target is The normality determination instruction signal including the measurement time and the generated estimated position data is output to the travel locus data normality determination unit 35.

The processing of steps Sb5, Sb6, and Sb7 is the same as that of steps Sa5, Sa6, and Sa7 shown in FIG. 8, and the traveling locus data normality determination unit 35, the road traffic information acquisition unit 34, and the correction processing unit 36 perform the same processing.

On the other hand, when the travel locus analysis unit 32a determines that the difference between the two speeds is within the predetermined range and the difference between the two travel directions is within the predetermined range (steps Sb4, Yes), the figure. The method shown in 4 is extended in the vertical direction, and the running state data 42a at the measurement time to be processed is applied to the estimated position data, the estimated measurement position interval data, and the estimated running state data, and the estimated position is again estimated. The data, the estimated measurement position interval data, and the estimated running state data are generated (step Sb8).

The traveling locus analysis unit 32a reads out the collected data 44 corresponding to the measurement time immediately before the measurement time of the processing target from the storage unit 25a. In the travel locus analysis unit 32a, the measurement position interval included in the read collected data 44 and the travel direction of the vehicle 1a are each the measurement position interval indicated by the estimated measurement position interval data and the travel direction of the vehicle 1a indicated by the estimated travel state data. It is determined whether or not each of the above is matched (step Sb9).

When the travel locus analysis unit 32a determines that the distance between the two measurement positions matches and the travel directions of the two vehicles 1a match (steps Sb9, Yes), the vehicle 1a travels at a constant speed at the measurement time to be processed. And you are going straight. Therefore, it is not necessary to correct the collected data 44 at the measurement time of the processing target. Therefore, the traveling locus analysis unit 32a does not proceed to the process of correcting the collected data 44, but proceeds to the process of step Sb13.

On the other hand, when the travel locus analysis unit 32a determines that the distance between the two measurement positions does not match or the travel directions of the two vehicles 1a do not match (steps Sb9, No), the estimated position data generated again and the estimated position data are used. The estimated measurement position interval data, the estimated running state data, and the measurement time of the processing target are output to the measurement condition generation unit 33a.

The measurement condition generation unit 33a is based on the estimated position data output by the traveling locus analysis unit 32a and the positional relationship of the laser radar device 21 fixedly installed in the vehicle 1a, and the laser radar device 21 at the measurement time of the processing target. Calculate the position of the above in three-dimensional space. The measurement condition generation unit 33a uses the direction 180 ° opposite to the horizontal component direction of the vehicle 1a included in the estimated travel state data output by the travel locus analysis unit 32a as the horizontal component in the direction of the laser radar device 21. The direction 180 ° opposite to the direction of the vertical component in the traveling direction of the vehicle 1a is defined as the vertical component in the direction of the laser radar device 21. That is, when the traveling direction of the vehicle 1a is indicated by a vector, the direction of the inverse vector of the vector indicates the direction of the laser radar device 21. Therefore, the measurement condition generation unit 33a obtains the direction of the horizontal component and the direction of the vertical component of the laser radar device 21 as described above.

The measurement condition generation unit 33a uses the calculated position of the laser radar device 21, the direction of the laser radar device 21 on the horizontal plane and the vertical surface, the measurement position interval indicated by the estimated measurement position interval data, and the estimated running state data. Data including the speed and traveling direction of the included vehicle 1a and the measurement time to be processed is generated as measurement condition data (step Sb10). The measurement condition generation unit 33a outputs the generated measurement condition data to the correction processing unit 36.

The correction processing unit 36 takes in the measurement condition data output by the measurement condition generation unit 33a. The correction processing unit 36 selects the collection data 44 corresponding to the measurement time of the processing target included in the measurement condition data captured from the collection data 44 stored in the storage unit 25a. The correction processing unit 36 rewrites the data included in the selected collected data 44 with the data included in the measurement condition data, and corrects the collected data 44 (step Sb11). The correction processing unit 36 refers to the point cloud data 43 of the storage unit 25a, and corrects the reliability index of the point cloud data 43 corresponding to the measurement time of the processing target included in the measurement condition data to “high” (step). Sb12).

The processing of step Sb13 is the same as that of step Sa13 shown in FIG. 8, and the correction processing unit 36 performs the same processing.

In the configuration of the second embodiment described above, the travel locus data generation unit 31a acquires data indicating the position of the vehicle 1a on the vertical surface obtained at each observation time, and acquires the data indicating the position of the vehicle 1a on the vertical surface. Based on the data indicating the position of, the traveling locus data indicating the traveling locus of the vehicle 1a is generated. The travel locus analysis unit 32a analyzes the travel locus data and estimates the position and travel state of the vehicle 1a at each measurement time. The measurement condition generation unit 33a generates measurement condition data indicating the measurement conditions of the laser radar device 21 for each measurement time based on the position and running state of the vehicle 1a for each measurement time. The correction processing unit 36 corrects the collected data 44 based on the measurement condition data.

When the vehicle 1a travels with the altitude of the vehicle 1a fluctuating up and down, an error is included in the position and direction of the laser radar device 21 included in the collected data 44 generated by the collected data generation unit 26, but the error is included. An appropriate reliability index was not given to the point cloud data 43 corresponding to the collected data 44. Therefore, the point cloud data 43 having a low degree of reliability may be used in the station design. On the other hand, in the second embodiment, since the altitude of the vehicle 1a is detected, the position of the vehicle 1a and the horizontal plane even when the vehicle 1a travels while the altitude of the vehicle 1a fluctuates up and down. And the traveling direction on the vertical surface can be estimated with high accuracy. Thereby, the position of the laser radar device 21 included in the collected data 44 and the horizontal and vertical directions of the laser radar device can be accurately corrected. Therefore, the reliability of the point cloud data 43 collected when the vehicle 1a travels with the altitude of the vehicle 1a fluctuating up and down can be increased, so that the vehicle 1a travels with the altitude of the vehicle 1a fluctuating up and down. In this case, the confidence index "high" can be given to the point cloud data 43 collected. Therefore, by estimating the position and running state of the vehicle in the place where the laser irradiation line is disturbed with higher accuracy than before, it is possible to use the point cloud data whose reliability is lowered due to the disturbance of the laser irradiation line. For example, it is possible to increase the reliability of the point cloud data 43 that can be used when designing the station.

In the second embodiment described above, the traveling state measuring unit 24a measures the steering angle of the steering at each measurement time based on the data obtained from the steering sensor, and measures based on the data obtained from the horizontal level. Although the inclination of the vehicle 1a is measured at each time, the present invention is not limited to the embodiment. The traveling state measuring unit 24a acquires data obtained from the three-dimensional gyro sensor instead of the steering sensor and the horizontal level, and based on the acquired data, determines the traveling direction of the vehicle 1a on the horizontal plane and on the vertical surface. It may be calculated.

(Third embodiment)
FIG. 14 is a block diagram showing a configuration of the point cloud data collection system γ according to the third embodiment. In the point cloud data collection system γ of the third embodiment, the same configurations as the point cloud data collection systems α and β of the first and second embodiments are designated by the same reference numerals, and the following, different configurations are used. explain. The point cloud data collection system γ includes a vehicle 1b equipped with a point cloud data collection device 2b, a plurality of GPS satellites 10-1, 10-2 and a quasi-zenith satellite 11, and a location information service provider server device 50. ing. The vehicle 1b is, for example, an automobile and corresponds to the above-mentioned MMS.

The point cloud data collecting device 2b includes a laser radar device 21, a satellite radio wave receiving antenna 22, a radio wave receiving antenna 27, an information receiving unit 23a, a traveling state measuring unit 24a, a storage unit 25a, a collected data generation unit 26b, and data correction. A part 3b is provided.

Unlike the first and second embodiments, the laser radar device 21 is installed at a constant angle indicated by reference numeral 100 with respect to the top of the vehicle 1b. Hereinafter, the constant angle indicated by the reference numeral 100 is referred to as an inclination angle 100. Therefore, the rotation axis of the laser radar device 21 is also tilted at an angle of inclination of 100, and the direction of the laser radar device 21 is the direction indicated by the arrow of reference numeral 4b.

For example, Japanese Patent Application Laid-Open No. 2017-156179 describes that "the device is arranged so that the angle of the scan line oscillated by the laser scanner is oblique with respect to the vertical direction" (paragraph [0018]). By inclining the angle of the scan line in this way, "the state of the equipment is accurately detected without slowing down the traveling speed of the inspection vehicle or increasing the number of irradiation points per second of the laser scanner. It is stated that the effect of "can be done" (paragraph [0019]) can be obtained.

In the third embodiment, the laser radar device 21 is tilted at an tilt angle of 100 in order to obtain the effect shown in Japanese Patent Application Laid-Open No. 2017-156179. As a result, the irradiation direction of the laser beam to be irradiated is also tilted at an inclination angle of 100. Therefore, it is necessary to generate the collected data 44 in consideration of the inclination angle 100. When the altitude of the vehicle 1b fluctuates up and down, it is necessary to correct the collected data 44 in consideration of the inclination angle 100.

The collected data generation unit 26b has the same configuration as the collected data generation unit 26a of the second embodiment except for the following points. Similar to the collected data generation unit 26 of the first embodiment, the collected data generation unit 26a of the second embodiment assumes that the rotation axis of the laser radar device 21 is always constant with respect to the horizontal plane, and the laser radar device. The collected data 44 is generated with the direction of 21 on the vertical plane always set to "0 °". On the other hand, the collected data generation unit 26b of the third embodiment always generates the collected data 44 with the direction of the laser radar device 21 on the vertical plane as the tilted direction by the tilt angle of 100.

The data correction unit 3b includes a travel locus data generation unit 31a, a travel locus analysis unit 32a, a measurement condition generation unit 33b, a road traffic information acquisition unit 34, a travel locus data normality determination unit 35, and a correction processing unit 36.

The measurement condition generation unit 33b is a laser radar at the measurement time based on the estimated position data for each measurement time, the estimated measurement position interval data, the estimated travel state data, and the inclination angle 100 generated by the travel locus analysis unit 32a. Generates measurement condition data indicating the measurement conditions of the device 21.

FIG. 15 is a diagram showing changes in the laser irradiation line when the vehicle 1b of the third embodiment travels on the road 300 shown in FIG. 11 of the second embodiment. Since the vehicle 1a of the second embodiment and the vehicle 1b of the third embodiment have the same shape, the average vehicle altitude is the same and the traveling locus is also the same. Therefore, the traveling locus 200 and the vehicle average altitude 210 are shown with the same reference numerals as those in FIG.

A plurality of dotted lines shown along the traveling locus 200 are laser irradiation lines indicating the irradiation direction of the laser light emitted by the laser radar device 21 mounted on the vehicle 1b at each measurement time. The laser radar device 21 is tilted at an inclination angle of 100. Therefore, all the laser irradiation lines are also inclined at an inclination angle of 100.

Similar to the second embodiment, the position of the vehicle 1b for each measurement time estimated by the travel locus analysis unit 32a is, for example, a laser at the actual measurement time as shown by a circle on the travel locus 200 in FIG. It coincides with the position where the radar device 21 irradiates the laser beam. For example, suppose that when the vehicle 1b is traveling straight at a constant speed, the traveling locus analysis unit 32a estimates the positions of the vehicle 1a at four consecutive measurement times as the vehicle positions 60, 61, 62, 63. A vector connecting the adjacent portions at the vehicle positions 60, 61, 62, and 63 is generated, and the generated vector as described in the first embodiment is used as one of the two sides sandwiching the right angle. Assume a right triangle. In the third embodiment, since the laser radar device 21 is tilted at an angle of inclination of 100, tilting the other side of the right triangle sandwiching the right angle at an angle of tilt of 100 coincides with the laser irradiation line. become.

It is assumed that when the altitude of the vehicle 1b changes, the traveling locus analysis unit 32a estimates the positions of the vehicle 1b at four consecutive measurement times as the vehicle positions 64, 65, 66, 67. Also in this case, after generating a vector connecting the adjacent ones at the vehicle positions 64, 65, 66, 67, the vector generated as described above is set as one of the two sides sandwiching the right angle. Assume a right triangle. If the other side of the right triangle sandwiching the right angle is tilted at an tilt angle of 100, it coincides with the laser irradiation line.

The measurement condition generation unit 33b is based on the travel direction of the vehicle 1b included in the estimated travel state data generated by the travel locus analysis unit 32a, and takes into consideration the inclination angle 100 which is the inclination of the laser radar device 21 and the laser radar device 21. Calculate the horizontal and vertical components in the direction of. By using the calculated horizontal component and vertical component in the direction of the laser radar device 21, it is possible to accurately specify the position of each point of the point cloud data 43 in the three-dimensional space.

(Processing of the third embodiment)
The collected data generation unit 26b performs the same processing as the collected data generation unit 26 of the first embodiment to generate the collected data 44. However, the collected data generation unit 26b generates the collected data 44 with the direction on the vertical surface of the direction of the laser radar device 21 always set as the tilted direction by the tilt angle of 100.

The process of correcting the collected data 44 in the third embodiment is the same as that of the second embodiment except for the process of step Sb10 among the processes of the second embodiment shown in FIG. ..

In the third embodiment, the following processing is performed as the processing of step Sb10. The measurement condition generation unit 33b is based on the estimated position data output by the traveling locus analysis unit 32a and the positional relationship of the laser radar device 21 fixedly installed in the vehicle 1b, and the laser radar device 21 at the measurement time of the processing target. Calculate the position of the above in three-dimensional space. The measurement condition generation unit 33b sets the direction 180 ° opposite to the direction of the horizontal component in the travel direction of the vehicle 1b included in the estimated travel state data output by the travel locus analysis unit 32a as the direction on the horizontal plane of the laser radar device 21. The direction obtained by adding the inclination angle 100 to the direction 180 ° opposite to the direction of the vertical component in the traveling direction of the vehicle 1b is defined as the vertical component in the direction of the laser radar device 21. That is, when the traveling direction of the vehicle 1b is indicated by a vector, the vector obtained by tilting the inverse vector of the vector in the vertical direction by the inclination angle of 100 indicates the direction of the laser radar device 21. Therefore, the measurement condition generation unit 33b obtains the direction of the horizontal component and the direction of the vertical component of the laser radar device 21 as described above.

The measurement condition generation unit 33b uses the calculated position of the laser radar device 21, the direction of the laser radar device 21 on the horizontal plane and the vertical surface, the measurement position interval indicated by the estimated measurement position interval data, and the estimated running state data. Data including the speed and traveling direction of the included vehicle 1b and the measurement time to be processed is generated as measurement condition data.

In the third embodiment, in addition to the effects obtained in the second embodiment, the following effects are exhibited. When the laser radar device 21 is tilted and attached to the top of the vehicle 1b and the vehicle 1b travels in which the altitude of the vehicle 1b fluctuates up and down, the vertical component in the direction of the laser radar device 21 changes in a complicated manner. It will be. On the other hand, by applying the configuration of the third embodiment, it is possible to calculate an accurate vertical component in the direction of the laser radar device 21 in consideration of the inclination of the laser radar device 21. Thereby, the vertical component in the direction of the laser radar device 21 included in the collected data 44 can be accurately corrected. Therefore, the reliability of the point cloud data 43 collected when the vehicle 1b travels in which the altitude of the vehicle 1b fluctuates up and down can be increased, and the reliability index "high" is given to the point cloud data 43. be able to. Therefore, by estimating the position and running state of the vehicle in the place where the laser irradiation line is disturbed with higher accuracy than before, it is possible to use the point cloud data whose reliability is lowered due to the disturbance of the laser irradiation line. For example, it is possible to increase the reliability of the point cloud data 43 that can be used when designing the station.

In the second and third embodiments described above, the traveling locus analysis unit 32a detects the altitudes of the

vehicles

1a and 1b by using the network type RTK-GNSS positioning method, but other methods are used. You may. For example, the optical lattice clock shown in References 2-1, 2, 2 and 2-3 may be used. An optical lattice clock is an atomic clock in which atoms are placed one by one in an optical lattice, which is a large number of regions smaller than the wavelength of light created by interfering with laser light, and another laser light is applied to measure the resonance frequency. In Einstein's general theory of relativity, time is delayed where gravity is strong, that is, at low altitudes. This extremely slight delay measurement is made possible by a clock with an accuracy of 18 digits, and if there is a difference of several centimeters in height between the two clocks, the time difference can be measured. Therefore, by measuring the time difference from the reference altitude, relativistic geodesy, that is, the measurement of the altitude difference becomes possible. While GNSS can detect the altitude with an accuracy of about 3 to 4 cm, the optical lattice clock can detect the altitude with an accuracy of 5 cm.

[Reference 2-1: "Successful measurement of elevation difference with ultra-high-precision" optical grid clock "-Expected to be new applications beyond the common sense of clocks such as monitoring of volcanic activity-", Japan Science and Technology Agency (JST) ), The University of Tokyo, RIKEN, Geospatial Information Authority of Japan, Advanced Photon Science Alliance, August 16, 2016, [Search on July 12, 2016], Internet (URL: https://www.jst. go.jp/pr/announce/20160816/)]

[Reference 2-2: "Optical lattice clock from Japan, measuring elevation difference by the theory of relativity, expectation for the next generation world standard", Sankei News, October 24, 2016, [Rewa 2 July 12, 2016 Search], Internet (URL: https://www.sankei.com/life/news/161024/lif1610240029-n1.html)]]

[Reference 2-3: "The world's first" optical grid clock "achieves a measurement accuracy of 5 cm above sea level, such as the University of Tokyo", University Journal, August 22, 2016, [Search on July 12, 2016 ], Internet (URL: https://univ-journal.jp/9185/)]]

In the first to third embodiments described above, when the travel locus

data generation units

31, 31a generate travel locus data based on the quasi-zenith satellite observation data 41, for example, MADOCA (Multi-GNSS Advanced Demonstration tool for) A correction algorithm for correcting the positioning data of the quasi-zenith satellite observation data 41 such as Orbit and Clock Analysis) (Reference 3-1) and the QZS quasi-static method (Reference 3-2) may be used.

The network type RTK-GNSS positioning method described in the second and third embodiments may be applied to the point cloud data collection system α in the first embodiment. In this case, the travel locus

data generation units

31, 31a should use a correction algorithm for correcting the positioning data of the quasi-zenith satellite observation data 41 such as the positioning process of the network type RTK-GPS positioning method (Reference 3-3). You may do it. By using the correction algorithm for correcting these positioning data, the estimation accuracy of the positions, speeds, and traveling directions of the

vehicles

1, 1a, 1b performed by the traveling locus

data generation units

31, 31a is improved.

As a correction method for removing noise from point cloud data 43, which is three-dimensional data (3D data), for example, point cloud data processing software (Reference 3-4) that performs general-purpose processing, wall noise removal processing (reference). Documents 3-5) and the like are known. Therefore, the point cloud data 43 can be corrected with higher accuracy by applying the correction method for removing the noise of the point cloud data 43 after generating the traveling locus data with high accuracy.

[Reference 3-1: Satoshi Kogure (JAXA), "Application of High Precision Positioning Technology", 13th Critical Software Workshop, January 21, 2016, [Search on July 12, 2016], Internet (URL: https://www.ipa.go.jp/files/000050351.pdf), pages 21-25, "6. High-precision positioning technology (MADOCA development), precision startup clock estimation software for multiple GNSS" Is shown, and on page 34, the time required to reach a horizontal error of 6 cm and a vertical error of 10 cm is shown. ]

[Reference 3-2: Hiroshi Yaku, "Technology Development for High-precision Positioning Correction by Quasi-Zenith Satellite", 40th Geospatial Information Authority of Japan Report Meeting, June 3, 2011, [Search on July 12, 2011 ], Internet (URL: https://www.gsi.go.jp/common/000061189.pdf), pages 15-29 show "(2) Development of high-precision positioning correction technology for surveying". On page 18, the correction information reception / positioning process'QZS quasi-static method'or'QZS-QS (QZS-Quasi Static) method' is shown, and on page 20, "test observation result (standard area) RMS (Root Mean)" is shown. Square) (cm), horizontal: 1.2 to 6.7, vertical: 2.0 to 8.4, (mountainous area) RMS (cm), horizontal: 2.1 to 3.7, vertical: 3.4 ~ 5.3 "is shown. On page 24, it is shown that the permissible margin of error for the comparison of the two measurements with TS in the "Accuracy verification using the actual Michibiki machine" boundary survey is 5 mm. ]

[Reference 3-3: "Technical Development by Generating and Distributing High-precision Positioning Correction Information by Quasi-Zenith Satellite: Detailed Algorithm Manual", Geographical Survey Institute, Ministry of Land, Infrastructure, Transport and Tourism, March 25, 2008, [Reiwa 2nd year] Search on July 12], Internet (URL: https://www.gsi.go.jp/common/000065384.pdf), page 61 shows "4.4 Network-type RTK-GPS positioning method positioning process" Has been done. ]

[Reference 3-4: Keigen, Kubota, Keisuke Okamoto, "Examination of Road Maintenance System Using 3D Point Cloud Data by Ground Laser Scanner", Information Processing Society of Japan, 80th National Convention, 1ZC-01, p.4-539-540, 2018, [Search on July 12, 2nd year of Reiwa], Internet (URL: https://www.ipsj.or.jp/event/taikai/80/ipsj_web2018/data/pdf /1ZC-01.html), Fig. 1 shows the definition of the road maintenance system, point cloud data processing software (general-purpose processing). ]

[Reference 3-5: Shigenori Tanaka, Ryuichi Imai, Kenji Nakamura, Kohei Kono, "Study on Automatic Generation of 3D Models Using Point Cloud Coordinate Data", Intelligence and Knowledge (Journal of the Japan Intelligent Information Fuzzy Society), Vol. .23, No.4, pp.572-590, 2011, [Search on July 12, 2nd year of Reiwa], Internet (URL: http://www.nilim.go.jp/lab/qbg/ronbun/ H23_fajii01.pdf), “Wall noise removal processing” is shown in the processing flow of the proposed method in FIG. 6 on page 202. ]

The GPS satellites 10-1, 10-2 and the quasi-zenith satellite 11 are not used in the first embodiment described above, and the GPS satellites 10-1, 10-2 and the quasi-zenith satellite 11 are used in the second and third embodiments. , The location information service provider server device 50 may not be used, and instead, the camera shake detection function in the camera shake correction by a digital camera or a video camera may be used (for example, References 4-1 and 4-2, 4). See -3, 4-4). In this case, the

vehicles

1, 1a and 1b include, for example, an image pickup unit such as a video camera and a video analysis unit that analyzes an image obtained from the image pickup unit. When the image analysis unit detects from the image obtained from the image pickup unit that the trajectory of the

vehicles

1, 1a, 1b is bent or fluctuates up and down, it responds to small displacements, for example, camera shake. It is possible to detect displacement information such as vibration, rotation, and movement amount. In this case, the analysis speed of the video analysis unit can be arbitrarily adjusted, and the time interval of the analysis of the video analysis unit is shorter than the observation time interval of 1 second in the first to third embodiments. For example, since it is possible to match the measurement time interval of the laser radar device 21, it is possible to detect the displacement information with high accuracy. By using the detected displacement information for the correction of the collected data 44, it is possible to specify the detailed horizontal and vertical displacement amounts of the laser radar device 21 and the like. By using the displacement amount of the specified laser radar device 21, the reliability of the point cloud data can be increased, and the point cloud data with the increased reliability can be utilized in the station design. The camera shake detection function may be added to the configuration of the first to third embodiments. With the camera shake detection function, it is possible to obtain displacement information at intervals shorter than the observation time as described above, and the observation time. It can be used as information that complements the interval between the above, and it becomes possible to estimate the movement of the

vehicles

1, 1a, 1b during the observation time with higher accuracy.

In the first to third embodiments, when the laser radar device 21 is mounted on the

vehicles

1, 1a, 1b, vibration is suppressed in order to prevent the influence of displacement when traveling in a bent manner or traveling up and down. A configuration provided on a pedestal such as a hexapod or a gimbal may be applied. Hexapods, gimbals, etc. are equipped with correction control technology that suppresses vibration based on camera images and sensor information that detects acceleration and tilt, and maintains the level of the camera (for example, References 4-2, 4-5). , 4-6). Therefore, by providing a pedestal such as a hexapod or a gimbal in the

vehicles

1, 1a, 1b and attaching the laser radar device 21 to the pedestal, the vibration is suppressed. Therefore, since the direction of the rotation axis of the laser radar device 21 can be maintained in a constant direction, the accuracy of specifying the position of each point in the point cloud data can be improved. In this configuration, the direction of the rotation axis of the laser radar device 21 is maintained in a constant direction by the hexapod, gimbal, etc., so that the traveling directions of the

vehicles

1, 1a, 1b and the direction of the laser radar device 21 Displacement occurs between them. Therefore, it is necessary to correct the collected data 44 in the first to third embodiments in consideration of the displacement that occurs. In addition, when the tilt exceeds the adjustment amount adjusted by the hexapod, gimbal, etc., or when the tilt is continuously adjusted, for example, when the tilt adjustment amount is small in a state close to horizontal. When the tilt is added and the adjustment cannot be performed, the laser radar device 21 cannot be maintained horizontally. In that case, the correction of the collected data according to the configurations of the first to third embodiments is utilized. Become.

[Reference 4-1: Yasuaki Serita, "Image Stabilizer for Digital Cameras", Optics, Vol. 33, No. 9, p.550 (26) -p.555 (31), 2004, [Reiwa 2nd Year 7] Search on 12th of March], Internet (URL: https://annex.jsap.or.jp/photonics/kogaku/public/33-09-kaisetsu5.pdf), Table 1 on page 551 shows the main camera shake detection technologies. "Image motion vector detection method: Image data is continuously read from the image sensor and blur is detected from the moving state of the image between each image". ]

[Reference 4-2: "Image Stabilizer", Wikipedia, [Search on July 12, 2nd year of Reiwa], Internet (URL: https://ja.wikipedia.org/wiki/Image Stabilizer), "Electronics Formula: Often installed in digital cameras and digital video cameras. The shootable area is narrowed to a certain size, the image is read into the buffer memory at the time of shooting, and the first shot image is taken after that. Comparison, calculation of the amount of movement, shooting and recording by automatically shifting the shootable area. Exterior type: Gimbal: Gimbal is a stabilizer that uses the principle of gyroscope and yajirobee. Horizontal of the attached camera. Because it keeps the image, it is possible to reduce blurring during moving shooting.]

[Reference 4-3: Kazuki Nishi, "Digital Movie Blur Measurement / Correction Evaluation System", Telecommunications University New Technology Briefing Session, May 13, 2008, [Search on July 12, 2008], Internet (URL: https://shingi.jst.go.jp/past_abst/abst/p/08/802/uec5.pdf), on page 10 as statistical processing of camera shake detection results, "3D trajectory of camera shake ⇒ many sheets Plot the camera shake detection results for the captured image + Principal component analysis (PCA) ⇒ Dispersion diagram of camera shake detection data and an ellipsoid showing the tendency "is shown. ]

[References 4-4: Hiroshi Maruyama, Yoshiki Yamaguchi, Yuetsu Kodama (University of Tsukuba), "Study on Image Stabilization Mechanism by FPGA", FIT2014 (13th Information Science and Technology Forum), Volume 1, RC-004, p23 -p28, 2014, [Search on July 12, 2014], Internet (URL: https://www.ipsj.or.jp/award/9faeag0000004eyo-att/RC-004.pdf), on page 1. "2. Optical flow detection and image stabilization method (Fig. 1. Difference between in-frame optical flow and image stabilization), 2.1. Frame movement amount estimation method (Fig. 2. Frame movement amount estimation method)" is shown. ing. ]

[Reference 4-5: Stephen Boldran, Scott Jordan, "Image Stabilization and Improvement of Image Resolution by Pietrical Hexapod Motion System", feature Imaging, Laser Focus World Japan, p.26-29, 2017 January, [Search on July 12, 2nd year of Reiwa], Internet (URL: http://ex-press.jp/wp-content/uploads/2017/01/LFWJ1701ft4.pdf), "Hexapod" on page 29 Drive types and applications: "PI H-900KSCO" Hexapod for industrial automation and motion stabilization of moving vehicles and ships, in addition to the image stabilization test of drone cameras using H-840 Hexapod. Is also designed. " The "PI H-900KSCO" hexapod has a maximum motion range of 200 mm (X, Y, Z axis direction) and 66 ° (pitch, yaw, roll) while supporting a load of up to 130 pounds (about 59 kg). It is provided at speeds of 80 mm / s and 30 ° / s, respectively. " ]

[Reference 4-6: Kazuhiro Kawamura, "You can easily take a movie-like smooth and camera-free image with one hand! Divert the drone camera and aerial photography technology! 3-axis electric stabilizer-equipped camera" Osmo " Appearance ", Price.com Magazine, October 26, 2015, [Search on July 12, 2015], Internet (URL: https://kakakumag.com/camera/?id=3477)," Camera It rotates with a muzzle, detects tilt and orientation, and keeps it horizontal. "]]

In the first to third embodiments described above, in steps Sa8 and Sb8 shown in FIGS. 8 and 13, the running

state data

42 and 42a are applied to the estimated position data, the estimated measurement position interval data and the estimated running state data. Therefore, the estimated position data, the estimated measurement position interval data, and the estimated running state data are generated again, but the configuration of the present invention is not limited to the embodiment. In the processing of steps Sa3 and Sb3, the estimated position data, the estimated measurement position interval data and the estimated traveling state data are generated based on the traveling locus data and the traveling

state data

42 and 42a, and the processing of steps Sa8 and Sb8 is performed. You may not have it. In this case, since the processing for generating the estimated position data, the estimated measurement position interval data, and the estimated running state data is performed once, the processing amount can be reduced. On the other hand, in the determination process of steps Sa4 and Sb4, the difference between the estimated running state data to be compared and the running state data is reduced. Therefore, in steps Sa4 and Sb4, the traveling

locus analysis units

32 and 32a may often determine "Yes". Therefore, from the viewpoint of the accuracy of the estimated position data, the estimated measurement position interval data, and the estimated running state data finally obtained, the process of steps Sa8 and Sb8 is not collectively performed as the process of steps Sa3 and Sb3. 8 and the process shown in FIG. 13 are superior. The estimated position data, the estimated measurement position interval data, and the estimated traveling state data may be generated only from the traveling locus data, and the steps Sa8 and Sb8 may not be performed.

In the first to third embodiments described above, the collected data 44 includes the measurement time, the position of the laser radar device 21, the direction including the horizontal component and the vertical component of the laser radar device 21, and the measurement position interval. , The speed of the

vehicles

1, 1a, 1b and the traveling direction of the

vehicles

1, 1a, 1b are included, but the configuration of the present invention is not limited to the embodiment. Since it is sufficient that the position of each point in the point group data 43 in the three-dimensional space can be specified, in the first embodiment, the collected data 44 includes at least the measurement time, the position of the laser radar device 21, and the laser radar device 21. The direction of the horizontal component may be included, in which case the travel locus analysis unit 32 of the first embodiment estimates the position of the vehicle 1 on the horizontal plane at each measurement time. Next, the traveling locus analysis unit 32 estimates only the traveling direction of the vehicle 1 on the horizontal plane at each measurement time as the traveling state of the vehicle 1. The measurement condition generation unit 33 may generate measurement condition data including the measurement time, the position of the laser radar device 21, and the direction of the horizontal component of the laser radar device 21. In the second and third embodiments, the collected data 44 may include at least the measurement time, the position of the laser radar device 21, and the directions of the horizontal and vertical components of the laser radar device 21. In this case, the traveling locus analysis unit 32a of the second and third embodiments first estimates the position of the vehicle 1a at each measurement time. Next, the traveling locus analysis unit 32a estimates only the traveling direction of the vehicle 1a on the horizontal plane and on the vertical surface at each measurement time as the traveling state of the vehicle 1a. The measurement condition generation units 33a and 33b may generate measurement condition data including the measurement time, the position of the laser radar device 21, and the directions of the horizontal component and the vertical component of the laser radar device 21.

The

data correction units

3, 3a, 3b in the point cloud

data collection devices

2, 2a, 2b of the first to third embodiments may be configured as a single device, the data correction device.
In the first to third embodiments described above, the vehicle 1 traveling on the road has been described as an example, but the vehicle 1 is not limited to the vehicle 1, and a moving body such as a drone may be used. When configured in this way, a moving object such as a drone is provided with a point cloud data collecting device 2.

The

data correction units

3, 3a, 3b in the above-described embodiment may be realized by a computer. In that case, a program for realizing this function may be recorded on a computer-readable recording medium, and the program recorded on the recording medium may be read by a computer system and executed. The term "computer system" as used herein includes hardware such as an OS and peripheral devices. Further, the "computer-readable recording medium" refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, or a CD-ROM, and a storage device such as a hard disk built in a computer system. Further, a "computer-readable recording medium" is a communication line for transmitting a program via a network such as the Internet or a communication line such as a telephone line, and dynamically holds the program for a short period of time. It may also include a program that holds a program for a certain period of time, such as a volatile memory inside a computer system that is a server or a client in that case. Further, the above program may be for realizing a part of the above-mentioned functions, and may be further realized for realizing the above-mentioned functions in combination with a program already recorded in the computer system. It may be realized by using a programmable logic device such as FPGA (Field Programmable Gate Array).

As described above, the embodiment of the present invention has been described in detail with reference to the drawings, but the specific configuration is not limited to this embodiment, and the design and the like within a range not deviating from the gist of the present invention are also included.

It can be applied to MMS that collects point cloud data.

1 ... Vehicle, 2 ... Point group data collection device, 3 ... Data correction unit, 4 ... Direction of laser radar device, 10-1, 10-2 ... GPS satellite, 11 ... Quasi-zenith satellite, 21 ... Laser radar device, 22 ... Satellite radio wave receiving antenna, 23 ... Information receiving unit, 24 ... Driving state measurement unit, 25 ... Storage unit, 26 ... Collected data generation unit, 31 ... Travel locus data generation unit, 32 ... Travel locus analysis unit, 33 ... Measurement Condition generation unit, 34 ... Road traffic information acquisition unit, 35 ... Travel locus data normality determination unit, 36 ... Correction processing unit

Claims

When the laser radar device mounted on the moving body measures the distance to the object to be measured by irradiating the laser beam at each measurement time and generates the point group data, it is generated in association with the point group data. This is a data correction method for correcting the collected data indicating the measurement conditions of the laser radar device estimated from the horizontal plane position data indicating the position of the moving object on the horizontal plane obtained at each observation time. ,
The horizontal plane position data with higher accuracy than the horizontal plane position data obtained at each observation time is acquired, or the vertical plane position data indicating the position of the moving object on the vertical plane is acquired and the acquired height is obtained. A movement locus data generation data generation step that generates movement locus data indicating the movement locus of the moving body based on accurate horizontal plane position data or the vertical plane position data.
A movement locus analysis step that analyzes the movement locus data and estimates the position and movement state of the moving body at each measurement time, and a movement locus analysis step.
A measurement condition generation step for generating measurement condition data indicating measurement conditions of the laser radar device for each measurement time based on the position and moving state of the moving body at each measurement time.
A correction processing step for correcting the collected data based on the measurement condition data, and
Data correction method including.
The high-precision horizontal plane position data is data obtained at the same time interval as the observation time interval, and has less error than the horizontal plane position data used when generating the collected data. Or, it is horizontal plane position data obtained at a time interval shorter than the observation time interval.
The data correction method according to claim 1.
When the laser radar device is mounted at an angle with respect to the moving body,
In the measurement condition generation step, the measurement condition data for each measurement time is generated based on the position and the moving state of the moving body and the angle of inclination.
The data correction method according to claim 1 or 2.
When the collected data is corrected in the correction processing step, a reliability index indicating a high degree of reliability is given to the point cloud data corresponding to the corrected collected data.
The data correction method according to any one of claims 1 to 3.
When it is determined in the movement locus analysis step that the moving body is moving at a constant speed and traveling straight at the measurement time, the measurement condition data is not generated in the measurement condition generation step.
When it is determined in the movement locus analysis step that the moving body is not moving at a constant speed or is not traveling straight at the measurement time, the measurement condition data is generated in the measurement condition generation step.
The data correction method according to any one of claims 1 to 4.
In the movement locus analysis step, the estimated movement state data indicating the estimated movement state of the moving body is generated, and the movement state data measured and generated by the movement state measuring unit provided inside the moving body is the estimation. When the difference between the movement state data corresponding to the measurement time of the movement state data and the estimated movement state data is within a predetermined range, the measurement is performed based on the movement trajectory data and the movement state data. Estimate the position and moving state of the moving body for each time,
The data correction method according to any one of claims 1 to 5.
When the difference between the movement state data and the estimated movement state data is not within a predetermined range, the movement trajectory data is generated based on the road traffic information acquired by the road traffic information acquisition unit provided inside the moving body. The data correction method according to claim 6, further comprising a movement locus data normality determination step of determining whether or not the data is normal, and if it is determined that the abnormality is not normal, an abnormality is output to the outside.
When the laser radar device mounted on the moving body measures the distance to the object to be measured by irradiating the laser beam at each measurement time and generates the point group data, it is generated in association with the point group data. It is a data correction device that corrects the collected data indicating the measurement conditions of the laser radar device estimated from the horizontal plane position data indicating the position of the moving object on the horizontal plane obtained at each observation time. ,
The horizontal plane position data with higher accuracy than the horizontal plane position data obtained at each observation time is acquired, or the vertical plane position data indicating the position of the moving object on the vertical plane is acquired and the acquired height is obtained. A movement locus data generation unit that generates movement locus data indicating the movement locus of the moving body based on accurate horizontal plane position data or the vertical plane position data.
A movement locus analysis unit that analyzes the movement locus data and estimates the position and movement state of the moving body at each measurement time.
A measurement condition generation unit that generates measurement condition data indicating measurement conditions of the laser radar device for each measurement time based on the position and movement state of the moving body at each measurement time.
A correction processing unit that corrects the collected data based on the measurement condition data,
A data correction device equipped with.