JP7770696B2 - System, measuring device, map data generating device, map data generating method and program - Google Patents

Description

The present invention relates to a system, a measurement device, a map data generation device, a map data generation method, and a program.

Patent Document 1 describes an automated driving system that realizes stable automated driving in various environments by appropriately using satellite positioning and SLAM (Simultaneous Localization and Mapping).
[Prior art documents]
[Patent Documents]
[Patent Document 1] JP 2022-146457 A

According to one embodiment of the present invention, a system is provided. The system may include a measurement device including a LiDAR (Light Detection and Ranging), an IMU (Inertial Measurement Unit), a base having a mounting portion on which the IMU is mounted and a support portion supporting the LiDAR, and a motor for rotating the base, wherein a support surface on which the support portion supports the LiDAR is perpendicular to a mounting surface on which the mounting portion mounts the IMU, and the center of the IMU passes through an axis of rotation of the LiDAR when the base portion rotates due to the mechanical energy of the motor, or wherein a support surface on which the support portion supports the LiDAR is parallel to the mounting surface on which the mounting portion mounts the IMU, and the center of the IMU passes through a line perpendicular to the axis of rotation of the LiDAR when the base portion rotates due to the mechanical energy of the motor. The system may include a first acquisition unit that acquires a plurality of first scan data measured by the LiDAR during a predetermined period while the base unit is rotating and not translating due to the mechanical energy of the motor, and motion data of the LiDAR measured by the IMU during the predetermined period when the LiDAR measured each of the plurality of first scan data, to generate reference data to be referenced when generating map data using a SLAM algorithm. The system may include a reference data generation unit that generates the reference data based on the plurality of first scan data and the motion data of the LiDAR measured by the LiDAR during the predetermined period. The system may include a second acquisition unit that acquires a plurality of second scan data measured by the LiDAR while the base unit is rotating and translating due to the mechanical energy of the motor. The system may include a map data generation unit that generates the map data based on the reference data generated by the reference data generation unit and the plurality of second scan data.

In the system, the LiDAR may output a first laser light whose output direction is the direction of a first vector contained in a plane perpendicular to the support surface, and a second laser light whose output direction is the direction of a second vector contained in a plane perpendicular to the support surface and opposite to the direction of the first vector, and the first acquisition unit may acquire the plurality of first scan data measured during the predetermined period that is equal to or longer than the period for the LiDAR to rotate half a revolution around the rotation axis, and the movement data of the LiDAR when the LiDAR measured each of the first scan data.

In any of the above systems, the reference data generation unit may generate the reference data by identifying the rotation angle of the LiDAR when the LiDAR measured each of the first scan data based on the movement data of the LiDAR when the LiDAR measured each of the first scan data, and integrating each of the first scan data based on the rotation angle of the LiDAR when the LiDAR measured each of the first scan data.

In any of the above systems, the reference data generation unit may generate the reference data by integrating the first scan data based on the results of aligning the first scan data using an ICP (Iterative Closest Point) algorithm.

In any of the above systems, the second acquisition unit may further acquire movement data of the LiDAR measured by the IMU when the LiDAR measured each of the second scan data of the plurality of second scan data, and the system may further include a grouping unit that groups the plurality of second scan data based on the movement data of the LiDAR when the LiDAR measured each of the second scan data of the plurality of second scan data, and the map data generation unit may generate the map data by aligning the second scan data grouped into the same group by the grouping unit with the reference data.

In any of the above systems, the LiDAR may output a first laser beam having an output direction aligned with a first vector contained in a plane perpendicular to the support surface, and a second laser beam having an output direction aligned with a second vector contained in a plane perpendicular to the support surface and having an opposite direction to the first vector, and the grouping unit may group the plurality of second scan data by grouping the second scan data measured during a period in which the LiDAR makes one-half rotation around the rotation axis into the same group.

In any of the above systems, the reference data generation unit may update the reference data based on the second scan data grouped into the same group by the grouping unit, and the map data generation unit may generate the map data based on the reference data updated by the reference data generation unit and the plurality of second scan data.

Any of the above systems may further include a pose graph generation unit that generates a pose graph that represents the trajectory of the measuring device in a graph structure for each group based on the reference data generated by the reference data generation unit and the plurality of second scan data grouped by the grouping unit, and the map data generation unit may generate the map data further based on the pose graph generated by the pose graph generation unit.

In any of the above systems, the second acquisition unit may further acquire movement data of the LiDAR measured by the IMU when the LiDAR measured each of the second scan data of the plurality of second scan data, and the map data generation unit may generate the map data further based on the movement data of the LiDAR when the LiDAR measured each of the second scan data of the plurality of second scan data.

According to one embodiment of the present invention, a measurement device is provided. The measurement device may include a LiDAR. The measurement device may include an IMU. The measurement device may include a base having a mounting portion on which the IMU is mounted and a support portion that supports the LiDAR. The measurement device may include a motor that rotates the base. The support surface on which the support portion supports the LiDAR may be perpendicular to the mounting surface on which the mounting portion mounts the IMU, and the center of the IMU may pass through the axis of rotation of the LiDAR when the base portion rotates due to the mechanical energy of the motor, or the support surface on which the support portion supports the LiDAR may be parallel to the mounting surface on which the mounting portion mounts the IMU, and the center of the IMU may pass through a line perpendicular to the axis of rotation of the LiDAR when the base portion rotates due to the mechanical energy of the motor.

The measurement device may further include a first acquisition unit that acquires a plurality of first scan data measured by the LiDAR during a predetermined period while the base portion is rotating and not translating due to the mechanical energy of the motor, and movement data of the LiDAR when the LiDAR measured each of the plurality of first scan data measured by the IMU during the predetermined period, in order to generate reference data that is referenced when generating map data using a SLAM algorithm; a reference data generation unit that generates the reference data based on the plurality of first scan data and the movement data of the LiDAR when the LiDAR measured each of the first scan data; a second acquisition unit that acquires a plurality of second scan data measured by the LiDAR while the base portion is rotating and translating due to the mechanical energy of the motor; and a map data generation unit that generates the map data based on the reference data generated by the reference data generation unit and the plurality of second scan data.

According to one embodiment of the present invention, there is provided a map data generation device that generates map data using a SLAM algorithm based on measurement data measured by a measurement device, the map data generation device comprising: a LiDAR; an IMU; a base unit having a mounting unit on which the IMU is mounted and a support unit that supports the LiDAR; and a motor that rotates the base unit, wherein the support surface on which the support unit supports the LiDAR is perpendicular to the mounting surface on which the mounting unit mounts the IMU, and the center of the IMU passes through the axis of rotation of the LiDAR when the base unit rotates due to the mechanical energy of the motor; or the support surface on which the support unit supports the LiDAR is parallel to the mounting surface on which the mounting unit mounts the IMU, and the center of the IMU passes through a line perpendicular to the axis of rotation of the LiDAR when the base unit rotates due to the mechanical energy of the motor. The map data generation device may include a first acquisition unit that acquires a plurality of first scan data measured by the LiDAR during a predetermined period while the base unit is rotating and not translating due to the mechanical energy of the motor, and motion data of the LiDAR measured by the IMU during the predetermined period when the LiDAR measured each of the plurality of first scan data, in order to generate reference data to be referenced when generating the map data. The map data generation device may include a reference data generation unit that generates the reference data based on the plurality of first scan data and the motion data of the LiDAR measured by the LiDAR during the predetermined period. The map data generation device may include a second acquisition unit that acquires a plurality of second scan data measured by the LiDAR while the base unit is rotating and translating due to the mechanical energy of the motor. The map data generation device may include a map data generation unit that generates the map data based on the reference data generated by the reference data generation unit and the plurality of second scan data.

According to one embodiment of the present invention, there is provided a map data generation method executed by a measurement device comprising: a LiDAR; an IMU; a base having a mounting portion on which the IMU is mounted and a support portion supporting the LiDAR; and a motor for rotating the base, wherein the support surface on which the support portion supports the LiDAR is perpendicular to the mounting surface on which the mounting portion mounts the IMU, and the center of the IMU passes through the axis of rotation of the LiDAR when the base portion rotates due to the mechanical energy of the motor; or the support surface on which the support portion supports the LiDAR is parallel to the mounting surface on which the mounting portion mounts the IMU, and the center of the IMU passes through a line perpendicular to the axis of rotation of the LiDAR when the base portion rotates due to the mechanical energy of the motor. The map data generation method may include a first acquisition step of acquiring a plurality of first scan data measured by the LiDAR during a predetermined period while the base portion is rotating due to the mechanical energy of the motor and is not moving in a translational manner, and motion data of the LiDAR measured by the IMU during the predetermined period when the LiDAR measured each of the plurality of first scan data, to generate reference data to be referenced when generating map data using a SLAM algorithm. The map data generation method may include a reference data generation step of generating the reference data based on the plurality of first scan data and the motion data of the LiDAR measured by the LiDAR during the predetermined period. The map data generation method may include a second acquisition step of acquiring a plurality of second scan data measured by the LiDAR while the base portion is rotating due to the mechanical energy of the motor and is moving in a translational manner. The map data generation method may include a map data generation step of generating the map data based on the reference data generated in the reference data generation step and the plurality of second scan data.

According to one embodiment of the present invention, a program for causing a computer to execute the map data generation method is provided.

According to one embodiment of the present invention, there is provided a map data generation method executed by a map data generation device that generates map data using a SLAM algorithm based on measurement data measured by a measurement device, the map data generation method comprising: a LiDAR; an IMU; a base unit having a mounting unit on which the IMU is mounted and a support unit that supports the LiDAR; and a motor that rotates the base unit, wherein the support surface on which the support unit supports the LiDAR is perpendicular to the mounting surface on which the mounting unit mounts the IMU, and the center of the IMU passes through the axis of rotation of the LiDAR when the base unit rotates due to the mechanical energy of the motor; or the support surface on which the support unit supports the LiDAR is parallel to the mounting surface on which the mounting unit mounts the IMU, and the center of the IMU passes through a line perpendicular to the axis of rotation of the LiDAR when the base unit rotates due to the mechanical energy of the motor. The map data generation method may include a first acquisition step of acquiring a plurality of first scan data measured by the LiDAR during a predetermined period while the base portion is rotating and not translating due to the mechanical energy of the motor, and motion data of the LiDAR measured by the IMU during the predetermined period when the LiDAR measured each of the plurality of first scan data, to generate reference data to be referenced when generating the map data. The map data generation method may include a reference data generation step of generating the reference data based on the plurality of first scan data and the motion data of the LiDAR measured by the LiDAR during the predetermined period. The map data generation method may include a second acquisition step of acquiring a plurality of second scan data measured by the LiDAR while the base portion is rotating and translating due to the mechanical energy of the motor. The map data generation method may include a map data generation step of generating the map data based on the reference data generated in the reference data generation step and the plurality of second scan data.

According to one embodiment of the present invention, a program for causing a computer to execute the map data generation method is provided.

Note that the above summary of the invention does not list all of the necessary features of the present invention. Also, subcombinations of these features may also constitute inventions.

1 shows an example of a hardware configuration of a measurement device 100. 2 shows another example of the hardware configuration of the measurement device 100. FIG. 4 is an explanatory diagram illustrating an example of a processing flow for generating map data. FIG. 10 is an explanatory diagram illustrating an example of a process for generating reference data. FIG. 10 is an explanatory diagram illustrating an example of a relationship between reference data and second scan data. FIG. 10 is an explanatory diagram illustrating an example of a process for aligning scan data with reference data. 1 shows a schematic diagram of an example of a pose graph for the measurement device 100. 10A and 10B show an example of a pose graph of the measuring device 100 before pose adjustment and map data corresponding to the pose graph of the measuring device 100 before pose adjustment. 10A and 10B show an example of a pose graph of the measurement device 100 before and after pose adjustment. 10A and 10B show an example of a pose graph of the measuring device 100 after the pose adjustment and map data corresponding to the pose graph of the measuring device 100 after the pose adjustment. FIG. 10 is an explanatory diagram illustrating another example of the flow of the process for generating map data. 1 illustrates an example of a system 10 in schematic form. 2 shows an example of a functional configuration of the measurement device 100. 2 shows an example of a functional configuration of a map data generating device 200. 1 shows an example of a hardware configuration of a computer 1200 that functions as the measurement device 100 or the map data generating device 200.

The present invention will be described below through embodiments of the invention, but the following embodiments do not limit the invention as defined by the claims. Furthermore, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

Figure 1 shows an example of the hardware configuration of the measuring device 100. The measuring device 100 includes a box unit 110, a base unit 120, a LiDAR 130, an IMU 140, a motor 150, and a slip ring 160. The measuring device 100 may further include a processing unit such as a CPU (Central Processing Unit) (not shown). Note that it is not essential for the measuring device 100 to include all of these components.

The box section 110 houses various devices. For example, the box section 110 houses a motor 150 that rotates the base section 120. The box section 110 houses, for example, a slip ring 160 that connects the base section 120 and the motor 150.

A base unit 120 may be disposed on the upper surface of the box unit 110. For example, the surface on which the base unit 120 is disposed on the box unit 110 is a surface parallel to the horizontal plane. Figure 1 illustrates an example in which the surface on which the base unit 120 is disposed is a surface parallel to the xy plane. Note that the xyz coordinate system in Figure 1 is the coordinate system of the measurement device 100.

The box unit 110 may have a hand grip (not shown). A user of the measuring device 100 may move the measuring device 100 by holding the measuring device 100 by the hand grip.

The base unit 120 may have a mounting unit 122. The mounting unit 122 may mount the IMU 140. For example, the mounting surface of the mounting unit 122 on which the IMU 140 is mounted is a plane parallel to the placement surface of the base unit 120. Figure 1 illustrates an example in which the mounting surface of the mounting unit 122 is a plane parallel to the xy plane.

The mounting surface of the mounting portion 122 protrudes, for example, from the bottom surface 121 of the base portion 120. The mounting surface of the mounting portion 122 does not have to protrude from the bottom surface 121. Figure 1 shows an example in which the mounting surface of the mounting portion 122 protrudes from the bottom surface 121 in the positive direction of the z-axis.

The mounting portion 122 and the bottom portion 121 may be integrated. The mounting portion 122 and the bottom portion 121 may also be separate components.

The base unit 120 may have a support unit 124. The support unit 124 may support the LiDAR 130. For example, the support surface on which the support unit 124 supports the LiDAR 130 is perpendicular to the placement surface of the base unit 120. For example, the support surface of the support unit 124 is perpendicular to the bottom surface unit 121. For example, the support surface of the support unit 124 is perpendicular to the mounting surface of the mounting unit 122. Figure 1 illustrates an example in which the support surface of the support unit 124 is perpendicular to the xy plane.

The base unit 120, for example, rotates. The base unit 120, for example, rotates using mechanical energy obtained from the motor 150 via the slip ring 160.

The base unit 120 rotates, for example, clockwise around the rotation axis. The base unit 120 rotates, for example, counterclockwise around the rotation axis. Figure 1 shows an example in which the base unit 120 rotates in the xy plane around a rotation axis parallel to the z axis.

The base unit 120 rotates at a predetermined rotational speed, for example. The base unit 120 rotates at a rotational speed of 30 rpm (revolutions per minute), for example. The base unit 120 may rotate at any other rotational speed.

LiDAR130 may output laser light and measure the distance between the output position of the laser light and the reflection point based on the reflected light of the laser light reflected at the reflection point. For example, if the propagation speed of the laser light is c [m/s], and the time from the time the laser light is output to the time the reflected light is received is t [s], and the distance between the output position of the laser light and the reflection point is x [m], LiDAR130 measures the distance between the output position of the laser light and the reflection point using x [m] = 1/2 × t [s] × c [m/s].

LiDAR130 may acquire point cloud data consisting of a collection of reflection points by measuring the distance between the output position of the laser light and each reflection point. Here, acquiring point cloud data by LiDAR130 is considered to be performing a scan by LiDAR130. Note that point cloud data may also be referred to as scan data.

LiDAR130 performs scans at, for example, a predetermined scan interval. LiDAR130 performs scans at, for example, a scan interval of 100 ms. LiDAR130 may also perform scans at any other scan interval.

LiDAR 130 outputs laser light from, for example, output surface 135. LiDAR 130 outputs multiple laser light beams from, for example, output surface 135. LiDAR 130 outputs, for example, a first laser light beam whose output direction is the direction of a first vector contained in a plane perpendicular to the support surface of support portion 124, and a second laser light beam whose output direction is the direction of a second vector contained in a plane perpendicular to the support surface of support portion 124 and has a direction opposite to that of the first vector.

LiDAR 130 has, for example, a cylindrical shape. In this case, LiDAR 130 outputs laser light from output surface 135, which is the side surface of the cylinder. LiDAR 130 may also have any other shape.

LiDAR130 includes, for example, a laser light source that outputs laser light with a wavelength in the infrared region. LiDAR130 includes, for example, a laser light source that outputs laser light with a wavelength in the visible light region. LiDAR130 may also include a laser light source that outputs laser light with a wavelength in the ultraviolet region.

The LiDAR 130 may rotate integrally with the base 120, which rotates using the mechanical energy of the motor 150. Therefore, the LiDAR 130 may be a rotating LiDAR.

IMU 140 measures the movement data of LiDAR 130. For example, IMU 140 measures the movement data of LiDAR 130 when LiDAR 130 measures scan data.

The IMU 140 measures, for example, motion data of the LiDAR 130 indicating the rotational motion of the LiDAR 130. The IMU 140 measures, for example, motion data of the LiDAR 130 indicating the translational motion of the LiDAR 130. The IMU 140 measures, for example, motion data of the LiDAR 130 indicating both the rotational motion and translational motion of the LiDAR 130.

For example, the center of IMU 140 passes through the rotation axis _lr of LiDAR 130 when base unit 120 rotates due to the mechanical energy of motor 150. For example, the support surface of support unit 124 is perpendicular to the mounting surface of mounting unit 122, and the center of IMU 140 passes through the rotation axis lr of LiDAR 130. Figure 1 illustrates an example where the center of IMU 140 passes through the rotation axis _lr parallel to the z-axis.

For example, the center line l _c passing through the center of the IMU 140 is parallel to the rotation axis l _r . Fig. 1 illustrates an example in which the center line l _c coincides with the rotation axis l _r .

The measuring device 100 may further include a camera (not shown). The camera may be mounted at any position on the measuring device 100. The camera is, for example, an RGB camera. The camera may also be an infrared camera.

The measuring device 100 generates, for example, map data. The measuring device 100 generates, for example, three-dimensional map data.

The measuring device 100 generates map data, for example, based on various acquired data. The measuring device 100 generates map data, for example, based on measured measurement data. The measurement data includes, for example, scan data measured by the LiDAR 130. The measurement data includes, for example, movement data of the LiDAR 130 measured by the IMU 140. The measurement data includes, for example, both scan data and movement data of the LiDAR 130. The measuring device 100 may generate map data based on image data captured by a camera provided in the measuring device 100.

The measuring device 100 generates map data using, for example, the SLAM algorithm. The SLAM algorithm is an algorithm that simultaneously estimates its own position and generates map data using sensors such as LiDAR and cameras. The SLAM algorithm allows a device whose position changes, such as the measuring device 100, to calculate its own trajectory while creating map data, even in an unknown environment where no map data exists, and estimate its own position on the generated map. Details of generating map data using the SLAM algorithm will be discussed later.

The measurement device 100 may generate a pose graph for the measurement device 100. The pose graph for the measurement device 100 is a graph structure that represents the trajectory of the measurement device 100. The measurement device 100 generates the pose graph for the measurement device 100 using, for example, the SLAM algorithm. Details of generating a pose graph for the measurement device 100 using the SLAM algorithm will be described later.

The measuring device 100 may be mounted on a mobile object such as a robot. In this case, the measuring device 100 can be moved without the user of the measuring device 100 having to move the measuring device 100.

A typical conventional LiDAR, in which the laser light output surface is located on the side and the bottom surface is fixedly positioned so that it is parallel to the horizontal plane, has a horizontal FOV (Field of View) of 360 degrees and a vertical FOV of 30 degrees. Because this LiDAR is fixedly positioned and has a narrow vertical FOV, when scanning a structure with a long vertical length, such as a high-rise building, the LiDAR user must manually tilt the LiDAR to expand the LiDAR's vertical FOV. Therefore, when using this LiDAR to scan a measurement environment containing a structure with a long vertical length, the LiDAR user must place a heavy burden on the LiDAR user. Given the above, it is desirable to be able to reduce the burden on the user when using LiDAR to scan a measurement environment containing a structure with a long vertical length.

In contrast, according to the measurement device 100 of this embodiment, the LiDAR 130, which outputs laser light from the output surface 135, is positioned so that its bottom surface is perpendicular to the horizontal plane. In this case, for example, the horizontal FOV of the stationary LiDAR 130 is 30 degrees, and its vertical FOV is 270 degrees. Then, by using the mechanical energy of the motor 150 to rotate the LiDAR 130 around a rotation axis perpendicular to the horizontal plane, the horizontal FOV of the LiDAR 130 can be expanded to 360 degrees without manual operation by the user of the measurement device 100. Thus, by positioning the LiDAR 130 so that its bottom surface is perpendicular to the horizontal plane and so that it can rotate around a rotation axis perpendicular to the horizontal plane, the measurement device 100 of this embodiment can ensure the horizontal FOV of the LiDAR while reducing the burden on the user when scanning a measurement environment containing structures with long vertical lengths using the LiDAR. Furthermore, the measurement device 100 according to this embodiment has a structure in which the center of the IMU 140 passes through the rotation axis of the LiDAR 130, making it possible to measure with high accuracy the movement data of the LiDAR 130, which may include the rotational movement of the LiDAR 130.

Figure 2 shows a schematic diagram of another example of the hardware configuration of the measuring device 100. Differences from the measuring device 100 shown in Figure 1 will be mainly explained here. Note that the box unit 110, motor 150, and slip ring 160 are omitted from Figure 2.

The IMU 140 may be mounted on the support portion 124. The IMU 140 may be mounted, for example, on the support surface of the support portion 124. Note that in Figure 2, the LiDAR 130 is drawn with dotted lines to indicate that the IMU 140 is mounted on the support portion 124.

The mounting portion 122 and the support portion 124 may be integral. The mounting portion 122 and the support portion 124 may also be separate components.

For example, the support section 124 has a gap between the LiDAR 130 and the support surface in which the IMU 140 is mounted, while the LiDAR 130 is supported on the support surface. In this case, the IMU 140 may be mounted within the gap.

For example, the support surface of the support portion 124 may be parallel to the mounting surface of the mounting portion 122. For example, the support surface of the support portion 124 may be flush with the mounting surface of the mounting portion 122. Figure 2 illustrates an example in which the support surface of the support portion 124 is flush with the mounting surface of the mounting portion 122 and is perpendicular to the xy plane.

For example, the center of IMU 140 passes through a line perpendicular to the rotation axis l _r . For example, the support surface of the support unit 124 is parallel to the mounting surface of the mounting unit 122, and the center of IMU 140 passes through a line perpendicular to the rotation axis l _r . For example, the support surface of the support unit 124 is flush with the mounting surface of the mounting unit 122, and the center of IMU 140 passes through a line perpendicular to the rotation axis l _r . Figure 2 illustrates an example where the center of IMU 140 is perpendicular to the rotation axis l _r , which is parallel to the z-axis, and passes through a line parallel to the x-axis.

For example, the center line _{l_c} is perpendicular to the rotation axis _{l_r} . Figure 2 illustrates an example in which the center line _{l_c} is perpendicular to the rotation axis _{l_r} , which is parallel to the z-axis, and is parallel to the x-axis.

Figure 3 is an explanatory diagram illustrating an example of the process flow for generating map data. Here, an example of the process flow for generating map data using the SLAM algorithm is explained in three steps.

In Step 1, reference data is generated that is referenced when generating map data using the SLAM algorithm. The reference data is generated, for example, based on scan data measured by LiDAR 130 when the base unit 120 is rotating due to the mechanical energy of the motor 150 and is not moving in a translational manner. Note that the scan data measured by LiDAR 130 when the base unit 120 is rotating due to the mechanical energy of the motor 150 and is not moving in a translational manner may be referred to as first scan data.

In Step 2, multiple scan data measured by LiDAR 130 while the base unit 120 is rotating and translating due to the mechanical energy of motor 150 are grouped. Note that the scan data measured by LiDAR 130 while the base unit 120 is rotating and translating due to the mechanical energy of motor 150 may be referred to as second scan data.

In Step 3, map data is generated using a SLAM algorithm based on the reference data generated in Step 1 and the multiple second scan data groups grouped in Step 2. For example, map data is generated by aligning the second scan data with the reference data. Here, "align" refers to aligning in the coordinate system of the map data. The coordinate system of the map data may also be referred to as map space.

In Step 3, a pose graph for the measurement device 100 may be further generated using the SLAM algorithm based on the reference data generated in Step 1 and the multiple second scan data groups grouped in Step 2.

Figure 4 is an explanatory diagram illustrating an example of the process of generating reference data. It is assumed that the LiDAR 130 simultaneously outputs two laser beams, a first laser beam and a second laser beam whose output direction is opposite to that of the first laser beam, that the rotation speed of the LiDAR 130 is 30 rpm, the scan interval of the LiDAR 130 is 100 ms, and the horizontal FOV of the stationary LiDAR 130 is 30 degrees.

The range indicated by the dashed line in Figure 4 is the range of the nth scan. The range indicated by the dashed line in Figure 4 is the range of the n+1th scan. n is a natural number. Note that the xyz coordinate system in Figure 4 is the coordinate system of the measurement device 100.

If the rotation speed of the LiDAR 130 is 30 rpm, the scan interval of the LiDAR 130 is 100 ms, and the horizontal FOV of the stationary LiDAR 130 is 30 degrees, the LiDAR 130 rotates 18 degrees around the rotation axis _lr between the nth scan and the n+1th scan. Therefore, the rotation angle of the LiDAR 130 between the nth scan and the n+1th scan is smaller than the horizontal FOV of the stationary LiDAR 130. Therefore, the range of the nth scan and the range of the n+1th scan overlap. As described above, the LiDAR 130 that simultaneously outputs two laser beams can expand its horizontal FOV to 360 degrees by measuring first scan data for a period of time longer than the LiDAR 130 rotates 180 degrees. In FIG. 4, the overlapping range where the range of the nth scan and the range of the (n+1)th scan overlap is indicated by diagonal lines.

The following formula can be used to derive the number of scans that LiDAR130 will perform during the period in which it rotates 180 degrees. Note that the right-hand side of the formula below is a ceiling function that derives the smallest integer greater than or equal to a given real number.

Since the scanning frequency of LiDAR130 is the reciprocal of the scanning interval of LiDAR130, when the scanning interval of LiDAR130 is 100 ms, the scanning frequency of LiDAR130 is 10 Hz. By substituting the rotation speed of LiDAR130 = 30 rpm and the scanning frequency of LiDAR130 = 10 Hz into the above equation, it can be derived that when the rotation speed of LiDAR130 is 30 rpm and the scanning interval of LiDAR130 is 100 ms, the number of scans that LiDAR130 performs during the period in which LiDAR130 rotates 180 degrees is 10.

The measurement device 100 generates reference data, for example, based on multiple first scan data measured by the LiDAR 130 over a predetermined period of time and movement data of the LiDAR 130 measured by the IMU 140 over the predetermined period of time when the LiDAR 130 measured each of the multiple first scan data. The measurement device 100 generates reference data by, for example, determining the rotation angle of the LiDAR 130 when the LiDAR 130 measured each of the first scan data based on the movement data of the LiDAR 130 when the LiDAR 130 measured each of the first scan data, and integrating each of the first scan data based on the rotation angle of the LiDAR 130 when the LiDAR 130 measured each of the first scan data.

The predetermined period may be set so that the horizontal FOV of the LiDAR 130 is 360 degrees. For example, the predetermined period may be set so that it is equal to or greater than the period required for the LiDAR 130, which simultaneously outputs two laser beams, to rotate 180 degrees. For example, if the rotation speed of the LiDAR 130 is 30 rpm, the scan interval of the LiDAR 130 is 100 ms, and the horizontal FOV of the stationary LiDAR 130 is 30 degrees, the predetermined period may be set to 10 seconds, which is longer than the 1 second required for the LiDAR 130 to rotate 180 degrees. In this case, the LiDAR 130 measures the first scan data for five rotations of the LiDAR 130. The predetermined period may be set to 1 second, which is the period required for the LiDAR 130 to rotate 180 degrees. In this case, the LiDAR 130 measures the first scan data while the LiDAR 130 rotates half a revolution.

Here, an example of a process for integrating the first scan data measured the nth time by the LiDAR 130 and the first scan data measured the n+1th time by the LiDAR 130 will be described. If the rotation angle of the LiDAR 130 when the LiDAR 130 measured the nth first scan data is θ _n , and the rotation angle of the LiDAR 130 when the LiDAR 130 measured the n+1th first scan data is θ _n+1 , the measurement device 100 determines that the LiDAR 130 rotated by Δθ = θ _n+1 - θ n between the time when the LiDAR 130 measured the nth first scan data and the time when the LiDAR 130 measured the _n +1th first scan data. Δθ represents the relative rotation angle between the rotation angle of LiDAR 130 when LiDAR 130 measures the first scan data the nth time and the rotation angle of LiDAR 130 when LiDAR 130 measures the first scan data the n+1th time. Next, measurement device 100 aligns the first scan data measured by LiDAR 130 the n+1th time with the first scan data measured by LiDAR 130 the nth time by applying the inverse rotation matrix of Δθ to the first scan data measured by LiDAR 130 the n+1th time. Thereafter, the first scan data measured by LiDAR 130 the nth time and the first scan data measured by LiDAR 130 the n+1th time are integrated.

The LiDAR 130 included in the measurement device 100 is positioned so that its bottom surface is perpendicular to the horizontal plane to widen the vertical FOV. The measurement device 100 rotates the LiDAR 130, thereby expanding the horizontal FOV of the LiDAR 130 to 360 degrees. However, due to the rotational motion of the LiDAR, the overlapping area between two scan data measured adjacently in time by a rotating LiDAR is lower than the overlapping area between two scan data measured adjacently in time by a stationary LiDAR. For a given LiDAR scan interval, the faster the LiDAR rotation speed, the lower the overlapping area between two scan data measured adjacently in time. Note that "two scan data measured adjacently in time" refers to two scan data measured consecutively in time, such as the scan data measured by LiDAR130 the nth time and the scan data measured by LiDAR130 the n+1th time.

For example, if the rotation speed of the LiDAR 130 is 30 rpm and the scan interval of the LiDAR 130 is 100 ms, as described above, the LiDAR 130 rotates 18 degrees around the rotation axis _lr between the nth scan and the n+1th scan. In this case, assuming that the LiDAR 130 is not performing any movement other than rotation, the overlapping range ratio between the range of the nth scan and the range of the n+1th scan is (30 degrees - 18 degrees) ÷ 30 degrees × 100 = 40%.

When generating map data using the SLAM algorithm, an alignment algorithm for aligning scan data, such as the ICP algorithm or the NDT (Normal Distributions Transform) algorithm, is essential. If the overlapping area between scan data is low, the stability and reliability of the alignment algorithm decreases. This decrease in the stability and reliability of the alignment algorithm adversely affects the accuracy of map data generated using the SLAM algorithm. Therefore, in order to generate highly accurate map data using the SLAM algorithm, it is important to increase the stability and reliability of the alignment algorithm. For these reasons, it is desirable to increase the stability and reliability of the alignment algorithm by increasing the overlapping area between scan data measured by a rotating LiDAR.

In contrast, the measuring device 100 according to this embodiment measures multiple first scan data sets using the LiDAR 130 while it is only performing rotational motion prior to the start of movement of the measuring device 100. The measuring device 100 then generates reference data based on the multiple first scan data sets and the movement data of the LiDAR 130 measured by the IMU 140 at the time the LiDAR 130 measured each of the multiple first scan data sets. As described above, the measuring device 100 has a structure in which the center of the IMU 140 passes through the rotation axis of the LiDAR 130. Therefore, the measuring device 100 can accurately determine the relative rotation angles between the LiDAR 130 rotation angles when the LiDAR 130 measured each of the first scan data sets, thereby generating reference data with high accuracy for aligning the first scan data sets. Then, when the alignment algorithm is executed after the measuring device 100 begins movement, the second scan data sets are aligned with the reference data. Because the reference data is generated based on the multiple first scan data measured by LiDAR 130 over a period of time or longer that is set so that the horizontal FOV of LiDAR 130 is 360 degrees, there is a high proportion of overlap between the second scan data and the reference data. Furthermore, because the alignment between the first scan data that make up the reference data is highly accurate, the measuring device 100 can align the second scan data with the highly accurately aligned reference data. As described above, the measuring device 100 according to this embodiment generates reference data prior to the start of movement of the measuring device 100, thereby improving the stability and reliability of the alignment algorithm and contributing to the generation of highly accurate map data using the SLAM algorithm.

Figure 5 is an explanatory diagram illustrating an example of the relationship between the reference data and the second scan data. Here, it is assumed that the second scan data is measured by the LiDAR 130 immediately after the measurement device 100 begins to move.

The range indicated by the solid line in Figure 5 is the range of the reference data. The range indicated by the dotted line in Figure 5 is the range of the second scan data. Note that the xyz coordinate system in Figure 5 is the coordinate system of the measurement device 100.

In the example shown in Figure 5, the range of the reference data encompasses the range of the second scan data. This means that the overlapping range between the reference data and the second scan data shown in Figure 5 is extremely high.

6 is an explanatory diagram illustrating an example of a process for aligning scan data with reference data. Here, an example of a process for aligning scan data with reference data using the ICP algorithm by the measurement device 100 will be mainly described. Note that the "◯" in FIG. 6 represents point _Pr constituting the reference data, and the "▽" in FIG. 6 represents point _Ps constituting the scan data.

6 is an explanatory diagram for explaining an example of a process (sometimes referred to as "Process 1") for matching each of a plurality of points _Ps of the scan data with one of a plurality of points _Pr of the reference data. Here, the explanation will be continued assuming that the measurement device 100 has previously aligned the scan data with the reference data using a RANSAC (Random Sample Consensus) algorithm.

The measurement apparatus 100 matches point _Ps to point _Pr , for example, by identifying point _Pr that is the closest distance from point _Ps in map space. The measurement apparatus 100 identifies point _Pr that is the closest distance from point _Ps in map space by, for example, performing a nearest neighbor search. When performing the nearest neighbor search, the measurement apparatus 100, for example, constructs a k-d tree from multiple points _Ps of the scan data and performs the nearest neighbor search using the constructed k-d tree. The description will continue assuming _that , in the example shown in the upper part of FIG. 6 , the measurement apparatus 100 matches point _Ps1 to point _Pr1 , point _Ps2 to point _Pr3 , point _Ps3 to point _Pr3 , point _{Ps4 to point Pr4, point Ps5 to point Pr4} , and point Ps6 to point _Pr6 .

The diagram in the middle of Figure 6 is an explanatory diagram illustrating an example of a process (sometimes referred to as "Process 2") for determining an alignment matrix for aligning scan data with reference data. Here, we will mainly explain an example in which the measurement device 100 determines an alignment matrix based on the matching results in Process 1.

For each of a plurality of points _Ps in the scan data, the measurement device 100 determines a position error in the map space of the point _Ps relative to a point _Pr matched to the point _Ps , and determines a registration matrix such that the sum of the position errors is minimized. The registration matrix includes, for example, a rotation matrix. The registration matrix includes, for example, a translation matrix. In an example shown in the middle of Figure 6, the measurement device 100 determines an alignment matrix based on the matching results in process ₁ so that the sum e total = e ₁ + e ₂ + e ₃ + e 4 + e ₅ + e ₆ of the positional error e 1 in map space of point P _s1 relative to point P _r1 , the positional error e ₂ in map space of point P _s2 relative to point P _r3 , the positional error e ₃ in map space of point P _s3 relative to point P _r3 , the positional error e ₄ in map space of point P _s4 relative to point P _r4 , the _positional error e ₅ in map space of point _{P s5 relative} to point _{P r4 ,} and the positional error e ₆ in map space of point P _s6 relative to point P r6 is minimized.

The diagram shown in the lower part of Figure 6 is an explanatory diagram illustrating an example of a process (sometimes referred to as "Process 3") that determines whether the positional relationship between the scan data and the reference data satisfies a predetermined termination condition. Here, the explanation will continue assuming that the measurement device 100 aligns the scan data with the reference data using the alignment matrix determined by Process 2.

If the measuring device 100 determines that the positional relationship between the scan data and the reference data satisfies the termination condition, it may terminate the process of aligning the scan data with the reference data using the ICP algorithm. On the other hand, if the measuring device 100 determines that the positional relationship between the scan data and the reference data does not satisfy the termination condition, it may execute process 1 again. In this case, the measuring device 100 repeatedly executes processes 1 to 3 until it determines that the positional relationship between the scan data and the reference data satisfies the termination condition.

The termination condition may be, for example, that the sum of positional errors in map space between each point _Ps of the scan data after the scan data has been aligned with the reference data and each point _Pr of the reference data is smaller than a predetermined error threshold. The measurement device 100, for example, performs a nearest neighbor search to identify a point _Pr that is the shortest distance in map space from the point _Ps of the scan data after the scan data has been aligned with the reference data, and matches the point _Ps to the point _Pr . In an example shown in the lower part of FIG. 6 , the measurement device 100 may match point _Ps1 to point _Pr1 , point _Ps2 to point _Pr2 , point _Ps3 to point _Pr3 , point _Ps4 to point _Pr4 , point _Ps5 to point _Pr5 , and point _{Ps6 to point Pr6 .} Then, the measurement device 100 may determine that the positional relationship between the scan _data and the reference _data satisfies the termination condition if the sum _e'total = _e'1 + e'2 + e'3 + e'4 + e'5 + _e'6 of the positional error _e'1 in map space of _{point Ps1} relative to point _Pr1 , the positional error _e'2 in map space of point _Ps2 relative to point _Pr2 , the positional error _e'3 in map space of _{point Ps3} relative to point Pr3, the _positional error _e'4 in map space of point _Ps4 relative to _{point Pr4} , the _positional error _e'5 in map space of point Ps5 relative to point _Pr5 , and the positional error _e'6 in map space of point Ps6 relative to point _Pr6 is less than the error threshold.

Figure 7 shows an example of a pose graph for the measurement device 100. Here, we will explain each component of the pose graph for the measurement device 100.

The black circles in Figure 7 are vertices of the pose graph of the measurement device 100. The vertices represent the position of the measurement device 100 when the LiDAR 130 measured the second scan data.

The solid lines in Figure 7 are odometry edges in the pose graph of the measurement device 100. The odometry edges represent the relative positions between the position of the measurement device 100 when the LiDAR 130 measured the second scan data the ath time and the position of the measurement device 100 when the LiDAR 130 measured the second scan data the a+1th time. a is a natural number.

The vertex that is the starting point of the odometry edge may correspond to the position of the measurement device 100 when the LiDAR 130 measured the second scan data for the ath time. The vertex that is the end point of the odometry edge may correspond to the position of the measurement device 100 when the LiDAR 130 measured the second scan data for the a+1th time.

The dotted lines in Figure 7 are loop edges in the pose graph of the measurement device 100. Loop edges are edges that are generated when a loop, which is a circular path, is detected. Note that detecting a loop is sometimes referred to as loop detection.

The measurement device 100 performs loop detection based on, for example, the position of the measurement device 100 when the LiDAR 130 measured the second scan data. For example, the measurement device 100 detects a loop when the relative position between the position of the measurement device 100 when the LiDAR 130 measured the second scan data the ath time (represented by a first vertex) and the position of the measurement device 100 when the LiDAR 130 measured the second scan data the a+αth time (represented by a second vertex) is shorter than a predetermined relative position threshold. In this case, the vertex that is the start point of the loop edge corresponds to the position of the measurement device 100 when the LiDAR 130 measured the second scan data the ath time, and the vertex that is the end point of the loop edge corresponds to the position of the measurement device 100 when the LiDAR 130 measured the second scan data the a+αth time. α is a natural number that satisfies α≧ _αth . _αth may be a measurement interval threshold.

In an example of a pose graph of the measurement device 100 shown in Figure 7, vertex _Vm is the start point of a loop edge, and vertex _Vn is the end point of the loop edge. Vertex _Vm represents the position of the measurement device 100 when the LiDAR 130 measured the second scan data for the mth time, and vertex _Vn represents the position of the measurement device 100 when the LiDAR 130 measured the second scan data for the nth time. m and n are natural numbers, and the relationship n = m + α holds.

The measurement device 100 adjusts the pose graph in response to generating the loop edge. Note that adjusting the pose graph may also be referred to as pose adjustment.

In pose adjustment, the position of the measurement device 100 when the LiDAR 130, represented by the vertex, measured the second scan data for the ath time is a variable. On the other hand, in pose adjustment, the relative position between the position of the measurement device 100 when the LiDAR 130, represented by the odometry edge, measured the second scan data for the ath time and the position of the measurement device 100 when the LiDAR 130 measured the second scan data for the (a+1)th time is a constant.

For example, the measuring device 100 derives the relative position between the position of the measuring device 100 when LiDAR 130 measured the second scan data on the a+αth time at the position of the measuring device 100 represented by the vertex that is the end point of the loop edge and the position of the measuring device 100 when LiDAR 130 measured the second scan data on the a+αth time by aligning the second scan data measured by LiDAR 130 on the ath time at the position of the measuring device 100 represented by the vertex that is the start point of the loop edge. In the example pose graph of the measurement device 100 shown in FIG. 7, the relative position between the position of the measurement device 100 when the LiDAR 130 measured the second scan data for the mth time and the position of the measurement device 100 when the LiDAR 130 measured the second scan data for the nth time is aligned with an alignment algorithm to derive the relative position between the position of the measurement device 100 when the LiDAR 130 measured the second scan data for the mth time and the position of the measurement device 100 when the LiDAR 130 measured the second scan data for the nth time.

Next, the measurement device 100 adjusts the vertex that is the end point of the loop edge based on the derived relative position. In the example of the pose graph of the measurement device 100 shown in Figure 7, the vertex _Vn that is the end point of the loop edge is adjusted to the vertex _V'n .

Then, the measuring device 100 adjusts the vertex representing the position of the measuring device 100 when the second scan data was measured on the (a+α-1)th time, based on the position of the measuring device 100 when the second scan data was measured on the (a+α)th time, which is represented by the adjusted vertex, and the relative position between the position of the measuring device 100 when the second scan data was measured on the (a+α)th time and the position of the measuring device 100 when the second scan data was measured on the (a+α-1)th time. Thereafter, the vertices are adjusted in the same manner, successively adjusting the vertices representing the position of the measuring device 100 when the second scan data was measured on the (a+α-2)th time, the vertex representing the position of the measuring device 100 when the second scan data was measured on the (a+α-3)th time, and so on.

Pose adjustment optimizes the pose graph. Note that optimizing the pose graph is sometimes referred to as pose graph optimization.

Figure 8 schematically shows an example of a pose graph of the measuring device 100 before pose adjustment and map data corresponding to the pose graph of the measuring device 100 before pose adjustment. The diagram shown in the upper part of Figure 8 is an example of a pose graph of the measuring device 100 before pose adjustment. The diagram shown in the lower part of Figure 8 is an example of map data corresponding to the pose graph of the measuring device 100 before pose adjustment.

The pose graph of the measuring device 100 before pose adjustment shown in the upper part of Figure 8 is identical to the pose graph of the measuring device 100 shown in Figure 7. Therefore, a description of the main points regarding the pose graph of the measuring device 100 before pose adjustment shown in the upper part of Figure 8 will be omitted.

Point _Vm in the diagram shown in the lower part of Fig. 8 indicates the position in map space of the measurement device 100 when the LiDAR 130 measured the second scan data for the mth time. Point _Vm in the diagram shown in the lower part of Fig. 8 corresponds to vertex _Vm in the pose graph of the measurement device 100 before pose adjustment shown in the upper part of Fig. 8.

Point _Vn in the diagram shown in the lower part of Fig. 8 indicates the position in map space of the measurement device 100 when the LiDAR 130 measured the second scan data for the nth time. Point _Vn in the diagram shown in the lower part of Fig. 8 corresponds to vertex _Vn in the pose graph of the measurement device 100 before pose adjustment shown in the upper part of Fig. 8.

The dotted line in the diagram shown in the lower part of Figure 8 represents the trajectory of the measurement device 100 in map space. The trajectory of the measurement device 100 in map space shown in the lower part of Figure 8 corresponds to the odometry edge of the pose graph of the measurement device 100 before pose adjustment shown in the upper part of Figure 8.

In the diagram shown in the lower part of Fig. 8, the region _Rm surrounded by a dashed line is the range of the mth scan by the LiDAR 130. In the diagram shown in the lower part of Fig. 8, the region _Rn surrounded by a dashed line is the range of the nth scan by the LiDAR 130. The thick solid line in the diagram shown in the lower part of Fig. 8 is the second scan data measured by the LiDAR 130.

According to the pose graph of the measurement device 100 before pose adjustment shown in the upper part of Figure 8, the loop is not closed at the stage when the loop edge is detected by loop detection. As a result, the map data shown in the lower part of Figure 8 shows multiple instances of the same real-world location.

Figure 9 shows an example of a pose graph of the measurement device 100 before and after pose adjustment. The diagram shown in the upper part of Figure 9 is an example of a pose graph of the measurement device 100 before pose adjustment. The diagram shown in the lower part of Figure 9 is an example of a pose graph of the measurement device 100 after pose adjustment.

The pose graph of the measuring device 100 before pose adjustment shown in the upper part of Figure 9 is identical to the pose graph of the measuring device 100 shown in Figure 7. Therefore, a description of the main points regarding the pose graph of the measuring device 100 before pose adjustment shown in the upper part of Figure 9 will be omitted.

The measurement device 100 performs pose adjustment in response to generating the loop edge. For example, the measurement device 100 aligns the second scan data measured by the LiDAR 130 the nth time, which is included in the region R _n illustrated in the lower part of Fig. 8 , with the second scan data measured by the LiDAR 130 the mth time, which is included in the region R _m illustrated in the lower part of Fig. 8 , thereby deriving the relative position between the position of the measurement device 100 when the LiDAR 130 measured the second scan data the mth time and the position of the measurement device 100 when the LiDAR 130 measured the second scan data the nth time. Based on the derived relative positions, the measurement device 100 adjusts vertex _Vn , which is the end point of the loop edge, to vertex _V'n , and then adjusts vertex Vn _-1 to vertex _V'n-1 , vertex Vn _-2 to vertex V'n _-2 , ..., and adjusts vertex _Vm+1 to vertex _V'm+1 .

The black circles in the diagram at the bottom of Figure 9 are the vertices of the pose graph of the measurement device 100 after the pose adjustment. The solid lines in the diagram at the bottom of Figure 9 are the odometry edges of the pose graph of the measurement device 100 after the pose adjustment.

Figure 10 shows an example of a pose graph of the measuring device 100 after pose adjustment and map data corresponding to the pose graph of the measuring device 100 after pose adjustment. The diagram shown in the upper part of Figure 10 is the pose graph of the measuring device 100 after pose adjustment. The diagram shown in the lower part of Figure 10 is map data corresponding to the pose graph of the measuring device 100 after pose adjustment.

The pose graph of the measuring device 100 after pose adjustment shown in the upper part of Figure 10 is identical to the pose graph of the measuring device 100 after pose adjustment shown in the lower part of Figure 9. Therefore, a description of the main points regarding the pose graph of the measuring device 100 after pose adjustment shown in the upper part of Figure 10 will be omitted.

Point _Vm in the diagram shown in the lower part of Fig. 10 indicates the position in map space of the measurement device 100 when the LiDAR 130 measured the second scan data for the mth time. Point _Vm in the diagram shown in the lower part of Fig. 10 corresponds to vertex _Vm in the pose graph of the measurement device 100 after the pose adjustment shown in the upper part of Fig. 10.

Point V' _n in the diagram shown in the lower part of Fig. 10 indicates the position in map space of the measurement device 100 when the LiDAR 130 measured the second scan data for the nth time. Point V' _n in the diagram shown in the lower part of Fig. 10 corresponds to vertex V' _n in the pose graph of the measurement device 100 after the pose adjustment shown in the upper part of Fig. 10.

The dotted line in the diagram shown in the lower part of Figure 10 represents the trajectory of the measurement device 100 in map space. The trajectory of the measurement device 100 in map space shown in the lower part of Figure 10 corresponds to the odometry edge of the pose graph of the measurement device 100 after pose adjustment shown in the upper part of Figure 10.

The pose graph of the measuring device 100 before pose adjustment, shown in the upper part of Figure 10, shows that the loop was closed by the pose adjustment. Therefore, as a result of the pose adjustment, the same location in the real world no longer exists multiple times on the map data shown in the lower part of Figure 10. From the above, it can be said that pose graph optimization improves the accuracy of map data.

Figure 11 is an explanatory diagram illustrating another example of the process flow for generating map data. Here, an example of the process flow for generating map data using the SLAM algorithm is explained in three steps. It is assumed that LiDAR 130 simultaneously outputs two laser beams: a first laser beam and a second laser beam whose output direction is opposite to that of the first laser beam, that the rotation speed of LiDAR 130 is 30 rpm, that the scan interval of LiDAR 130 is 100 ms, and that the horizontal FOV of LiDAR 130 in a stationary state is 30 degrees.

The diagram shown in the upper part of Fig. 11 is an explanatory diagram for explaining Step 1, which is an example of a process for generating reference data. "S _i '" in the diagram shown in the upper part of Fig. 11 is the first scan data measured by the LiDAR 130 for the i-th time, where i is a natural number. Here, the explanation will continue assuming that the number of scan data constituting the reference data is 50, and that the LiDAR 130 measures the first scan data for 10 seconds.

Since the scan interval of LiDAR 130 is 100 ms and LiDAR 130 measures the first scan data for 10 seconds, LiDAR 130 measures the first scan data 100 times. Therefore, there are 100 pieces of first scan data. In this case, the 100 pieces of first scan data may be ranked in descending order of the time they were measured by LiDAR 130, and the 50 first scan data with the highest rankings may be selected to generate reference data.

The diagram shown in the middle of Fig. 11 is an explanatory diagram for explaining Step 2, which is an example of a process for grouping a plurality of second scan data. "S _j " in the diagram shown in the middle of Fig. 11 is the second scan data measured the jth time by the LiDAR 130. "G _k " in the diagram shown in the middle of Fig. 11 is the kth group. j and k are natural numbers.

The multiple second scan data are grouped, for example, based on the movement data of LiDAR 130 measured by IMU 140 when LiDAR 130 measured each of the multiple second scan data. The multiple second scan data are grouped based on the rotation angle of LiDAR 130 when LiDAR 130 measured each of the multiple second scan data. Second scan data grouped into the same group may be integrated based on the rotation angle of the LiDAR.

The plurality of second scan data are grouped, for example, so that the horizontal FOV of the LiDAR 130 becomes 360 degrees when the second scan data grouped in the same group are combined. The plurality of second scan data are grouped, for example, so that second scan data measured during a period in which the LiDAR 130 rotates halfway around the rotation axis _lr are grouped in the same group. For example, if the rotation speed of the LiDAR 130 is 30 rpm and the scan interval of the LiDAR 130 is 100 ms, 10 pieces of second scan data measured adjacent in time during a period in which the LiDAR 130 rotates halfway around the rotation axis _lr are grouped in the same group.

Each of the multiple second scan data may be aligned to the reference data, for example. Each of the multiple second scan data may be aligned to the reference data before the multiple second scan data are grouped, for example. Each of the multiple second scan data may be aligned to the reference data after the multiple second scan data are grouped.

The reference data is updated, for example, based on second scan data aligned with the reference data. The reference data is updated, for example, by ranking the multiple scan data that make up the reference data in descending order of the time measured by LiDAR 130, and replacing the lowest-ranked scan data with second scan data aligned with the reference data. Updating the reference data may be an example of generating reference data.

The diagram shown in the lower part of Fig. 11 is an explanatory diagram for explaining Step 3, which is an example of a process for generating map data and a pose graph of the measuring device 100 using the SLAM algorithm. "G _k " in the diagram shown in the lower part of Fig. 11 is the kth group formed in Step 2 shown in the middle part of Fig. 11.

The map data is generated in groups based on the multiple second scan data grouped in Step 2 shown in the middle of Figure 11. The map data is generated by aligning the second scan data grouped in the same group with the reference data on a group-by-group basis, for example.

The reference data is updated, for example, based on second scan data grouped into the same group and aligned with the reference data. The reference data is updated, for example, by ranking the multiple scan data that make up the reference data in descending order of the time measured by LiDAR 130, and replacing the number of scan data that make up the lowest-ranked group with second scan data grouped into the same group and aligned with the reference data.

The pose graph of the measuring device 100 is generated in groups, for example, based on the multiple second scan data grouped in Step 2 shown in the middle of Figure 11. For example, when the pose graph of the measuring device 100 is generated in groups, the second scan data grouped in the same group are ranked in descending order of the time measured by the LiDAR 130, and the position of the measuring device 100 when the LiDAR 130 measured the second scan data with the highest ranking is set as the vertex of the pose graph of the measuring device 100.

By generating map data and the pose graph of the measuring device 100 in groups based on multiple grouped second scan data, the number of vertices that make up the pose graph of the measuring device 100 can be reduced. This reduces the processing burden of the pose graph generation process and pose graph optimization process, while allowing map data to be generated based on scan data measured by a rotating LiDAR. Because the SLAM algorithm is an algorithm that can process a huge amount of data and perform a huge number of calculations, it is important to be able to reduce the processing burden of the pose graph generation process and pose graph optimization process.

Figure 12 shows a schematic diagram of an example of the system 10. The system 10 may include a measurement device 100. The system 10 may include a map data generation device 200.

The measuring device 100 communicates with, for example, the map data generation device 200. The measuring device 100 communicates with the map data generation device 200 via, for example, the network 20.

The network 20 may include, for example, a LAN (Local Area Network). The LAN may include, for example, a wireless LAN conforming to standards such as Wi-Fi (registered trademark). The LAN may also include a wired LAN conforming to standards such as Ethernet (registered trademark). The network 20 may include a mobile communication network. The network 20 may include the Internet. The network 20 may also include cables.

The measuring device 100, for example, transmits various data to the map data generation device 200. The measuring device 100, for example, transmits measurement data measured by the measuring device 100 to the map data generation device 200. The measuring device 100 may also transmit captured image data captured by a camera provided in the measuring device 100 to the map data generation device 200.

The map data generation device 200 generates map data. The map data generation device 200 may generate map data in the same manner as when the measurement device 100 generates map data.

The map data generation device 200 generates map data based on, for example, various data acquired from the measuring device 100. The map data generation device 200 acquires various data from the measuring device 100 by, for example, receiving the various data from the measuring device 100 via the network 20.

The map data generation device 200 generates map data based on, for example, measurement data measured by the measuring device 100. The map data generation device 200 may also generate map data based on captured image data captured by a camera provided in the measuring device 100.

The map data generation device 200 may generate a pose graph for the measurement device 100. The map data generation device 200 may generate a pose graph for the measurement device 100 in the same manner as the measurement device 100 generates a pose graph.

The map data generation device 200 may transmit the various generated data to the measuring device 100 via the network 20. The map data generation device 200 may, for example, transmit map data to the measuring device 100. The map data generation device 200 may also transmit the pose graph of the measuring device 100 to the measuring device 100.

Figure 13 shows an example of the functional configuration of the measuring device 100. The measuring device 100 includes a generating unit 101, a storage unit 102, an acquiring unit 104, a grouping unit 106, and a communication unit 108. The generating unit 101 includes a measurement data generating unit 103, a reference data generating unit 105, a map data generating unit 107, and a pose graph generating unit 109. Note that it is not essential for the measuring device 100 to include all of these components.

The storage unit 102 stores various data. The storage unit 102 stores, for example, various thresholds. The storage unit 102 stores, for example, an error threshold. The storage unit 102 stores, for example, a relative position threshold. The storage unit 102 may store a measurement interval threshold. The storage unit 102 may also store scan number information indicating the number of scan data that make up the reference data.

The acquisition unit 104 acquires various data. For example, the acquisition unit 104 acquires various data from the LiDAR 130. For example, the acquisition unit 104 acquires various data from the IMU 140. The acquisition unit 104 may store the acquired various data in the storage unit 102.

The acquisition unit 104 acquires, for example, multiple pieces of first scan data measured by the LiDAR 130 over a predetermined period of time. The acquisition unit 104 acquires, for example, movement data of the LiDAR 130 measured by the IMU 140 over the predetermined period of time, each piece of first scan data from the multiple pieces of first scan data. The acquisition unit 104 may be an example of a first acquisition unit.

The acquisition unit 104 acquires, for example, a plurality of first scan data measured by the LiDAR 130 during a predetermined period that is equal to or longer than the period for the LiDAR 130 to rotate half a revolution around the rotation axis _lr , and movement data of the LiDAR 130 measured during the predetermined period by the IMU 140 when the LiDAR 130 measured each of the first scan data of the plurality of first scan data. The acquisition unit 104 acquires, for example, a plurality of first scan data measured by the LiDAR 130 during a predetermined period that is longer than the period for the LiDAR 130 to rotate half a revolution around the rotation axis _lr , and movement data of the LiDAR 130 measured during the predetermined period by the IMU 140 when the LiDAR 130 measured each of the first scan data of the plurality of first scan data. The acquisition unit 104 may acquire a plurality of first scan data measured by the LiDAR 130 during a predetermined period during which the LiDAR 130 rotates half a revolution around the rotation axis l/ _r , and movement data of the LiDAR 130 measured by the IMU 140 during the predetermined period when the LiDAR 130 measured each of the first scan data of the plurality of first scan data.

The acquisition unit 104 acquires, for example, second scan data measured by the LiDAR 130. The acquisition unit 104 acquires, for example, movement data of the LiDAR 130 measured by the IMU 140 when the LiDAR 130 measured each second scan data of the plurality of second scan data. The acquisition unit 104 may be an example of a second acquisition unit.

The generation unit 101 generates various data. The generation unit 101 generates various data, for example, based on various data stored in the storage unit 102. The generation unit 101 generates various data, for example, based on various data acquired by the acquisition unit 104. The generation unit 101 may generate various data based on the generated various data. The generation unit 101 may store the generated various data in the storage unit 102.

The measurement data generation unit 103 generates measurement data. For example, the measurement data generation unit 103 generates measurement data in response to the LiDAR 130 measuring scan data. The measurement data generation unit 103 may also generate measurement data in response to the IMU 140 measuring the movement of the LiDAR 130.

The reference data generation unit 105 generates reference data. The reference data generation unit 105 generates reference data based on, for example, scan data measured by the LiDAR 130.

The reference data generation unit 105 generates reference data, for example, based on multiple first scan data measured by the LiDAR 130 and the movement data of the LiDAR 130 measured by the IMU 140 when the LiDAR 130 measured each of the multiple first scan data. The reference data generation unit 105 generates reference data, for example, by identifying the rotation angle of the LiDAR 130 when the LiDAR 130 measured each of the first scan data based on the movement data of the LiDAR 130 when the LiDAR 130 measured each of the first scan data, and integrating each of the first scan data based on the rotation angle of the LiDAR 130 when the LiDAR 130 measured each of the first scan data. For example, when the LiDAR 130 measures one piece of first scan data and then measures other pieces of first scan data adjacent in time, the reference data generation unit 105 identifies the relative rotation angle between the rotation angle of the LiDAR 130 when the LiDAR 130 measured the one piece of first scan data and the rotation angle of the LiDAR 130 when the LiDAR 130 measured the other piece of first scan data, and integrates the first scan data by applying an inverse rotation matrix of the relative rotation angle to the other piece of first scan data to align the other piece of first scan data with the one piece of first scan data.

The reference data generation unit 105 may generate reference data by integrating each of the first scan data based on the result of aligning each of the first scan data of the multiple first scan data using an ICP algorithm. The reference data generation unit 105 integrates each of the first scan data, for example, by aligning other first scan data measured after one of the first scan data with the first scan data using an ICP algorithm. The reference data generation unit 105 may generate reference data by integrating each of the first scan data based on the result of aligning each of the first scan data of the multiple first scan data using an NDT algorithm.

For example, when the number of the plurality of first scan data is greater than the number of scan data indicated by the scan number information stored in the storage unit 102, the reference data generation unit 105 generates the reference data by selecting, from the plurality of first scan data, first scan data that constitute the reference data. Here, the explanation will continue assuming that the number of scan data indicated by the scan number information is _Nr , where _Nr is a natural number.

For example, the reference data generation unit 105 ranks the plurality of first scan data in descending order of the time measured by the LiDAR 130, and selects the _Nr first scan data with the highest rank as first scan data constituting the reference data. Then, the reference data generation unit 105 generates the reference data by integrating the selected _Nr first scan data.

The map data generation unit 107 generates map data. The map data generation unit 107 generates map data using, for example, a SLAM algorithm.

The map data generation unit 107 generates map data based on, for example, the reference data generated by the reference data generation unit 105 and multiple pieces of second scan data measured by the LiDAR 130. The map data generation unit 107 generates map data by, for example, aligning the second scan data with the reference data using an ICP algorithm. The map data generation unit 107 may also generate map data by aligning the second scan data with the reference data using an NDT algorithm.

The map data generation unit 107 generates map data further based on, for example, movement data of the LiDAR 130 measured by the IMU 140 when the LiDAR 130 measured each of the multiple pieces of second scan data. The map data generation unit 107 generates map data by, for example, identifying the position of the measuring device 100 and the rotation angle of the LiDAR 130 when the LiDAR 130 measured each piece of second scan data based on the movement data of the LiDAR 130 when the LiDAR 130 measured each piece of second scan data, and integrating each piece of second scan data based on the position of the measuring device 100 and the rotation angle of the LiDAR 130 when the LiDAR 130 measured each piece of second scan data. For example, in a case where the LiDAR 130 measures one piece of second scan data and then measures other pieces of second scan data adjacent in time, the map data generation unit 107 identifies the relative position between the position of the measurement device 100 when the LiDAR 130 measured the one piece of second scan data and the position of the measurement device 100 when the LiDAR 130 measured the other piece of second scan data, and the relative rotation angle between the rotation angle of the LiDAR 130 when the LiDAR 130 measured the one piece of second scan data and the rotation angle of the LiDAR 130 when the LiDAR 130 measured the other piece of second scan data, and integrates the second scan data by applying a translation matrix corresponding to the relative position and an inverse rotation matrix of the relative rotation angle to the other piece of second scan data to align the other piece of second scan data with the one piece of second scan data.

The reference data generation unit 105 updates the reference data in response to, for example, the map data generation unit 107 aligning the second scan data with the reference data. The reference data generation unit 105 updates the reference data, for example, by replacing the scan data measured by the LiDAR 130 earliest among the _Nr scan data constituting the reference data with the second scan data aligned with the reference data by the map data generation unit 107. The reference data generation unit 105 updates the reference data by, for example, ranking the _Nr scan data constituting the reference data in descending order of the time measured by the LiDAR 130 and replacing the scan data with the lowest ranking with the second scan data aligned with the reference data by the map data generation unit 107. When the reference data generation unit 105 updates the reference data, the map data generation unit 107 may generate map data based on the reference data updated by the reference data generation unit 105 and the multiple second scan data measured by the LiDAR 130.

The storage unit 102 may store the updated reference data. The storage unit 102 may also store the reference data before the update.

The pose graph generation unit 109 generates a pose graph for the measuring device 100. The pose graph generation unit 109 generates a pose graph for the measuring device 100 using, for example, a SLAM algorithm. The map data generation unit 107 may generate map data further based on the pose graph for the measuring device 100 generated by the pose graph generation unit 109.

The pose graph generation unit 109 generates a pose graph for the measurement device 100 based on, for example, the reference data generated by the reference data generation unit 105 and multiple pieces of second scan data measured by the LiDAR 130. The pose graph generation unit 109 generates the pose graph for the measurement device 100 by setting the position of the measurement device 100 when the LiDAR 130 measured the scan data as the vertex of the pose graph for the measurement device 100. For example, when the LiDAR 130 measures one piece of scan data and then measures another piece of scan data adjacent in time, the pose graph generation unit 109 generates the pose graph for the measurement device 100 by setting the relative positions between the position of the measurement device 100 when the LiDAR 130 measured the one piece of scan data and the position of the measurement device 100 when the LiDAR 130 measured the other piece of scan data as odometry edges of the pose graph for the measurement device 100.

The pose graph generation unit 109 performs loop detection, for example, based on the position of the measurement device 100 when the LiDAR 130 measured each of the multiple pieces of second scan data. The pose graph generation unit 109 detects a loop when, for example, the measurement interval from the time when the LiDAR 130 measured the second scan data at a first vertex of the pose graph of the measurement device 100 to the time when the LiDAR 130 measured the second scan data at a second vertex of the pose graph of the measurement device 100 is longer than the measurement interval threshold stored in the storage unit 102, and the relative position between the position of the measurement device 100 represented by the first vertex and the position of the measurement device 100 represented by the second vertex is shorter than the relative position threshold stored in the storage unit 102.

The pose graph generation unit 109 may generate a loop edge in response to detecting a loop. Here, we will continue the explanation assuming that the vertex that is the start point of the loop edge is the first vertex, and the vertex that is the end point of the loop edge is the second vertex.

The pose graph generator 109 may perform pose adjustment in response to generating a loop edge. For example, the pose graph generator 109 aligns the second scan data measured by the LiDAR 130 at the second vertex, which is the end point of the loop edge, with the second scan data measured by the LiDAR 130 at the first vertex, which is the start point of the loop edge, to derive the relative position between the position of the measurement device 100 represented by the first vertex and the position of the measurement device 100 represented by the second vertex. Next, the pose graph generator 109 adjusts the second vertex based on the derived relative position. Thereafter, the pose graph generator 109 sequentially adjusts other vertices in the pose graph of the measurement device 100 that correspond to second scan data measured after the time when the LiDAR 130 measured the second scan data at the first vertex, based on the adjusted second vertex.

The grouping unit 106 groups multiple pieces of second scan data measured by the LiDAR 130. The grouping unit 106 groups the multiple pieces of second scan data, for example, based on the movement data of the LiDAR 130 measured by the IMU 140 when the LiDAR 130 measured each piece of second scan data. The grouping unit 106 groups the multiple pieces of second scan data, for example, based on the rotation angle of the LiDAR 130 when the LiDAR 130 measured each piece of second scan data.

The grouping unit 106 groups the plurality of second scan data, for example, so that the horizontal FOV of the LiDAR 130 becomes 360 degrees when the grouped second scan data are integrated. The grouping unit 106 groups the plurality of second scan data, for example, by grouping the second scan data measured during a period in which the LiDAR 130 makes a half rotation around the rotation axis _lr into the same group. Here, the description will continue assuming that _Ng pieces of second scan data are grouped into the same group, where _Ng is a natural number.

The map data generation unit 107 may, for example, align each piece of second scan data with the reference data before the grouping unit 106 groups the pieces of second scan data. The map data generation unit 107 may, for example, align each piece of second scan data with the reference data after the grouping unit 106 groups the pieces of second scan data.

The map data generation unit 107 generates map data in groups, for example, based on the reference data generated by the reference data generation unit 105 and the multiple second scan data grouped by the grouping unit 106. The map data generation unit 107 generates map data in groups, for example, by aligning the second scan data grouped into the same group by the grouping unit 106 with the reference data.

For example, the map data generation unit 107 simultaneously aligns the _Ng pieces of second scan data grouped into the same group with the reference data. Alternatively, the map data generation unit 107 may individually align the _Ng pieces of second scan data grouped into the same group with the reference data.

The reference data generation unit 105 updates the reference data, for example, based on the second scan data grouped into the same group by the grouping unit 106. The reference data generation unit 105 updates the reference data, for example, in response to the map data generation unit 107 aligning the second scan data grouped into the same group with the reference data. The reference data generation unit 105 updates the reference data, for example, by ranking the Nr _scan data constituting the reference data in descending order of the time measured by the LiDAR 130 and replacing the _Ng scan data with the lowest rankings with the _Ng second scan data that have been grouped into the same group by the grouping unit 106 and aligned with the reference data by the map data generation unit 107.

The pose graph generation unit 109 generates a pose graph for the measurement device 100 for each group, for example, based on the reference data generated by the reference data generation unit 105 and the plurality of second scan data grouped by the grouping unit 106. The pose graph generation unit 109 generates a pose _graph for the measurement device 100 for each group, for example, by setting the position of the measurement device 100 when the LiDAR 130 measured the second scan data that was measured earliest among the N g pieces of second scan data grouped into the same group as a vertex of the pose graph for the measurement device 100. The pose graph generation unit 109 generates a pose graph for the measurement device 100 for each group, for example, by ranking the N _g pieces of second scan data grouped into the same group in descending order of the time of measurement by the LiDAR 130, and setting the position of the measurement device 100 when the LiDAR 130 measured the second scan data with the highest ranking as a vertex of the pose graph for the measurement device 100.

The communication unit 108 communicates with the map data generation device 200. The communication unit 108 communicates with the map data generation device 200, for example, via the network 20.

The communication unit 108 transmits various data to the map data generation device 200, for example. The communication unit 108 transmits various data generated by the generation unit 101, for example. The communication unit 108 transmits various data stored in the storage unit 102, for example. The communication unit 108 transmits various data acquired by the acquisition unit 104, for example.

The communication unit 108 receives various data from, for example, the map data generation device 200. The communication unit 108 receives, for example, map data generated by the map data generation device 200. The communication unit 108 may also receive a pose graph of the measuring device 100 generated by the map data generation device 200.

Figure 14 shows an example of the functional configuration of the map data generation device 200. The map data generation device 200 includes a generation unit 201, a storage unit 202, an acquisition unit 204, a grouping unit 206, and a data transmission unit 208. The generation unit 201 includes a reference data generation unit 205, a map data generation unit 207, and a pose graph generation unit 209. Note that it is not essential for the map data generation device 200 to include all of these components.

The storage unit 202 stores various data. The storage unit 202 stores various data similar to the various data stored by the storage unit 102.

The acquisition unit 204 acquires various data. The acquisition unit 204 may store the acquired various data in the storage unit 202.

The acquisition unit 204 acquires various data from, for example, the measurement device 100. The acquisition unit 204 acquires various data from the measurement device 100, for example, by receiving the various data from the measurement device 100 via the network 20.

The acquisition unit 204 acquires, for example, various data similar to the various data acquired by the acquisition unit 104. The acquisition unit 204 acquires, for example, multiple pieces of first scan data measured by the LiDAR 130 over a predetermined period of time. The acquisition unit 204 acquires, for example, movement data of the LiDAR 130 measured by the IMU 140 over the predetermined period of time when the LiDAR 130 measured each piece of first scan data from the multiple pieces of first scan data. The acquisition unit 204 acquires, for example, second scan data measured by the LiDAR 130. The acquisition unit 204 acquires, for example, movement data of the LiDAR 130 measured by the IMU 140 when the LiDAR 130 measured each piece of second scan data from the multiple pieces of second scan data. The acquisition unit 204 may also acquire measurement data.

The acquisition unit 204 may be an example of a first acquisition unit. The acquisition unit 204 may be an example of a second acquisition unit.

The generation unit 201 generates various data. For example, the generation unit 201 generates various data based on various data stored in the storage unit 202. For example, the generation unit 201 generates various data based on various data acquired by the acquisition unit 204. The generation unit 201 may generate various data based on the generated various data. The generation unit 201 may store the generated various data in the storage unit 202.

The reference data generation unit 205 generates reference data. The reference data generation unit 205 generates the reference data based on, for example, multiple pieces of first scan data measured by the LiDAR 130 and the movement data of the LiDAR 130 measured by the IMU 140 when the LiDAR 130 measured each piece of first scan data of the multiple pieces of first scan data. The reference data generation unit 205 may have the same functions as the reference data generation unit 105.

The map data generation unit 207 generates map data. The map data generation unit 207 generates map data using, for example, the SLAM algorithm.

The map data generation unit 207 generates map data based on, for example, the reference data generated by the reference data generation unit 205 and multiple pieces of second scan data measured by the LiDAR 130. The map data generation unit 207 may have the same functions as the map data generation unit 107.

The pose graph generation unit 209 generates a pose graph for the measurement device 100. The pose graph generation unit 209 generates a pose graph for the measurement device 100 using, for example, the SLAM algorithm.

The pose graph generator 209 generates a pose graph for the measurement device 100 based on, for example, the reference data generated by the reference data generator 205 and multiple pieces of second scan data measured by the LiDAR 130. The pose graph generator 209 may have the same functions as the pose graph generator 109.

The data transmission unit 208 transmits various data to the measurement device 100 via the network 20. The data transmission unit 208 transmits, for example, various data generated by the generation unit 201 to the measurement device 100.

Figure 15 shows a schematic diagram of an example of the hardware configuration of a computer 1200 that functions as the measurement device 100 or the map data generation device 200. A program installed on the computer 1200 can cause the computer 1200 to function as one or more "parts" of the device according to the above embodiments, or can cause the computer 1200 to perform operations or one or more "parts" associated with the device according to the above embodiments, and/or can cause the computer 1200 to perform a process or steps of the process according to the above embodiments. Such a program can be executed by the CPU 1212 to cause the computer 1200 to perform specific operations associated with some or all of the blocks in the flowcharts and block diagrams described herein.

The computer 1200 according to this embodiment includes a CPU 1212, RAM 1214, and graphics controller 1216, which are interconnected by a host controller 1210. The computer 1200 also includes input/output units such as a communications interface 1222, a storage device 1224, a DVD drive 1226, and an IC card drive, which are connected to the host controller 1210 via an input/output controller 1220. The DVD drive 1226 may be a DVD-ROM drive, a DVD-RAM drive, or the like. The storage device 1224 may be a hard disk drive, a solid-state drive, or the like. The computer 1200 also includes legacy input/output units such as a ROM 1230 and a keyboard 1242, which are connected to the input/output controller 1220 via an input/output chip 1240.

The CPU 1212 operates according to programs stored in the ROM 1230 and RAM 1214, thereby controlling each unit. The graphics controller 1216 acquires image data generated by the CPU 1212 into a frame buffer provided in the RAM 1214 or into the graphics controller itself, and causes the image data to be displayed on the display device 1218.

The communication interface 1222 communicates with other electronic devices via a network. The storage device 1224 stores programs and data used by the CPU 1212 in the computer 1200. The DVD drive 1226 reads programs or data from a DVD-ROM 1227 or the like and provides them to the storage device 1224. The IC card drive reads programs and data from an IC card and/or writes programs and data to an IC card.

ROM 1230 stores therein a boot program and the like that is executed by computer 1200 upon activation, and/or programs that depend on the hardware of computer 1200. I/O chip 1240 may also connect various I/O units to I/O controller 1220 via USB ports, parallel ports, serial ports, keyboard ports, mouse ports, etc.

The programs are provided by a computer-readable storage medium such as a DVD-ROM 1227 or an IC card. The programs are read from the computer-readable storage medium, installed in storage device 1224, RAM 1214, or ROM 1230, which are also examples of computer-readable storage media, and executed by CPU 1212. The information processing described in these programs is read by computer 1200, resulting in cooperation between the programs and the various types of hardware resources described above. An apparatus or method may be configured by implementing the operation or processing of information in accordance with the use of computer 1200.

For example, when communication is performed between computer 1200 and an external device, CPU 1212 may execute a communication program loaded into RAM 1214 and instruct communication interface 1222 to perform communication processing based on the processing described in the communication program. Under the control of CPU 1212, communication interface 1222 reads transmission data stored in a transmission buffer area provided in RAM 1214, storage device 1224, DVD-ROM 1227, or a recording medium such as an IC card, and transmits the read transmission data to the network, or writes received data received from the network to a reception buffer area or the like provided on the recording medium.

The CPU 1212 may also cause all or a necessary portion of a file or database stored on an external recording medium such as the storage device 1224, DVD drive 1226 (DVD-ROM 1227), IC card, etc. to be read into the RAM 1214, and perform various types of processing on the data on the RAM 1214. The CPU 1212 may then write the processed data back to the external recording medium.

Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and may undergo information processing. CPU 1212 may perform various types of processing on data read from RAM 1214, including various types of operations, information processing, conditional judgment, conditional branching, unconditional branching, information search/replacement, etc., as described throughout this disclosure and specified by the program's instruction sequence, and write the results back to RAM 1214. CPU 1212 may also search for information in files, databases, etc. on the recording medium. For example, if multiple entries each having an attribute value of a first attribute associated with an attribute value of a second attribute are stored on the recording medium, CPU 1212 may search for an entry whose attribute value of the first attribute matches a specified condition from among the multiple entries, read the attribute value of the second attribute stored in the entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

The programs or software modules described above may be stored on computer-readable storage media on or near computer 1200. Recording media such as a hard disk or RAM provided within a server system connected to a dedicated communications network or the Internet can also be used as computer-readable storage media, thereby providing the programs to computer 1200 via the network.

The blocks in the flowcharts and block diagrams in this embodiment may represent stages of a process in which an operation is performed or "parts" of a device responsible for performing the operation. Particular stages and "parts" may be implemented by dedicated circuitry, programmable circuitry provided with computer-readable instructions stored on a computer-readable storage medium, and/or a processor provided with computer-readable instructions stored on a computer-readable storage medium. Dedicated circuitry may include digital and/or analog hardware circuitry, and may include integrated circuits (ICs) and/or discrete circuits. Programmable circuitry may include reconfigurable hardware circuitry including AND, OR, XOR, NAND, NOR, and other logical operations, flip-flops, registers, and memory elements, such as field programmable gate arrays (FPGAs) and programmable logic arrays (PLAs).

A computer-readable storage medium may include any tangible device capable of storing instructions executed by a suitable device, such that a computer-readable storage medium having instructions stored thereon comprises an article of manufacture, including instructions that can be executed to create means for performing the operations specified in the flowchart or block diagram. Examples of computer-readable storage media may include electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, etc. More specific examples of computer-readable storage media may include floppy disks, diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), electrically erasable programmable read-only memory (EEPROM), static random access memory (SRAM), compact disc read-only memory (CD-ROM), digital versatile disc (DVD), Blu-ray disc, memory stick, integrated circuit card, etc.

The computer-readable instructions may include either assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state-setting data, or source or object code written in any combination of one or more programming languages, including object-oriented programming languages such as Smalltalk®, JAVA®, C++, etc., and conventional procedural programming languages such as the "C" programming language or similar programming languages.

The computer-readable instructions may be provided locally or over a wide area network (WAN) such as a local area network (LAN), the Internet, etc. to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, or to a programmable circuit, such that the processor or programmable circuit executes the computer-readable instructions to generate means for performing the operations specified in the flowcharts or block diagrams. Examples of processors include computer processors, processing units, microprocessors, digital signal processors, controllers, microcontrollers, etc.

The present invention has been described above using embodiments, but the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be clear to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the claims that such modifications and improvements can also be included within the technical scope of the present invention.

The order of execution of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not specifically stated as "before" or "prior to," and it should be noted that processes can be performed in any order unless the output of a previous process is used in a subsequent process. Even if the operational flow in the claims, specifications, and drawings is described using terms such as "first," "next," etc. for convenience, this does not mean that the processes must be performed in that order.

10 System, 20 Network, 100 Measurement device, 101 Generation unit, 102 Storage unit, 103 Measurement data generation unit, 104 Acquisition unit, 105 Reference data generation unit, 106 Grouping unit, 107 Map data generation unit, 108 Communication unit, 109 Pose graph generation unit, 110 Box unit, 120 Base unit, 121 Bottom unit, 122 Mounting unit, 124 Support unit, 130 LiDAR, 135 Output surface, 140 IMU, 150 Motor, 160 Slip ring, 200 Map data generation device, 201 Generation unit, 202 Storage unit, 204 Acquisition unit, 205 Reference data generation unit, 206 Grouping unit, 207 Map data generation unit, 208 Data transmission unit, 209 Pose graph generation unit, 1200 Computer, 1210 Host controller, 1212 CPU, 1214 RAM, 1216 graphics controller, 1218 display device, 1220 input/output controller, 1222 communication interface, 1224 storage device, 1226 DVD drive, 1227 DVD-ROM, 1230 ROM, 1240 input/output chip, 1242 keyboard

Claims (16)

a measurement device comprising: a LiDAR (Light Detection and Ranging); an IMU (Inertial Measurement Unit); a base portion having a mounting portion on which the IMU is mounted and a support portion that supports the LiDAR; and a motor that rotates the base portion, wherein a support surface on which the support portion supports the LiDAR is perpendicular to a mounting surface on which the mounting portion mounts the IMU, and the center of the IMU passes through an axis of rotation of the LiDAR when the base portion rotates due to the mechanical energy of the motor; or a support surface on which the support portion supports the LiDAR is parallel to the mounting surface on which the mounting portion mounts the IMU, and the center of the IMU passes through a line that is perpendicular to the axis of rotation of the LiDAR when the base portion rotates due to the mechanical energy of the motor;
a first acquisition unit that acquires a plurality of first scan data measured by the LiDAR during a predetermined period in a state in which the base unit is rotating and not translating due to the mechanical energy of the motor, and motion data of the LiDAR when the LiDAR measured each of the plurality of first scan data, measured by the IMU during the predetermined period, in order to generate reference data that is referenced when generating map data using a SLAM (Simultaneous Localization and Mapping) algorithm;
a reference data generation unit that generates the reference data based on the plurality of first scan data and the movement data of the LiDAR when the LiDAR measured each of the first scan data;
a second acquisition unit that acquires a plurality of second scan data measured by the LiDAR while the base unit is rotating and translating due to the mechanical energy of the motor;
a map data generation unit that generates the map data based on the reference data generated by the reference data generation unit and the plurality of second scan data. The LiDAR outputs a first laser light having an output direction corresponding to a direction of a first vector included in a plane perpendicular to the support surface, and a second laser light having an output direction corresponding to a direction of a second vector included in a plane perpendicular to the support surface and having an opposite direction to the first vector;
The first acquisition unit acquires the plurality of first scan data measured during the predetermined period that is equal to or longer than a period for the LiDAR to rotate halfway around the rotation axis, and the movement data of the LiDAR when the LiDAR measured each of the first scan data.
The system of claim 1 . The system of claim 1, wherein the reference data generation unit determines the rotation angle of the LiDAR when the LiDAR measured each of the first scan data based on the movement data of the LiDAR when the LiDAR measured each of the first scan data, and generates the reference data by integrating each of the first scan data based on the rotation angle of the LiDAR when the LiDAR measured each of the first scan data. The system described in claim 3, wherein the reference data generation unit generates the reference data by integrating the first scan data based further on the result of aligning the first scan data using an ICP (Iterative Closest Point) algorithm. The second acquisition unit further acquires motion data of the LiDAR measured by the IMU when the LiDAR measured each of the plurality of second scan data,
The system comprises:
a grouping unit that groups the plurality of second scan data based on the movement data of the LiDAR when the LiDAR measured each of the plurality of second scan data,
the map data generating unit generates the map data by aligning the second scan data grouped into the same group by the grouping unit with the reference data;
A system according to any one of claims 1 to 4. The LiDAR outputs a first laser light having an output direction corresponding to a direction of a first vector included in a plane perpendicular to the support surface, and a second laser light having an output direction corresponding to a direction of a second vector included in a plane perpendicular to the support surface and having an opposite direction to the first vector;
The grouping unit groups the plurality of second scan data by grouping the second scan data measured during a period in which the LiDAR rotates halfway around the rotation axis into the same group.
The system of claim 5. the reference data generation unit updates the reference data based on the second scan data grouped into the same group by the grouping unit;
the map data generation unit generates the map data based on the reference data updated by the reference data generation unit and the plurality of second scan data.
The system of claim 5. a pose graph generation unit that generates, for each group, a pose graph that represents a trajectory of the measuring device in a graph structure based on the reference data generated by the reference data generation unit and the plurality of second scan data grouped by the grouping unit,
the map data generation unit generates the map data further based on the pose graph generated by the pose graph generation unit.
The system of claim 5. The second acquisition unit further acquires motion data of the LiDAR measured by the IMU when the LiDAR measured each of the plurality of second scan data,
The map data generation unit generates the map data further based on the movement data of the LiDAR when the LiDAR measured each of the plurality of second scan data.
A system according to any one of claims 1 to 4. LiDAR and
IMU and
a base portion having a mounting portion for mounting the IMU and a support portion for supporting the LiDAR;
a motor that rotates the base portion,
The support surface on which the support portion supports the LiDAR is perpendicular to the mounting surface on which the mounting portion mounts the IMU, and the center of the IMU passes through the rotation axis of the LiDAR when the base portion rotates due to the mechanical energy of the motor, or
The support surface on which the support part supports the LiDAR is parallel to the mounting surface on which the mounting part mounts the IMU, and the center of the IMU passes through a straight line perpendicular to the rotation axis of the LiDAR when the base part rotates by the mechanical energy of the motor.
Measuring device. a first acquisition unit that acquires a plurality of first scan data measured by the LiDAR during a predetermined period in a state in which the base unit is rotating and not translating due to the mechanical energy of the motor, and motion data of the LiDAR measured by the IMU during the predetermined period when the LiDAR measured each of the plurality of first scan data, in order to generate reference data that is referenced when generating map data using a SLAM algorithm;
a reference data generation unit that generates the reference data based on the plurality of first scan data and the movement data of the LiDAR when the LiDAR measured each of the first scan data;
a second acquisition unit that acquires a plurality of second scan data measured by the LiDAR while the base unit is rotating and translating due to the mechanical energy of the motor;
The measuring device according to claim 10 , further comprising: a map data generating unit that generates the map data based on the reference data generated by the reference data generating unit and the plurality of second scan data. a LiDAR, an IMU, a base unit having a mounting unit on which the IMU is mounted and a support unit that supports the LiDAR, and a motor that rotates the base unit, wherein a support surface on which the support unit supports the LiDAR is perpendicular to a mounting surface on which the mounting unit mounts the IMU, and a center of the IMU passes through an axis of rotation of the LiDAR when the base unit rotates due to the mechanical energy of the motor, or a support surface on which the support unit supports the LiDAR is parallel to the mounting surface on which the mounting unit mounts the IMU, and a center of the IMU passes through a line that is perpendicular to the axis of rotation of the LiDAR when the base unit rotates due to the mechanical energy of the motor,
a first acquisition unit that acquires a plurality of first scan data measured by the LiDAR during a predetermined period in a state in which the base unit is rotating and not translating due to the mechanical energy of the motor, and motion data of the LiDAR measured by the IMU during the predetermined period when the LiDAR measured each of the plurality of first scan data, in order to generate reference data that is referenced when generating the map data;
a reference data generation unit that generates the reference data based on the plurality of first scan data and the movement data of the LiDAR when the LiDAR measured each of the first scan data;
a second acquisition unit that acquires a plurality of second scan data measured by the LiDAR while the base unit is rotating and translating due to the mechanical energy of the motor;
a map data generating unit configured to generate the map data based on the reference data generated by the reference data generating unit and the plurality of second scan data. A map data generation method executed by a measurement device comprising: a LiDAR; an IMU; a base unit having a mounting unit on which the IMU is mounted and a support unit that supports the LiDAR; and a motor that rotates the base unit, wherein a support surface on which the support unit supports the LiDAR is perpendicular to a mounting surface on which the mounting unit mounts the IMU, and the center of the IMU passes through an axis of rotation of the LiDAR when the base unit rotates due to the mechanical energy of the motor; or a support surface on which the support unit supports the LiDAR is parallel to the mounting surface on which the mounting unit mounts the IMU, and the center of the IMU passes through a line that is perpendicular to the axis of rotation of the LiDAR when the base unit rotates due to the mechanical energy of the motor,
a first acquisition step of acquiring a plurality of first scan data measured by the LiDAR during a predetermined period in a state in which the base unit is rotating and not translating due to the mechanical energy of the motor, and movement data of the LiDAR measured by the IMU during the predetermined period when the LiDAR measured each of the plurality of first scan data, in order to generate reference data to be referenced when generating map data using a SLAM algorithm;
a reference data generation step of generating the reference data based on the plurality of first scan data and the movement data of the LiDAR when the LiDAR measured each of the first scan data;
a second acquisition step of acquiring a plurality of second scan data measured by the LiDAR while the base portion is rotating and translating due to the mechanical energy of the motor;
a map data generating step of generating the map data based on the reference data generated in the reference data generating step and the plurality of second scan data. A program for causing a computer to execute the map data generation method described in claim 13. a base unit having a LiDAR, an IMU, a mounting unit on which the IMU is mounted and a support unit that supports the LiDAR; and a motor that rotates the base unit, wherein a support surface on which the support unit supports the LiDAR is perpendicular to a mounting surface on which the mounting unit mounts the IMU, and a center of the IMU passes through an axis of rotation of the LiDAR when the base unit rotates due to the mechanical energy of the motor, or a support surface on which the support unit supports the LiDAR is parallel to the mounting surface on which the mounting unit mounts the IMU, and a center of the IMU passes through a line that is perpendicular to the axis of rotation of the LiDAR when the base unit rotates due to the mechanical energy of the motor,
a first acquisition step of acquiring a plurality of first scan data measured by the LiDAR during a predetermined period while the base unit is rotating and not translating due to the mechanical energy of the motor, and movement data of the LiDAR measured by the IMU during the predetermined period when the LiDAR measured each of the plurality of first scan data, in order to generate reference data to be referenced when generating the map data;
a reference data generation step of generating the reference data based on the plurality of first scan data and the movement data of the LiDAR when the LiDAR measured each of the first scan data;
a second acquisition step of acquiring a plurality of second scan data measured by the LiDAR while the base portion is rotating and translating due to the mechanical energy of the motor;
a map data generating step of generating the map data based on the reference data generated in the reference data generating step and the plurality of second scan data. A program for causing a computer to execute the map data generation method described in claim 15.