JP7209755B2 - CALIBRATION METHOD, PROGRAM AND CALIBRATION DEVICE - Google Patents

Description

The present disclosure relates to calibration methods, programs, and calibration devices.

A moving object may be equipped with a sensor that detects its surroundings. Japanese Patent Laid-Open No. 2002-200000 describes that each sensor provided in a vehicle is calibrated. Patent Literature 1 describes that landmarks provided with reflectors are installed at the intersections of scanning lines with which the sensors scan the ground, and each sensor detects the position of the intersection of the scanning lines. In Patent Document 1, sensor calibration is performed based on the position of the intersection of scanning lines in the coordinate system of each sensor.

Japanese Patent Application Laid-Open No. 2020-064011

However, in Patent Document 1, it is necessary to perform the work of specifying the position of the intersection of the scanning lines using a dedicated device for specifying the position of the intersection of the scanning lines, which increases the burden on the operator. . Therefore, when calibrating a plurality of sensors provided on a moving body, it is required to reduce the burden on the operator.

An object of the present disclosure is to solve the problems described above, and to provide a calibration method, a program, and a calibration device that can reduce the burden on workers.

In order to solve the above-described problems and achieve the object, a calibration method according to the present disclosure is a method for calibrating a plurality of sensors provided in a moving body for detecting the surroundings, and obtaining inter-sensor position information, which is information on the relative positions and relative orientations of the sensors, based on the position information of the same measurement object. and acquiring reference sensor position information, which is information on the position and orientation of a reference sensor, which is a reference sensor among the plurality of sensors, with respect to the reference position of the moving body, based on the design information of the moving body. and acquiring sensor position information, which is information on the position and orientation of each of the sensors with respect to the reference position of the moving object, based on the inter-sensor position information and the reference sensor position information.

In order to solve the above-described problems and achieve an object, a program according to the present disclosure is a program for causing a computer to execute a calibration method for a plurality of sensors provided in a mobile body for detecting the surroundings, the program comprising: a step of causing at least two of the sensors to detect the position information of the same measurement object; a step of acquiring position information; and a reference sensor that is position and orientation information of a reference sensor that is a reference sensor among the plurality of sensors with respect to a reference position of the moving body based on the design information of the moving body. a step of acquiring position information; and a step of acquiring sensor position information, which is information on the position and orientation of each of the sensors with respect to the reference position of the moving object, based on the inter-sensor position information and the reference sensor position information. , is executed by the computer.

In order to solve the above-described problems and achieve the object, a calibration device according to the present disclosure is a calibration device for a plurality of sensors provided in a moving body for detecting the surroundings, wherein among the plurality of sensors, A measurement object position acquisition unit that causes at least two sensors of the above to detect the position information of the same measurement object, and a sensor that is information on the relative position and relative orientation of the sensors based on the position information of the same measurement object an inter-sensor position information acquisition unit that acquires inter-position information; and a position and orientation of a reference sensor, which is a reference sensor among the plurality of sensors, with respect to a reference position of the moving body based on design information of the moving body. a reference sensor position information acquisition unit that acquires reference sensor position information that is information of the position and orientation of each of the sensors with respect to the reference position of the moving object based on the inter-sensor position information and the reference sensor position information; a sensor position information acquisition unit that acquires sensor position information, which is information.

In order to solve the above-described problems and achieve the object, a calibration method according to the present disclosure is a method of calibrating a sensor that is provided in a moving body and detects the surroundings. fixing the position and orientation of the reference position of the moving body; causing the sensor to detect position information of an object to be measured whose position and orientation with respect to the reference position of the jig are known; Obtaining sensor position information, which is information on the position and orientation of the sensor with respect to the reference position of the moving body, based on the detection result of the position information of the object and the position information of the measurement object with respect to the reference position of the moving body. and

According to the present disclosure, the burden on workers can be reduced.

FIG. 1 is a schematic diagram of a moving body according to the present embodiment. FIG. 2 is a schematic diagram explaining the detection range of the sensor of the moving object. FIG. 3 is a schematic block diagram of the calibration device according to this embodiment. FIG. 4 is a schematic diagram for explaining detection of position information of a measurement object. FIG. 5 is a schematic diagram for explaining calculation of inter-sensor position information. FIG. 6 is a schematic diagram for explaining reference sensor position information. FIG. 7 is a schematic diagram for explaining the second pass. FIG. 8 is a flowchart for explaining the flow of calibration processing according to this embodiment. FIG. 9 is a schematic diagram illustrating another example of calculation of reference sensor position information. FIG. 10 is a schematic diagram illustrating another example of calculation of reference sensor position information. FIG. 11 is a diagram for explaining an example of the shape of the object to be measured. 12A and 12B are schematic diagrams for explaining a method of detecting a measurement object. 13A and 13B are diagrams for explaining a method of calculating the measurement object position information. FIG. 14 is a diagram for explaining calculation of a conversion amount for converting from the coordinate system of one sensor to the coordinate system of the other sensor. FIG. 15 is a diagram for explaining another example of conversion amount calculation. FIG. 16 is a schematic diagram for explaining calculation of inter-sensor position information. FIG. 17 is a schematic diagram for explaining a point group selection method. FIG. 18 is a flowchart for explaining a point group selection method according to the fourth embodiment. FIG. 19 is a schematic diagram of a jig according to the fifth embodiment. FIG. 20 is a flowchart for explaining the flow of calibration processing according to the fifth embodiment.

Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that the present invention is not limited by this embodiment, and when there are a plurality of embodiments, the present invention includes a combination of each embodiment.

(First embodiment)
(moving body)
FIG. 1 is a schematic diagram of a moving body according to the present embodiment. In this embodiment, the moving body 10 is an automatically movable device. More specifically, the moving body 10 is a forklift, more specifically, a so-called AGF (Automated Guided Forklift). However, the moving body 10 may be any movable device, and is not limited to a forklift. Further, the moving body 10 is not limited to a device that moves automatically, and may be a device that moves by being driven by the user.

(Composition of moving body)
As shown in FIG. 1, the moving body 10 includes a vehicle body 20, a mast 22, a fork 24, a plurality of sensors 26, and a control device 28. The vehicle body 20 has wheels 20A. The mast 22 is provided at one end of the vehicle body 20 in the front-rear direction. The mast 22 extends along the vertical direction orthogonal to the front-rear direction. A fork 24 is mounted movably in the direction Z on the mast 22 . The fork 24 may also be movable with respect to the mast 22 in the lateral direction of the vehicle body 20 (the direction intersecting the vertical direction and the front-rear direction). The fork 24 has a pair of claws 24A, 24B. The claws 24A and 24B extend from the mast 22 toward the front of the vehicle body 20. As shown in FIG. The claws 24A and 24B are spaced apart from each other in the lateral direction of the mast 22. As shown in FIG. Hereinafter, among the front-to-rear directions, the direction on which the fork 24 is provided in the moving body 10 is referred to as the forward direction, and the direction on which the fork 24 is not provided is referred to as the rearward direction.

The control device 28 controls movement of the moving body 10 . The control device 28 is a computer and includes, for example, a storage section and a control section (not shown). A control unit of the control device 28 is an arithmetic unit, that is, a CPU (Central Processing Unit), and executes processing by reading and executing a program (software) from a storage unit. The processing of the control unit includes, for example, processing for automatically moving the moving body 10 .

(sensor)
The sensor 26 detects at least one of the position and orientation of an object existing around the vehicle body 20 . It can also be said that the sensor 26 detects the position of the object with respect to the moving body 10 and the posture of the object with respect to the moving body 10 . In this embodiment, the sensors 26 are provided at the mast 22 and the four corners of the vehicle body 20 , that is, at the left and right ends of the vehicle body 20 in the front direction and the left and right ends of the vehicle body 20 in the rear direction. That is, four sensors 26A, 26B, 26C, and 26D are provided as the sensors 26 in this embodiment. The sensors 26A, 26B are provided at the tip of the support portion 25 facing the fork 24, and the sensors 26C, 26D are provided on the rearward side of the vehicle body 20. As shown in FIG. However, the position where the sensor 26 is provided is not limited to this, and may be provided at any position. Also, the number of sensors 26 is not limited to four, and may be arbitrary. Further, for example, a safety sensor provided on the moving body 10 may be used as the sensor 26 . Diversion of the safety sensor eliminates the need to provide a new sensor.

FIG. 2 is a schematic diagram explaining the detection range of the sensor of the moving object. As shown in FIG. 2, hereinafter, the upward direction in the vertical direction is defined as a direction Z, one direction orthogonal to the direction Z is defined as a direction X, and one direction orthogonal to the directions X and Z is defined as a direction Y. As shown in FIG. In the present embodiment, it is preferable that the floor surface on which the moving body 10 travels extends along the X direction and the Y direction. In this embodiment, unless otherwise specified, "position" refers to a position (coordinates) on a plane perpendicular to the Z direction, and "attitude" refers to an orientation in the plane perpendicular to the Z direction. (orientation of the object when the object is viewed from the Z direction).

The sensor 26 detects the position and orientation of the object by detecting (receiving) reflected light from surrounding objects. Furthermore, the sensor 26 is a sensor that emits laser light. The sensor 26 detects the position and orientation of the object by detecting the reflected light of the irradiated laser beam. The sensor 26 emits laser light while scanning in one direction, and detects the position and orientation of the object from the reflected light of the emitted laser light. That is, it can be said that the sensor 26 is a so-called 2D-LiDAR (Light Detection And Ranging). In this embodiment, the sensor 26 scans the laser beam horizontally, ie in a direction perpendicular to the Z direction. In the present embodiment, each sensor 26 is arranged so that its optical axis lies along a plane orthogonal to the Z direction, that is, the sensor 26 faces the horizontal direction instead of facing vertically upward or downward. 10 is preferably attached. In other words, it is preferable that the vertical axis of the sensor 26 and the vertical axis of the moving body 10 are oriented in the same direction.

In this embodiment, the detection ranges of adjacent sensors 26 overlap. The detection range refers to the range in which the sensor 26 can irradiate laser light, in other words, the range in which the position and orientation of the target object can be detected. That is, the sensor 26 can detect the position and orientation of the object by receiving reflected light reflected from the object within the detection range. In the example of FIG. 2, the detection range CRA of the sensor 26A and the detection range CRB of the sensor 26B partially overlap, and the detection range CRB of the sensor 26B and the detection range CRC of the sensor 26C partially overlap. The detection range CRC of the sensor 26C and the detection range CRD of the sensor 26D partially overlap, and the detection range CRD of the sensor 26D and the detection range CRA of the sensor 26A partially overlap. Both the sensors 26A and 26B are capable of detecting the position and orientation of the object in the region where the detection range CRA and the detection range CRB overlap. , 26C can detect the position and orientation of the object, and in the region where the detection range CRC and the detection range CRD overlap, both the sensors 26C and 26D can detect the position and orientation of the object. Yes, and in a region where the detection range CRD and the detection range CRA overlap, both the sensors 26D and 26A can detect the position and orientation of the object. In the example of FIG. 2, only the detection ranges of two adjacent sensors 26 overlap, but the detection ranges of three or more sensors 26 may overlap.

In addition, in this embodiment, it is preferable that the detection range of all the sensors 26 is combined so that the entire area around the moving body 10 can be detected. That is, for example, it is preferable that the circumference CI at a position radially outwardly spaced from the center of the moving body 10 by a predetermined distance when viewed in the Z direction is within the detection range of at least one sensor 26 over the entire section. .

Note that the sensor 26 is not limited to the above, and may be a sensor that detects an object by any method. For example, it may be a so-called 3D-LiDAR that scans in multiple directions, or it may be a camera. may

(Calibration device)
The calibration device 30 according to this embodiment is a device that performs calibration of the sensor 26 provided on the moving body 10 as described above. That is, the calibration device 30 is a device that executes the calibration method according to this embodiment. The calibration of the sensor 26 is to calculate the position and orientation of the sensor 26 mounted on the mobile body 10 with respect to the reference position P of the mobile body 10 (relative position and relative orientation between the reference position P and the sensor 26), 26 is aligned with the coordinate system of the reference position P of the moving body 10. FIG. For example, in the example of FIG. 2, the coordinates of the reference position P and the reference direction AXP are set in advance in the coordinate system of the moving body 10 . In the example of FIG. 2, the reference position P is a position between the forks 24A and 24B, but may be any position on the moving body 10. FIG. In the example of FIG. 2, the reference direction AXP is the traveling direction of the moving body 10, but may be any direction with respect to the moving body 10. FIG. Hereinafter, the position of the sensor 26A is defined as a position PA, the direction in which the sensor 26A is facing is defined as a direction AXA, the position of the sensor 26B is defined as a position PB, the direction in which the sensor 26B is facing is defined as a direction AXB, and the position of the sensor 26C is defined as a position Let PC be the direction in which the sensor 26C faces, direction AXC be the direction in which the sensor 26C is facing, position PD be the position of the sensor 26D, and direction AXD be the direction in which the sensor 26D is facing. In the example of FIG. 2, the direction in which each sensor 26 faces is the direction along the central line of the detection range when viewed from the Z direction, but it is not limited thereto.

The calibration device 30 performs calibration to convert the coordinates of the position PA of the sensor 26A with respect to the coordinates of the reference position P (coordinates of the position PA in the coordinate system of the moving body 10) to the position of the sensor 26A with respect to the reference position P. Then, the direction AXA in which the sensor 26A faces with respect to the reference direction AXP (orientation of the direction AXA in the coordinate system of the moving body 10) is calculated as the attitude of the sensor 26A with respect to the reference position P. Similarly, the calibration device 30 calculates the coordinates of the position PB of the sensor 26B with respect to the coordinates of the reference position P (coordinates of the position PB in the coordinate system of the moving body 10) as the position of the sensor 26B with respect to the reference position P, A direction AXB in which the sensor 26B faces with respect to the direction AXP (orientation of the direction AXB in the coordinate system of the moving body 10) is calculated as the attitude of the sensor 26B with respect to the reference position P. Further, the calibration device 30 calculates the coordinates of the position PC of the sensor 26C with respect to the coordinates of the reference position P (coordinates of the position PC in the coordinate system of the moving body 10) as the position of the sensor 26C with respect to the reference position P, A direction AXC in which the sensor 26C faces with respect to AXP (a direction of the direction AXC in the coordinate system of the moving body 10) is calculated as an orientation of the sensor 26C with respect to the reference position P. FIG. Further, the calibration device 30 calculates the coordinates of the position PD of the sensor 26D with respect to the coordinates of the reference position P (the coordinates of the position PD in the coordinate system of the moving body 10) as the position of the sensor 26D with respect to the reference position P, and calculates the position of the sensor 26D with respect to the reference direction. A direction AXD in which the sensor 26D is facing with respect to AXP (orientation of the direction AXD in the coordinate system of the moving body 10) is calculated as the orientation of the sensor 26D with respect to the reference position P.

(Configuration of calibration device)
FIG. 3 is a schematic block diagram of the calibration device according to this embodiment. The calibration device 30 is a computer, and includes a storage section 32, a communication section 34, and a control section 36, as shown in FIG. The storage unit 32 is a memory that stores various kinds of information such as calculation contents and programs of the control unit 36, and includes, for example, a main storage device such as a RAM (Random Access Memory), a ROM (Read Only Memory), At least one of an external storage device such as an HDD (Hard Disk Drive) is included. The program for the control unit 36 stored in the storage unit 32 may be stored in a recording medium readable by the calibration device 30 . The communication unit 34 is a module that is used by the control unit 36 to communicate with an external device such as the mobile unit 10, and may include an antenna, for example. The communication method by the communication unit 34 is wireless communication in this embodiment, but any communication method may be used. Note that the calibration device 30 may include an input unit that receives user input and an output unit that outputs information. Examples of the input unit include a touch panel, a keyboard, and a mouse, and examples of the output unit include a display for displaying images.

The control unit 36 is an arithmetic unit, that is, a CPU. The control unit 36 includes a measurement object position acquisition unit 40 , an inter-sensor position information acquisition unit 42 , a reference sensor position information acquisition unit 44 , and a sensor position information acquisition unit 46 . The control unit 36 reads out and executes a program (software) from the storage unit 32 to obtain a measurement object position acquisition unit 40, an inter-sensor position information acquisition unit 42, a reference sensor position information acquisition unit 44, and a sensor position information acquisition unit. 46 and perform those processes. Note that the control unit 36 may execute these processes by one CPU, or may be provided with a plurality of CPUs and may execute the processes by the plurality of CPUs. Moreover, at least a part of the measurement object position acquisition unit 40, the inter-sensor position information acquisition unit 42, the reference sensor position information acquisition unit 44, and the sensor position information acquisition unit 46 may be realized by hardware.

(Measurement object position acquisition unit)
The measurement object position acquisition unit 40 causes the sensor 26 to detect the position information of the measurement object B. FIG. More specifically, the measurement object position acquisition unit 40 causes the sensor 26 to irradiate the surface Ba of the measurement object B with laser light, and the sensor 26 receives the reflected light of the laser light from the surface Ba of the measurement object B. is received. The measurement object position acquisition unit 40 calculates the position information of the measurement object B based on the information of the reflected light of the laser beam from the measurement object B received by the sensor 26 . The position information of the measurement object B refers to information of the position of the measurement object B (coordinates of the measurement object B in the coordinate system of the sensor 26) with respect to the sensor 26 that detected the position information of the measurement object B. The position information of the measurement object B may also include information on the attitude of the measurement object B with respect to the sensor 26 . Hereinafter, the position information of the measurement object B will be referred to as measurement object position information as appropriate. Note that the measurement object B is a member used for calibration.

The measurement object position acquisition unit 40 causes at least two sensors 26 out of the plurality of sensors 26 to detect the position information of the same measurement object B. FIG. A specific description will be given below.

FIG. 4 is a schematic diagram for explaining detection of position information of a measurement object. As shown in FIG. 4 , the measurement object position acquisition unit 40 causes the sensor 26 to detect the measurement object position information while the measurement object B is placed near the moving body 10 . More specifically, the measurement object position acquiring unit 40 causes the sensors 26 to detect the measurement object position information in a state where the measurement object B is arranged in an area where the detection ranges of the adjacent sensors 26 overlap. In the example of FIG. 4, the measurement object B1 is arranged at a position where the detection ranges of the sensors 26A and 26B overlap, and the measurement object B2 is arranged at a position where the detection ranges of the sensors 26B and 26C overlap, A measuring object B3 is arranged at a position where the detection ranges of the sensors 26C and 26D overlap.

The measurement object position acquisition unit 40 causes at least two sensors 26 out of the plurality of sensors 26 to detect the same measurement object B. FIG. That is, the measurement object position acquisition unit 40 causes at least two sensors 26 out of the plurality of sensors 26 to irradiate the same measurement object B with the laser light L, and the laser light L from the measurement object B By receiving the reflected light, the measurement object position information of the same measurement object B is acquired. In the example of this embodiment, the measurement object position acquisition unit 40 causes the two sensors 26 to detect the same measurement object B. FIG. That is, the measurement object position acquisition unit 40 acquires the measurement object position information of the measurement object B detected by one sensor 26 and the measurement object position information of the same measurement object B detected by the other sensor 26. and get. In the example of FIG. 4, the measurement object position acquisition unit 40 causes the sensor 26A and the sensor 26B to irradiate the measurement object B1 with the laser light L, and obtains the reflected light of the laser light L from the measurement object B1. By receiving light, the measurement object position information of the measurement object B1 is acquired. The measurement object position acquisition unit 40 obtains the position information of the measurement object B1 with respect to the sensor 26A detected by the sensor 26A and the measurement object position information with respect to the sensor 26B detected by the sensor 26B as the measurement object position information about the measurement object B1. B1 position information (the position of the measurement object B1 with respect to the sensor 26B) is calculated. Similarly, the measurement object position acquisition unit 40 causes the sensors 26B and 26C to irradiate the measurement object B2 with the laser light L and receive the reflected light of the laser light L from the measurement object B2. , the measurement object position information of the measurement object B2 is acquired. The measurement object position acquisition unit 40 obtains the position information of the measurement object B2 with respect to the sensor 26B detected by the sensor 26B and the measurement object position information with respect to the sensor 26C detected by the sensor 26C as the measurement object position information of the measurement object B2. The position information of B2 is calculated. Similarly, the measurement object position acquisition unit 40 causes the sensors 26C and 26D to irradiate the measurement object B3 with the laser light L and receive the reflected light of the laser light L from the measurement object B3. , the measurement object position information of the same measurement object B3 is acquired. The measurement object position acquiring unit 40 obtains the position information of the measurement object B3 with respect to the sensor 26C detected by the sensor 26C and the measurement object B3 with respect to the sensor 26D detected by the sensor 26D as the measurement object position information of the measurement object B3. The position information of B3 is calculated.

Thus, in this embodiment, the sensors 26A and 26B are caused to detect the measurement object B1, and the sensors 26B and 26C are caused to detect the measurement object B2. However, it is not limited to this, and the same measurement object B may be detected by a pair of sensors 26 and another pair of sensors 26 . Also, in the present embodiment, two adjacent sensors 26 are caused to detect the same measurement object B, but the present invention is not limited to this, and three or more sensors 26 may be caused to detect the same measurement object B.

(Inter-sensor position information acquisition unit)
FIG. 5 is a schematic diagram for explaining calculation of inter-sensor position information. The inter-sensor position information acquisition unit 42 acquires inter-sensor position information, which is information on the relative positions and relative orientations of the sensors 26 based on the measurement object position information of the same measurement object B. FIG. The inter-sensor position information acquisition unit 42 calculates inter-sensor position information between the sensors 26 that have detected the same measurement object B based on the measurement object position information of the same measurement object B. FIG. For example, in the example of the present embodiment, the inter-sensor position information acquisition unit 42 is based on the measurement object position information of the measurement object B1, in other words, the position information of the measurement object B1 detected by the sensor 26A and Inter-sensor position information between the sensors 26A and 26B is calculated based on the detected position information of the measurement object B1. Since the sensors 26A and 26B detect the same measurement object B1, the position of the measurement object B1 relative to the sensor 26A detected by the sensor 26A and the position of the measurement object B1 relative to the sensor 26B detected by the sensor 26B are , the position and orientation of the sensor 26B with respect to the sensor 26A can be calculated. The inter-sensor position information acquisition unit 42 calculates the position and orientation of the sensor 26B with respect to the sensor 26A (the position and orientation of the sensor 26B in the coordinate system of the sensor 26A) as inter-sensor position information between the sensors 26A and 26B. Furthermore, as shown in FIG. 5, the inter-sensor position information acquisition unit 42 uses a transformation for converting the coordinate system of the sensor 26B into the coordinate system of the sensor 26A as inter-sensor position information between the sensor 26A and the sensor 26B. Calculate the quantities (t _XB , t _YB , θ _B ). Note that _tXB is the amount of displacement in the X-axis direction of the coordinate system of sensor 26B with respect to the coordinate system of sensor 26A, _tYB is the amount of displacement in the Y-axis direction of the coordinate system of sensor 26B with respect to the coordinate system of sensor 26A, θ _B is the amount of rotation of the coordinate system of sensor 26B with respect to the coordinate system of sensor 26A.

Similarly, the inter-sensor position information acquisition unit 42 obtains the position information of the measurement object B2 detected by the sensor 26B and the position information of the measurement object B2 detected by the sensor 26C based on the measurement object position information of the measurement object B2. Inter-sensor position information between the sensor 26B and the sensor 26C is calculated based on the position information of the sensor 26B and the sensor 26C. The inter-sensor position information acquisition unit 42 calculates the position and orientation of the sensor 26C with respect to the sensor 26B (the position and orientation of the sensor 26C in the coordinate system of the sensor 26B) as inter-sensor position information between the sensors 26B and 26C. In other words, the inter-sensor position information acquisition unit 42 obtains the inter-sensor position information between the sensors 26B and 26C by converting the coordinate system of the sensor 26C into the coordinate system of the sensor 26B (t _XC , t _YC , θ _C ). Similarly, the inter-sensor position information acquisition unit 42 obtains the position information of the measurement object B3 detected by the sensor 26C and the position information of the measurement object B3 detected by the sensor 26D based on the measurement object position information of the measurement object B3. Inter-sensor position information between the sensor 26C and the sensor 26D is calculated based on the position information of the sensor 26C and the sensor 26D. The inter-sensor position information acquisition unit 42 calculates the position and orientation of the sensor 26D with respect to the sensor 26C (the position and orientation of the sensor 26D in the coordinate system of the sensor 26C) as inter-sensor position information between the sensors 26C and 26D. Furthermore, the inter-sensor position information acquiring unit 42 obtains the inter-sensor position information between the sensor 26C and the sensor 26D by converting the coordinate system of the sensor 26D into the coordinate system of the sensor 26C (t _XD , t _YD , θ _D ).

(Reference sensor position information acquisition unit)
FIG. 6 is a schematic diagram for explaining reference sensor position information. The reference sensor position information acquisition unit 44 acquires reference sensor position information, which is information on the position and orientation of the reference sensor with respect to the reference position P of the moving body 10 . The reference sensor position information can also be said to be information on the position and orientation of the reference sensor in the coordinate system of the moving body 10 . A reference sensor is a sensor 26 that serves as a reference among the plurality of sensors 26 . In this embodiment, the sensor 26A is used as the reference sensor, but the sensor used as the reference sensor may be arbitrarily selected. The reference sensor position information acquisition unit 44 acquires reference sensor position information based on the design information of the moving body 10 . The reference sensor position information acquisition unit 44 may read the design information of the moving body 10 stored in advance in the storage unit 32, or may receive the design information of the moving body 10 from an external device via the communication unit 34. may be obtained. The design information of the moving body 10 is information including design values such as dimensions of the moving body 10. In the present embodiment, the mounting position (coordinates) and mounting angle (attitude) of the reference sensor to the moving body 10 are specified. Includes design values. More specifically, in the present embodiment, the design information of the moving body 10 includes the position of the reference sensor relative to the reference position P of the moving body 10 in the coordinate system of the moving body 10 (in this example, the position PA of the sensor 26A); It contains information about the angle of the direction in which the reference sensor faces (the direction AXA in this example) with respect to the reference direction AXP of the moving body 10 . The reference sensor position information acquisition unit 44 obtains the position (coordinates) of the reference sensor with respect to the reference position P of the mobile body 10 in the coordinate system of the mobile body 10 and the angle at which the reference sensor faces with respect to the reference direction AXP of the mobile body 10. is acquired as the reference sensor position information. In the example of FIG. 6, the reference sensor position information acquisition unit 44 uses the conversion amounts (t _XA , t _YA , θ _A ) are calculated. t _XA is a value for converting the X coordinate in the coordinate system of the sensor 26A into the X coordinate in the coordinate system of the moving body 10, and t _YA is the value for converting the Y coordinate in the coordinate system of the sensor 26A into the coordinate system of the moving body 10. , and θ _A is a value for converting the orientation (angle) of the sensor 26A in the coordinate system into the orientation (angle) of the moving body 10 in the coordinate system.

(Sensor position information acquisition unit)
FIG. 7 is a schematic diagram for explaining calculation of sensor position information. Based on the inter-sensor position information acquired by the inter-sensor position information acquisition unit 42 and the reference sensor position information acquired by the reference sensor position information acquisition unit 44, the sensor position information acquisition unit 46 determines the position of the moving body 10 relative to the reference position P. , to acquire sensor position information, which is information on the position and orientation of each sensor 26 . The sensor position information can be said to indicate the position and orientation of each sensor 26 in the coordinate system of the moving body 10, and can also be said to be a transformation amount for transforming the coordinate system of each sensor 26 into the coordinate system of the moving body 10. I can say. The reference sensor position information indicates the position and orientation of the reference sensor with respect to the reference position P of the moving body 10, and the inter-sensor position information indicates the position and orientation of the other sensor with respect to the reference sensor. The position and orientation of each sensor 26 relative to the reference position P can be calculated based on the information and inter-sensor position information.

In the present embodiment, the sensor position information acquisition unit 46 uses the reference sensor position information, that is, the conversion amounts (t _XA , t _YA , θ _A ) here as the sensor position information of the sensor 26A (reference sensor).

Further, in the present embodiment, the sensor position information acquisition unit 46 obtains sensor position information of the sensor 26B (in position and orientation of the sensor 26B). For example, as shown in FIG. 7, the sensor position information acquisition unit 46 obtains a value obtained by multiplying the conversion amount (t _XB , t _YB , θ _B ) by the conversion amount (t _XA , t _YA , θ _A ) in this order. , as sensor position information of the sensor 26B.

Further, in the present embodiment, the sensor position information acquisition unit 46, based on inter-sensor position information between the sensors 26A and 26B, inter-sensor position information between the sensors 26B and 26C, and reference sensor position information, Sensor position information of the sensor 26C (the position and orientation of the sensor 26C in the coordinate system of the moving body 10) is calculated. For example, the sensor position information acquisition unit 46 obtains the conversion amounts (t _XC , t _YC , θ _C ), the conversion amounts (t _XB , t _YB , θ _B ), and the conversion amounts (t _XA , t _YA , θ _A ). The value multiplied in this order is calculated as the sensor position information of the sensor 26C.

Further, in the present embodiment, the sensor position information acquisition unit 46 obtains inter-sensor position information between the sensors 26A and 26B, inter-sensor position information between the sensors 26B and 26C, and inter-sensor position information between the sensors 26C and 26D. Based on the position information and the reference sensor position information, sensor position information of the sensor 26D (the position and orientation of the sensor 26D in the coordinate system of the moving body 10) is calculated. For example, the sensor position information acquisition unit 46 converts the conversion amounts (t _XD , t _YD , θ _D ), the conversion amounts (t _XC , t _YC , θ _C ), the conversion amounts (t _XB , t _YB , θ _B ) and the conversion A value obtained by multiplying the amount (t _XA , t _YA , θ _A ) in this order is calculated as the sensor position information of the sensor 26D.

The sensor position information acquisition unit 46 causes the storage unit 32 to store the calculated sensor position information of each sensor 26 . Further, the sensor position information acquisition unit 46 may transmit the calculated sensor position information of each sensor 26 to the moving body 10 or another device via the communication unit 34 . This completes the calibration process for each sensor 26 . In the above description, coordinate conversion is performed in the order of sensor 26D, sensor 26C, sensor 26B, sensor 26A, and reference position P, but the order of coordinate conversion is not limited to this and may be arbitrary.

(Calibration flow)
A flow of calibration processing by the calibration device 30 described above will be described. FIG. 8 is a flowchart for explaining the flow of calibration processing according to this embodiment. As shown in FIG. 8, the calibration device 30 causes the pair of sensors 26 to detect the same measurement object B by the measurement object position acquisition unit 40, and calculates measurement object position information (step S10). The measurement object position acquisition unit 40 causes each pair of sensors 26 , that is, each two adjacent sensors 26 to detect the same measurement object B. FIG. The measurement object position acquisition unit 40 calculates measurement object position information of the measurement object from the detection results of the same measurement object B by the pair of sensors 26 .

The calibration device 30 uses the inter-sensor position information acquisition unit 42 to calculate inter-sensor position information between the sensors 26 based on the measurement object position information (step S12). The inter-sensor position information acquisition unit 42 calculates inter-sensor position information between the sensors 26 that detect the measurement object B based on the measurement object position information of the same measurement object B. FIG. The inter-sensor position information acquisition unit 42 calculates inter-sensor position information for each pair of sensors 26 .

The calibration device 30 acquires the reference sensor position information by the reference sensor position information acquisition unit 44 based on the design information of the moving body 10 (step S14). It should be noted that step S14 is not limited to be executed after steps S10 and S12, and the execution order of step S14 and steps S10 and S12 may be arbitrary. For example, step S14 may be performed before step S10.

Next, the calibration device 30 acquires the sensor position information of each sensor 26 (each sensor relative to the reference position P of the moving body 10) based on the inter-sensor position information and the reference sensor position information by the sensor position information acquisition unit 46 26) are calculated (step S16). This completes the calibration process.

(effect)
As described above, the calibration device 30 according to the present embodiment causes at least two sensors 26 to detect the same measurement object B, and based on the detection results, the relative positions and orientations of the sensors 26 (inter-sensor position information), and from the relative position and orientation of the sensors 26 and the relative position and orientation of the reference sensor with respect to the reference position P acquired from the design information (reference sensor position information), the position of each sensor 26 with respect to the reference position P Calculate the relative position and orientation. According to this embodiment, it is possible to calculate the relative position and orientation of each sensor 26 with respect to the reference position P simply by causing the pair of sensors 26 to detect the same object B to be measured. Therefore, according to the present embodiment, if the measurement object B is within the detection range of the pair of sensors 26, each sensor 26 can be calibrated. This eliminates the need for equipment, and prevents an increase in the work load for identifying the installation position of the measurement object B, thereby reducing the burden on the operator. Moreover, according to the present embodiment, it is possible to reduce the degree of dependence of the calibration accuracy on the skill level of the operator, and to suppress deterioration of the calibration accuracy.

In the example of the present embodiment, the calibration device 30 is not mounted on the moving body 10 and is a device separate from the moving body 10, but is not limited to this, and may be mounted on the moving body 10. good too. That is, the control device 28 of the moving body 10 may perform the processing of the calibration device 30 as the calibration device 30 .

(another example)
In the above description, the reference sensor position information was acquired only from the design information of the moving body 10, but it is not limited to this. For example, the reference sensor position information acquisition unit 44 may calculate the reference sensor position information based on the design information of the moving body 10 and inter-sensor position information between the reference sensor and the other sensor 26 . A specific description will be given below.

9 and 10 are schematic diagrams illustrating another example of calculation of reference sensor position information. 9 shows an example of a method of calculating the position of the sensor 26A (reference sensor) with respect to the reference position P. FIG. As shown in FIG. 9, the reference sensor position information acquisition unit 44 calculates the coordinates of the middle point P0 between the position PA of the sensor 26A and the position PD of the sensor 26D based on the inter-sensor position information. For example, the reference sensor position information acquisition unit 44, based on inter-sensor position information between the sensors 26A and 26B, inter-sensor position information between the sensors 26B and 26C, and inter-sensor position information between the sensors 26C and 26D, The coordinates of the position PD of the sensor 26D in the coordinate system of the sensor 26A are calculated. The reference sensor position information acquisition unit 44 obtains the coordinates of the midpoint P0 in the coordinate system of the sensor 26A from the coordinates of the position PA of the sensor 26A in the coordinate system of the sensor 26A and the coordinates of the position PD of the sensor 26D in the coordinate system of the sensor 26A. Calculate The distance D1 in FIG. 9 indicates half the distance between the positions PA and PD. Then, the reference sensor position information acquisition unit 44 acquires information on the reference direction AXP and information on the distance D2 from the design information. A distance D2 refers to the distance between the position PA of the sensor 26A and the reference position P in the direction along the reference direction AXP. The reference sensor position information acquiring unit 44 calculates, as the coordinates of the reference position P, a position separated by a distance D2 along the reference direction AXP from the coordinates of the midpoint P0. The reference sensor position information acquiring unit 44 calculates the position of the sensor 26A with respect to the reference position P from the calculated coordinates of the reference position P. FIG.

FIG. 10 shows an example of a method of calculating the orientation of the sensor 26A (reference sensor) with respect to the reference direction AXP. As shown in FIG. 10 , the reference sensor position information acquisition unit 44 calculates the angle α between the reference direction AXP and the line segment L0 from the design information of the moving body 10 . A line segment L0 is a line segment connecting the position PA of the sensor 26A and the position PD of the sensor 26D. The reference sensor position information acquisition unit 44 calculates the direction of the line segment L0 based on the coordinates of the position PA of the sensor 26A and the coordinates of the position PD of the sensor 26D acquired from the design information of the moving body 10. FIG. Then, the reference sensor position information obtaining unit 44 obtains the information of the reference direction AXP obtained from the design information of the moving body 10, and calculates the angle α between the reference direction AXP and the line segment L0. Further, the reference sensor position information acquisition unit 44 calculates an angle β formed between the direction AXA in which the sensor 26A is facing and the line segment L0 based on the inter-sensor position information. Based on inter-sensor position information between the sensors 26A and 26B, inter-sensor position information between the sensors 26B and 26C, and inter-sensor position information between the sensors 26C and 26D, the reference sensor position information acquisition unit 44 acquires the sensor 26A , and the direction of the line segment L0 is calculated from the coordinates of the position PA of the sensor 26 and the coordinates of the position PD of the sensor 26D. Then, the reference sensor position information acquiring unit 44 calculates an angle between the direction of the calculated line segment L0 and the direction AXA as an angle β. The reference sensor position information acquisition unit 44 calculates the orientation of the sensor 26A with respect to the reference direction AXP based on the angles α and β. That is, the reference sensor position information acquiring unit 44 calculates an angle obtained by adding the angle α and the angle β as the orientation of the sensor 26A with respect to the reference direction AXP.

As described above, by calculating the reference sensor position information based on the inter-sensor position information between the reference sensor and the other sensor 26, for example, even if the position or orientation of the reference sensor deviates from the design information, Since the position and orientation of the actual reference sensor can be reflected, calibration accuracy can be improved. The method of calculating the inter-sensor position information based on the inter-sensor position information between the reference sensor and the other sensor 26 described above is just an example.

(Second embodiment)
Next, a second embodiment will be described. The second embodiment differs from the first embodiment in that the shape of the object B to be measured is known. In the second embodiment, descriptions of the parts that are common to the first embodiment will be omitted.

FIG. 11 is a diagram for explaining an example of the shape of the object to be measured. In the second embodiment, the shape of the surface Ba of the measuring object B is known. That is, the calibration device 30 can acquire information about the shape of the surface Ba of the measurement object B. FIG. The object B to be measured may have any shape, and may be planar, for example. That is, the measurement object B may have a linear surface Ba when viewed from the Z direction, as shown on the left side of FIG. Further, for example, the measurement object B may have a shape in which planes intersect. That is, as shown in the center of FIG. 11, the measurement object B may have a surface Ba when viewed from the Z direction, in which straight lines intersect each other. Moreover, the measurement object B may be cylindrical. That is, as shown in the right side of FIG. 11, the measurement object B may have an arcuate surface Ba when viewed from the Z direction.

(Method for calculating position information of measurement object)
Next, a method of calculating the measurement object position information in the second embodiment will be described. 12A and 12B are schematic diagrams for explaining a method of detecting a measurement object. As shown in FIG. 12, the measurement object position acquisition unit 40 causes the sensor 26 to scan and irradiate the surface Ba of the measurement object B with the laser light L. As shown in FIG. That is, the measurement object position acquisition unit 40 irradiates the laser light L to each different position on the surface Ba. The measurement object position acquisition unit 40 causes the sensor 26 to receive the laser light L reflected at each position on the surface Ba as reflected light. Hereinafter, the position where the laser beam L is reflected on the surface Ba (the position where the laser beam L is irradiated) will be referred to as a reflection position as appropriate. The measurement object position acquisition unit 40 calculates the coordinates of each reflection position based on the information of each reflected light received by the sensor 26, and uses them as measurement object position information. Note that FIG. 12 illustrates a case where the same measurement object B1 is detected by the sensors 26A and 26B.

13A and 13B are diagrams for explaining a method of calculating the measurement object position information. (A) of FIG. 13 shows an example of the coordinates of the reflection position detected by the sensor 26A. That is, FIG. 13A shows the case where the sensor 26A irradiates the measurement object B1 with the laser light L as shown in FIG. 12 and receives the reflected light reflected at each reflection position of the measurement object B1. 4 shows a diagram in which the coordinates of the reflection positions calculated by the measurement object position acquisition unit 40 are plotted as a point group M. FIG. That is, the point group M can be said to be points (measurement points) indicating the coordinates of the reflection position. The horizontal axis in (A) of FIG. 13 indicates the X coordinate in the coordinate system of the sensor 26A, and the vertical axis indicates the Y coordinate in the coordinate system of the sensor 26A. The measurement object position acquisition unit 40 calculates the coordinates of each point group M as the measurement object position information of the measurement object B1. Further, for example, the measurement object position acquisition unit 40 generates a line segment LI connecting each point group M, and obtains the coordinates and direction of the line segment LI from the measurement object position information (the position and orientation of the measurement object B). information). The line segment LI is not limited to being a line segment passing through the point group M, and may be an approximation line between the point groups M, for example. The line segment LI may be generated by any method based on the coordinates of the point cloud M.

In this way, the measurement object position acquiring unit 40 causes the sensor 26 to detect the light reflected by the measurement object B, and based on the information of the light detected by the sensor 26, the sensor detects the reflection position of the measurement object B. It can be said that the information of the coordinates in the 26 coordinate system is calculated as the position information of the object B to be measured.

Here, the sensor 26 may irradiate the laser light L toward a portion other than the measurement object B and detect the reflected light from the portion other than the measurement object B. However, if the measurement object position information is calculated based on the reflected light reflected from a place other than the measurement object B, the error with respect to the surface Ba of the actual measurement object B becomes large. There is a possibility that the calculation accuracy of the inter-sensor position information calculated based on the information may be degraded. Therefore, the measurement object position acquisition unit 40 according to the present embodiment selects the point group M (coordinates of the reflection positions) used for calculating the inter-sensor position information based on the shape of the surface Ba of the measurement object B. do. Specifically, the measurement object position acquisition unit 40 extracts a plurality of point groups M corresponding to the shape of the surface Ba from the detected point group M, and obtains the coordinates of the plurality of extracted point groups M as This is the measurement object position information. Any method can be used to select the point group M corresponding to the shape of the surface Ba. A point group M having a predetermined value or less may be selected as the point group M corresponding to the shape of the surface Ba.

Note that the shape of the measurement object B is not limited to being known. In this case, the measurement object position acquisition unit 40 does not have to perform the process of extracting a plurality of point groups M corresponding to the shape of the surface Ba. In this case, for example, since the approximate position of the measurement object B with respect to the sensor 26 is known, the point group M at the approximate position of the measurement object B is selected from the detected point cloud M as the surface Ba , and the coordinates of the extracted point group M may be used as the measurement object position information.

Further, in the present embodiment, it is preferable to cause the pair of sensors 26 to detect the same measurement object B multiple times. For example, the measurement object position acquisition unit 40 causes the sensors 26A and 26B to detect the measurement object B1 for a plurality of scans. Then, at least one of the relative position and orientation of the object B1 to be measured with respect to the moving body 10 is changed, and the sensors 26A and 26B are made to detect the object B1 to be measured by a plurality of scans. In this case, the above processing may be repeated any number of times. The measurement object position acquisition unit 40 calculates the point group M (coordinates of the reflection positions) for each of the plurality of detection data thus detected, and uses the measurement object position information. The measurement object position acquisition unit 40 causes all the pairs of sensors 26 to perform these processes. In this way, by executing the detection of the same measurement object B multiple times and calculating a plurality of measurement object position information for the same measurement object B, the inter-sensor position calculated based on the measurement object position information Information calculation accuracy can be improved.

(Calculation method of inter-sensor position information)
Next, a method for calculating inter-sensor position information in the second embodiment will be described. FIG. 13A shows the point group M detected by the sensor 26A as described above. More specifically, as shown in FIG. The point group M indicating the coordinates of each reflection position on the surface Ba of the measurement object B1 is plotted in the coordinate system of the sensor 26A. On the other hand, (B) of FIG. 13 shows the point group M detected by the sensor 26B. The point group M indicating the coordinates of each reflection position on the surface Ba of the measurement object B1 is plotted in the coordinate system of the sensor 26B. Here, the sensors 26A and 26B detect the same measurement object B1. Therefore, when the point group M detected by the sensor 26A is transformed into the coordinate system of the sensor 26B, those point groups (the point group transformed into the coordinate system of the sensor 26B M) coincides with the line segment LI of the point group M in the coordinate system of the sensor 26B detected by the sensor 26B. The inter-sensor position information acquisition unit 42 according to the second embodiment calculates a conversion amount for converting from the coordinate system of one sensor 26 to the coordinate system of the other sensor 26 based on this principle.

A specific description will be given below. Here, let the coordinates of the reflection position of the measurement object B (for example, the measurement object B1) in the coordinate system of the one sensor 26 (for example, the sensor 26B) detected by the one sensor 26 (for example, the sensor 26B) be the first coordinates. Coordinates obtained by converting the first coordinates into the coordinate system of the other sensor 26 (for example, the sensor 26A) are referred to as first converted coordinates. Also, the coordinates of the same reflection position of the same measurement object B (for example, measurement object B1) in the coordinate system of the other sensor 26 detected by the other sensor 26 are defined as second coordinates. In this case, for example, the inter-sensor position information acquisition unit 42 is configured such that the difference between the first converted coordinates and the second coordinates, that is, the distance between the first converted coordinates and the second coordinates is equal to or less than a predetermined value. A conversion amount (for example, a conversion amount (t _XB , t _YB , θ _B )) from the coordinate system of one sensor 26 to the coordinate system of the other sensor 26 is calculated.

FIG. 14 is a diagram for explaining calculation of a conversion amount for converting from the coordinate system of one sensor to the coordinate system of the other sensor. When calculating the conversion amount from the first conversion coordinates and the second coordinates as described above, for example, as shown in FIG. , laser light L may be scanned. Then, the coordinates of the intersection point MA between the line segments LI of one sensor 26 are set as the first coordinates, and the coordinates of the intersection point MB of the line segments LI of the other sensor 26 are set as the second coordinates. Calculate

A more specific description will be given below by taking as an example a case of calculating the conversion amounts (t _X , t _Y , θ) for converting the coordinate system of the sensor 26B into the coordinate system of the sensor 26A. In the following, the first coordinates, which are the coordinates of the reflection position of the measurement object B1 detected by the sensor 26B, are defined as (x2, y2), and the second coordinates, which are the coordinates of the same reflection position of the measurement object B1 detected by the sensor 26A. be (x1, y1).

When the second coordinates (x1, y1) and the first coordinates (x2, y2) are the same point on the real space, the following equation (1) holds.

Assume that the sensors 26A and 26B are caused to detect the same measurement object B N times, and the first coordinates and the second coordinates are calculated N times. In this case, the inter-sensor position information acquisition unit 42 calculates the conversion amount (t _X , t _Y , θ) such that J (t _x , ty, θ) shown in the following equation (2) is minimized. . That is, the inter-sensor position information acquisition unit 42 uses the method of least squares to minimize the square of the total value of the differences between the first transformed coordinates and the second coordinates, the transformed amount (t _X , t _Y , θ) can be calculated. However, the inter-sensor position information acquisition unit 42 is not limited to using the least squares method. For example, the conversion amount (t _X , t _Y , θ) may be calculated.

Next, another specific example of the conversion amount calculation method will be described. FIG. 15 is a diagram for explaining another example of conversion amount calculation. Here, the line segment LI connecting the point group M of the measurement object B1 in the coordinate system of the one sensor 26 detected by the one sensor 26 is defined as the line segment LIB, and the other sensor 26 detected by the other sensor 26 A line segment LI connecting the point group M of the 26 coordinate systems is assumed to be a line segment LIA. In this case, the inter-sensor position information acquisition unit 42 of this example converts the line segment LIB (the line segment converted into the coordinate system of the other sensor 26) into the coordinate system of the other sensor 26. LIB) and the line segment LIA, the amount of conversion from the coordinate system of one sensor 26 to the coordinate system of the other sensor 26 is calculated such that the difference is equal to or less than a predetermined value. Note that the difference between the line segments LI may refer to, for example, the difference between the matrices of the line segments LI.

Here, it is assumed that the line segment LIA connecting the point group M of the measurement object B1 detected by the sensor 26A is represented by the following equation (3), and the line segment LIA connecting the point group M of the measurement object B1 detected by the sensor 26B is Suppose that the line segment LIB is represented by the following equation (4). In this case, when the line segment LIA and the line segment LIB match in real space, the following equation (5) holds.

Applying the equation (5) to the line segment LIA of the coordinate system of the sensor 26A results in the following equation (6).

Assume that the sensors 26A and 26B are caused to detect the same measurement object B N times, and the line segment LIA and the line segment LIB are calculated N times. In this case, the inter-sensor position information acquisition unit 42 calculates the conversion amount (t _X , t _Y , θ) such that J (t _x , ty, θ) shown in the following equation (7) is minimized. . That is, the inter-sensor position information acquisition unit 42 uses the method of least squares to obtain a transformation amount (t _X , t _Y , θ).

The inter-sensor position information acquisition unit 42 calculates the conversion amount for converting the coordinate system for each pair of sensors 26 by the method described above. However, the conversion amount calculation method described above is an example, and the conversion amount may be calculated by other methods.

(effect)
As described above, in the second embodiment, the shape of the measurement object B is known, and based on the shape of the measurement object B, the point group M (reflection position coordinates). As a result, it is possible to suppress a decrease in the calculation accuracy of the inter-sensor position information. Further, in the second embodiment, when the point group M and the line segment LI detected by one sensor 26 are converted into the coordinate system of the other sensor 26, the point group M and the line segment LI detected by the other sensor 26 are The amount of conversion from the coordinate system of one sensor 26 to the coordinate system of the other sensor 26 is calculated so that the difference from the line segment LI is equal to or less than a predetermined value. Therefore, according to the second embodiment, it is possible to suppress a decrease in the calculation accuracy of the inter-sensor position information.

(Third embodiment)
Next, a third embodiment will be described. The third embodiment differs from the second embodiment in that the amount of conversion from the coordinate system of the sensor 26D to the coordinate system of the sensor 26A is also calculated, that is, in that inter-sensor position information between all sensors is considered. is different. In the third embodiment, descriptions of parts that have the same configuration as in the second embodiment will be omitted. In addition, 3rd Embodiment is applicable also to 1st Embodiment.

In the third embodiment, in order to calculate the amount of conversion from the coordinate system of sensor 26D to the coordinate system of sensor 26A, sensor 26D and sensor 26A are caused to detect the same measurement object B4 (not shown). That is, in the fourth embodiment, the measurement object B4 as the measurement object B is arranged at a position where the detection ranges of the sensor 26D and the sensor 26A overlap. The measurement object position acquisition unit 40 causes the sensor 26D and the sensor 26A to irradiate the measurement object B4 with the laser light L and receive the reflected light of the laser light L from the measurement object B4. The measurement object position information of the measurement object B4 is acquired. The measurement object position acquisition unit 40 calculates the position information of the measurement object B4 with respect to the sensor 26D and the position information of the measurement object B4 with respect to the sensor 26A as the measurement object position information about the measurement object B4.

FIG. 16 is a schematic diagram for explaining calculation of inter-sensor position information. In the third embodiment, the inter-sensor position information acquisition unit 42 is based on the measurement object position information of the measurement object B4. Inter-sensor position information between the sensor 26D and the sensor 26A is calculated based on the position information of the sensor 26D and the sensor 26A. As shown in FIG. 16, the inter-sensor position information acquisition unit 42 obtains the position and orientation of the sensor 26A with respect to the sensor 26D (the position and orientation of the sensor 26A in the coordinate system of the sensor 26D) as inter-sensor position information between the sensor 26D and the sensor 26A. Posture) is calculated. The inter-sensor position information acquisition unit 42 obtains the conversion amounts (t _XA1 , t _YA1 , θ _A1 ) for converting the coordinate system of the sensor 26A into the coordinate system of the sensor 26D as inter-sensor position information between the sensor 26D and the sensor 26A. Calculate

Here, the position and orientation of the sensor 26D in the coordinate system of the sensor 26A are obtained by transforming the position and orientation of the sensor 26D in the coordinate system of the sensor 26D with the transformation amounts (t _XD , t _YD , θ _D ), and furthermore, the transformation amounts It is calculated by transforming with (t _XC , t _YC , θ _C ) and further transforming with the transformation amount (t _XB , t _YB , θ _B ). The position and orientation of the sensor 26D in the coordinate system of the sensor 26A calculated in this manner are converted by the conversion amounts (t _XA1 , t _YA1 , θ _A1 ) into the position and orientation of the sensor 26D in the coordinate system of the sensor 26D. When returned, it preferably matches the position and orientation of sensor 26D in the coordinate system of sensor 26D that was originally detected. That is, the transformation of the coordinate system using each transformation amount is preferably an identity transformation. Therefore, the inter-sensor position information acquisition unit 42 calculates each transformation amount so that the coordinate system transformation approaches the identity transformation, and calculates the inter-sensor position information between the sensors 26 . For example, an arbitrary position and orientation in the coordinate system of the sensor 26D is transformed into a position and orientation in the coordinate system of the sensor 26A by coordinate transformation, and further transformed by transformation amounts (t _XA1 , t _YA1 , θ _A1 ) to The position and orientation when returned to the 26D coordinate system are assumed to be the return position and return orientation. In this case, the inter-sensor position information acquisition unit 42 performs conversion so that the difference between the return position and return orientation and the original position and orientation in the coordinate system of the sensor 26D that has not undergone coordinate conversion is equal to or less than a predetermined value. It is preferred to calculate the amount. A specific example will be described below.

Here, let the transformation amount for transforming the coordinate system between the sensors 26 be (t _ix , t _iy , θ _i ). However, i is either 1, 2, 3, or 4. The transformation amounts (t _1x , t _1y , θ ₁ ) when i is 1 correspond to the transformation amounts (t _XB , t _YB , θ _B ) for transforming the coordinate system of the sensor 26B into the coordinate system of the sensor 26A. Then, the transformation amounts (t _2x , t _2y , θ ₂ ) when i is 2 are the transformation amounts (t _XC , t _YC , θ _C ) for transforming the coordinate system of the sensor 26C into the coordinate system of the sensor 26B. and the transformation amounts (t _3x , t _3y , θ ₃ ) when i is 3 are the transformation amounts (t _XD , t _YD , θ _D ), and the transformation amounts (t _4x , t _4y , θ ₄ ) when i is 4 are the transformation amounts (t _XA1 , t _YA1 , θ _A1 ). In this case, it can be said that the transformation of the coordinate system satisfies the identity transformation when the following equation (8) is satisfied.

The inter-sensor position information acquisition unit 42 according to the _third embodiment minimizes J(t _1x , t _1y , θ ₁ , t _2x , . , the transformation amount (t _ix , t _iy , θ _i ) is calculated. Note that one item (left term) on the right side of Equation (9) is the total value of J(t _x , ty, θ) shown in the first embodiment for each pair of sensors 26 .

(effect)
As described above, the inter-sensor position information acquisition unit 42 according to the third embodiment performs a conversion amount (for example, , the transformation amounts (t _XD , t _YD , θ _D ), the transformation amounts (t _XC , t _YC , θ _C ) and the transformation amounts (t _XB , t _YB , θ _B ), in the coordinate system of the first sensor An arbitrary position and orientation are converted into a position and orientation in the coordinate system of the second sensor, and the amount of conversion from the coordinate system of the second sensor to the coordinate system of the first sensor (for example, the amount of conversion (t _XA1 , t _YA1 , θ _A1 )) is used to transform the position and orientation in the coordinate system of the second sensor into the position and orientation in the coordinate system of the first sensor, the position and orientation in the coordinate system of the first sensor before conversion and the position and orientation in the coordinate system of the first sensor after conversion The amount of transformation from the coordinate system of the first sensor to the coordinate system of the second sensor is set such that the difference between the position and orientation in the coordinate system of the first sensor is equal to or less than a predetermined value. In the embodiment, since the transformation amount is calculated so that the transformation of the coordinate system approaches the identity transformation, it is possible to appropriately suppress the deterioration of the calculation accuracy of the inter-sensor position information.

(Fourth embodiment)
The fourth embodiment differs from the second embodiment in that the point clouds M at both ends in the scanning direction of the laser beam L are excluded and the remaining point clouds M are used as the measurement object position information. . In the fourth embodiment, descriptions of parts having the same configuration as in the second embodiment will be omitted. In addition, 4th Embodiment is applicable also to 1st Embodiment and 3rd Embodiment.

The sensor 26 detects the point group M by irradiating the measurement object B with the laser light L while scanning along the scanning direction. In this case, near the edge of the object B to be measured in the scanning direction, detection may become unstable and the calibration accuracy may decrease. On the other hand, in the fourth embodiment, the reduction in calibration accuracy is suppressed by excluding the point group M near the edge of the measurement object B in the scanning direction. Specific processing will be described below.

FIG. 17 is a schematic diagram for explaining the point group selection method, and FIG. 18 is a flowchart for explaining the point group selection method in the fourth embodiment. The measurement object position acquisition unit 40 causes the sensor 26 to irradiate the laser light L toward different positions on the surface Ba along the scanning direction, and the laser light L is reflected at each reflection position along the scanning direction. Reflected light is received. Therefore, the point group M (coordinates of the reflection positions) is arranged along the scanning direction SC as shown in FIG. In other words, the measurement object position acquisition unit 40 causes the sensor 26 to receive reflected light along the scanning direction SC so that the point group M is aligned in the scanning direction SC. As shown in FIG. 18, when selecting the point group M, the measurement object position acquisition unit 40 first selects a plurality of point groups detected by the sensor 26 based on the shape of the surface Ba of the measurement object B. A point group M to be used is extracted from M (coordinates of reflection positions) (step S20). The measurement object position acquisition unit 40 extracts a plurality of point groups M corresponding to the shape of the surface Ba as point groups M to be used. In the example of FIG. 17, the point group M0 that does not correspond to the shape of the surface Ba is excluded, and only the point group M that corresponds to the shape of the surface Ba is extracted as the point group M to be used.

Next, the measurement object position acquisition unit 40 sequentially assigns numbers to the extracted point group M from both ends in the scanning direction SC (the direction in which the point group M is arranged) (step S22). That is, the measurement object position acquisition unit 40 acquires the point group M1 located on the most one side along the scanning direction SC and the point group M2 located on the othermost side along the scanning direction SC with respect to the extracted point group M. , a point group M3 adjacent to the point group on the othermost side along the scanning direction SC on the other side, a point group M4 adjacent to the point group on the othermost side along the scanning direction SC on one side, A number is assigned to each point group M in the order of . Further, the measurement object position acquisition unit 40 estimates the shape of the measurement object B based on the extracted point group M (step S24). In other words, the measurement object position acquisition unit 40 generates a line segment LI connecting the extracted point group M. FIG. In the example of FIG. 17, the line segment LI is generated from the entire extracted point group M including the point groups M1 to M4. Note that step S24 is not limited to being performed after step S22, and the processing order of steps S22 and S24 may be arbitrary.

After that, the measurement object position acquisition unit 40 removes part of the extracted point group M in numerical order (step S26). The measurement object position acquisition unit 40 removes a predetermined number of point groups M from the point group M having the smallest number, that is, from both end sides along the scanning direction SC. The predetermined number here may be set arbitrarily. Then, the measurement object position acquiring unit 40 estimates the shape of the measurement object B based on the point group M that remains without being removed (step S28). In other words, the measurement object position acquisition unit 40 generates a line segment LI that connects the remaining point group M without being removed. In the example of FIG. 17, the point groups M1 to M4 are removed to generate a line segment LI connecting the point groups M other than the point groups M1 to M4.

The measurement object position acquisition unit 40 determines whether the shape before and after removal differs by a threshold value or more (step S30). That is, the measurement object position acquisition unit 40 determines that the difference between the line segment LI generated from the point group M in step S24 and the line segment LI generated from the point group M remaining without being removed in step S28 is equal to or greater than a threshold. to determine whether The difference between line segments LI may be, for example, the difference between matrices of line segments LI. Also, the threshold may be set arbitrarily. If the shapes before and after removal do not differ by a threshold value or more (step S30; No), that is, if the difference between the line segments LI is less than the threshold value, the remaining point cloud M is used as measurement object position information (step S32). . The inter-sensor position information acquisition unit 42 calculates inter-sensor position information based on the point group M determined to be used as the measurement object position information. That is, when the change in the line segment LI is small before and after removing the point group M at both ends, the measurement object position acquiring unit 40 determines that the point group M whose detection is not stable has been removed, and the remaining points The group M is used as measurement object position information. On the other hand, if the shapes before and after removal differ by a threshold value or more (step S30; Yes), that is, if the difference between the line segments LI is a threshold value or more, the process returns to step S26, and part of the remaining point group M is replaced with a number The process of removing a predetermined number in ascending order is repeated. In this case, in step S30, the measurement object position acquiring unit 40 determines whether the difference between the line segment LI generated last time and the line segment LI generated by further removing the point group M is a threshold value or more. That is, the measurement object position acquisition unit 40 removes a predetermined number of point groups M at both ends while comparing the shapes of the line segments LI each time. The remaining point cloud M is used as measurement object position information.

(effect)
Thus, in the fourth embodiment, inter-sensor position information is calculated based on point groups M other than a predetermined number of point groups M (coordinates of reflection positions) at both ends in the scanning direction SC (predetermined direction). According to the fourth embodiment, by excluding the point group M near the edge of the measurement object B in the scanning direction, it is possible to suppress deterioration in calibration accuracy.

(Fifth embodiment)
Next, a fifth embodiment will be described. The fifth embodiment differs from the first embodiment in that the moving body 10 is fixed to a jig whose position and orientation with respect to the measurement object B are known, and calibration is performed. In the fifth embodiment, descriptions of parts that are common to the first embodiment will be omitted. Note that the fifth embodiment can also be applied to the second to fourth embodiments.

(jig)
FIG. 19 is a schematic diagram of a jig according to the fifth embodiment. A jig 100 is a jig for fixing the position of the moving body 10 . The jig 100 is provided with a fixing portion 100A. The fixing section 100</b>A has a function of fixing the moving body 10 to the jig 100 . For example, the fixed part 100A may be a recess into which a part of the moving body 10 is inserted. In this case, the movable body 10 is fixed in position with respect to the jig 100 by being inserted into the fixing portion 100A, which is partially recessed. In the example of FIG. 19, the fixing portion 100A is a recess into which the tip of the support portion 25 can be inserted, and the moving body 10 is inserted into the fixing portion 100A, which is the recess, so that the jig 100 can be moved. position relative to is fixed.

As described above, since the jig 100 fixes the moving body 10 , in a state where the moving body 10 is fixed to the jig 100 , the reference position PJ of the moving body 10 is shifted from the reference position PJ of the jig 100 . It can be said that the position and orientation of the position P are fixed. In the example of FIG. 19, the reference position PJ overlaps with the reference position P of the moving body 10 when the moving body 10 is fixed to the jig 100, but it is not limited to this.

Further, the jig 100 has a known position and orientation of the measuring object B with respect to the reference position PJ. In the present embodiment, by attaching the measurement object B to the jig 100, the position and orientation of the measurement object B with respect to the reference position PJ of the jig 100 are fixed. However, the measurement object B is not limited to being attached to the jig 100 as long as the position and orientation with respect to the reference position PJ are known. Moreover, the position and orientation of the measurement object B are not limited to being fixed with respect to the reference position PJ of the jig 100 , and may be variable with respect to the reference position PJ of the jig 100 . For example, as shown in FIG. 19, the object B to be measured may be attached to a jig 100 via rails 102 and may be movable along the rails 102 . In this case, for example, the rail 102 is provided with a plurality of stoppers along the extending direction of the rail 102, and the position of the measuring object B is preferably fixed by one of the stoppers. Since the positions and postures of the stoppers are fixed with respect to the reference position PJ of the jig 100, even if the measuring object B is moved, the calibration is executed with the position fixed by one of the stoppers. Thus, calibration can be performed while the position and orientation of the measurement object B with respect to the reference position PJ are known.

It should be noted that the structure of the jig 100 and the attachment state of the measuring object B to the jig 100 described above are examples. For example, the jig 100 may have any structure as long as the position and attitude of the reference position P of the moving body 10 can be fixed with respect to the reference position PJ. Moreover, the mounting state of the measurement object B to the jig 100 may be any one in which the position and orientation of the measurement object B with respect to the reference position PJ are known.

(Calibration processing)
FIG. 20 is a flowchart for explaining the flow of calibration processing according to the fifth embodiment. As shown in FIG. 20, in the fifth embodiment, first, the reference position P of the moving body 10 is fixed with respect to the reference position PJ of the jig 100 (step S40). In this case, for example, by fixing the moving body 10 with the fixing portion 100A of the jig 100, the position and attitude of the reference position P of the moving body 10 are fixed with respect to the reference position PJ of the jig 100. FIG.

Next, in the state where the reference position P of the moving body 10 is fixed with respect to the reference position PJ of the jig 100, the calibration device 30 causes the sensor 26 to detect the position of the measurement object B by the measurement object position acquisition unit 40. Position information (measurement object position information) is detected (step S42). The measurement object position acquisition unit 40 acquires position information of the measurement object B in the coordinate system of the sensor 26 as measurement object position information. The measurement object position acquisition unit 40 causes the sensor 26 to irradiate the laser light L toward the measurement object B in a state where the reference position P of the moving body 10 is fixed with respect to the reference position PJ of the jig 100, Reflected light from the measurement object B is received to calculate position information of the measurement object. This processing is the same as in the first to fourth embodiments.

Next, the calibration device 30 acquires the position information of the measurement object B with respect to the reference position P of the moving body 10 and the measurement object position, which is the position information of the measurement object B with respect to the sensor 26, by the sensor position information acquisition unit 46. Based on the information, sensor position information, which is information on the position and attitude of the sensor 26 with respect to the reference position P of the moving body 10, is obtained (step S44).

The processing of step S44 will be specifically described below. Here, the position of the measurement object B with respect to the reference position P of the moving body 10 is obtained from the position of the reference position P of the moving body 10 with respect to the reference position PJ of the jig 100 and the position of the measurement object B with respect to the reference position PJ. can be calculated. When the moving body 10 is fixed to the jig 100, the position of the reference position P of the moving body 10 with respect to the reference position PJ of the jig 100 is uniquely defined and can be calculated in advance. Further, since the position of the measuring object B with respect to the reference position PJ is also known, it can be calculated in advance. Therefore, it can be said that the position of the measuring object B with respect to the reference position P of the moving body 10 can also be calculated in advance. The sensor position information acquisition unit 46 acquires the position information of the measurement object B with respect to the reference position P of the moving body 10, which has been calculated in advance. It can be said that the sensor position information acquisition unit 46 acquires the position information of the measurement object B in the coordinate system of the mobile body 10 as the position information of the measurement object B with respect to the reference position P of the mobile body 10 .

The sensor position information acquisition unit 46 acquires the position information of the measurement object B in the coordinate system of the moving body 10 and the position information of the measurement object B in the coordinate system of the sensor 26 (measurement object position information) to determine the position of the moving body 10. is calculated as sensor position information. In other words, the sensor position information acquisition unit 46 calculates the conversion amount for converting the coordinate system of the sensor 26 into the coordinate system of the moving body 10 as the sensor position information.

As described above, according to the fifth embodiment, since the calibration is executed with the position and orientation of the moving body 10 fixed with respect to the jig 100, the position of the measuring object B in the coordinate system of the moving body 10 is is known. Therefore, the sensor position information can be calculated and the sensor 26 can be calibrated simply by detecting the position information of the measuring object B in the coordinate system of the sensor 26 . That is, in the third embodiment, since calibration is performed using the jig 100, it is not necessary to use reference sensor position information using design information as in the first embodiment. According to the fifth embodiment, since it is not necessary to use design information, it is possible to suppress deterioration in the calculation accuracy of the sensor position information even when the design information and the actual state are deviated from each other.

Further, in the fifth embodiment, all the sensors 26 may detect the measuring object B whose position is fixed with respect to the jig 100 . In this case, it is possible to directly calculate the sensor position information for each sensor 26 by executing the processing described with reference to FIG. In this case, the calibration between the sensors 26, that is, the calculation of the inter-sensor positional information, as in the first to fourth embodiments, becomes unnecessary. When a plurality of sensors 26 are caused to detect the measuring object B whose position is fixed with respect to the jig 100, the plurality of sensors 26 may be made to detect the same measuring object B, or the jig may A different measuring object B whose position is fixed with respect to 100 may be detected.

However, in the fifth embodiment, at least some of the sensors 26 may be caused to detect the measuring object B whose position is fixed with respect to the jig 100 . In this case, for the sensor 26 that did not detect the measuring object B whose position is fixed with respect to the jig 100, the sensor position information can be calculated by the same method as in the first to fourth embodiments. good. That is, in this case, the sensor 26 whose sensor position information has been calculated by the processing described with reference to FIG. Position information can be calculated, and sensor position information of the other sensors 26 can be calculated based on the inter-sensor position information.

(effect)
As described above, the calibration method according to the present disclosure calibrates the plurality of sensors 26 provided in the moving body 10 to detect the surroundings. This calibration method includes a step of causing at least two sensors 26 out of the plurality of sensors 26 to detect the position information of the same measurement object B (measurement object position information); obtaining inter-sensor position information, which is information on the relative positions and relative orientations of the sensors 26; a step of acquiring reference sensor position information, which is information on the position and orientation of a reference sensor that serves as a reference; obtaining sensor location information, which is 26 position and orientation information. According to this method, it is possible to calculate the relative position and orientation of each sensor 26 with respect to the reference position P simply by causing at least two sensors 26 to detect the same object B to be measured. Therefore, according to this method, if the measurement object B is within the detection range of the pair of sensors 26, each sensor 26 can be calibrated, so that the burden on the operator can be reduced. Moreover, according to this method, it is possible to reduce the degree of dependence of the calibration accuracy on the skill level of the operator, and to suppress deterioration of the calibration accuracy.

In the step of acquiring the reference sensor position information, the reference sensor position information is also acquired based on inter-sensor position information between the reference sensor and the other sensor 26 . According to this method, the reference sensor position information is obtained based on the inter-sensor position information in addition to the design information of the moving body 10, so that deterioration in calibration accuracy can be suppressed.

In the step of detecting the position information of the measurement object B, the light reflected by the measurement object B is detected by the sensor 26, and the reflection position in the coordinate system of the sensor 26 is determined based on the information of the light detected by the sensor 26. The coordinates are calculated, and the position information of the measuring object B is calculated based on the coordinates of the reflection position. According to this method, since the position information of the measurement object B is calculated based on the coordinates of the reflection position of the measurement object B, the sensor 26 can be properly calibrated.

Further, in the present disclosure, the shape of the measurement object B is known, and in the step of acquiring the inter-sensor position information, based on the shape of the measurement object B, the reflection position used for calculating the inter-sensor position information Select coordinates (point cloud M). According to this method, it is possible to suppress a decrease in the calculation accuracy of the inter-sensor position information.

In the step of acquiring inter-sensor position information, the first coordinates of the reflected position in the coordinate system of the first sensor detected by the first sensor (eg, sensor 26B) are converted into the coordinate system of the second sensor (eg, sensor 26B). and the second coordinates of the reflection position in the coordinate system of the second sensor detected by the second sensor are equal to or less than a predetermined value. Sets the amount of transformation to the coordinate system of the sensor. According to this method, since the coordinate system conversion amount is set so that the difference between the first converted coordinates and the second coordinates is equal to or less than the predetermined value, deterioration in the calibration accuracy of the sensor 26 can be suppressed.

In the step of detecting the position information of the measurement object B, the first sensor (eg sensor 26B) and the second sensor (eg sensor 26A) are caused to detect the coordinates of the reflection position multiple times. Then, in the step of acquiring the inter-sensor position information, the conversion amount is set based on the difference between the first conversion coordinates and the second coordinates. According to this method, since the conversion amount is set using a plurality of data, deterioration in the calibration accuracy of the sensor 26 can be suppressed.

In the step of acquiring the inter-sensor position information, an arbitrary to the position and orientation in the coordinate system of the second sensor, and using the amount of transformation from the coordinate system of the second sensor to the coordinate system of the first sensor, the position and orientation in the coordinate system of the second sensor When the orientation is converted into the position and orientation in the coordinate system of the first sensor, the difference between the position and orientation in the coordinate system of the first sensor before conversion and the position and orientation in the coordinate system of the first sensor after conversion is A conversion amount from the coordinate system of the first sensor to the coordinate system of the second sensor is set so as to be equal to or less than a predetermined value. According to this method, since the transformation amount is set so that the transformation of the coordinate system approaches the identity transformation, deterioration in the calibration accuracy of the sensor 26 can be suppressed.

In the step of detecting the position information of the measuring object B, the sensor 26 is caused to detect light along a predetermined direction so that the point group M indicating the coordinates of the reflection position is aligned in the predetermined direction (scanning method SC). Further, in the step of acquiring the inter-sensor position information, the inter-sensor position information is calculated based on the coordinates of the reflection positions other than the predetermined number of point groups M (coordinates of the reflection positions) at both ends in the predetermined direction. According to this method, the reduction in calibration accuracy can be suppressed by excluding the point cloud M near the edge of the measurement object B in the scanning direction.

In the step of acquiring inter-sensor position information, the shape (line segment LI) of the measurement object B specified based on the point group M before removing the coordinates at both ends, and the point after removing the coordinates at both ends It is determined whether the difference from the shape (line segment LI) of the measurement object B specified based on the group M is equal to or greater than a threshold. Then, when the difference is equal to or greater than the threshold, the difference between the shape (line segment LI) of the measurement object B specified before and after the coordinates at both ends are removed is less than the threshold, until the difference after the coordinates at both ends are removed is less than the threshold. The processing of further removing the coordinates of both ends from the point group M is repeated, and inter-sensor position information is calculated based on the remaining point group M when the difference becomes less than the threshold value. According to this method, since the point group M is removed until the specified shape is stabilized, it is possible to appropriately remove the point group M with low accuracy, thereby suppressing deterioration in calibration accuracy.

A program according to the present disclosure causes a computer to execute a method of calibrating a plurality of sensors 26 provided in the mobile object 10 to detect the surroundings. This program comprises a step of causing at least two sensors 26 out of the plurality of sensors 26 to detect position information of the same measurement object B (measurement object position information), and based on the position information of the same measurement object B, a step of obtaining inter-sensor position information, which is information on the relative positions and relative orientations of the sensors 26; a step of acquiring reference sensor position information, which is information on the position and orientation of a reference sensor that is a sensor that is a sensor, and based on the inter-sensor position information and the reference sensor position information, the position of each sensor 26 with respect to the reference position P of the moving body 10 and obtaining sensor position information, which is position and orientation information. According to this program, the burden on workers can be reduced.

The calibration device 30 according to the present disclosure performs calibration of a plurality of sensors 26 that are provided in the mobile body 10 and detect the surroundings. 42 , a reference sensor position information acquisition unit 44 , and a sensor position information acquisition unit 46 . The measurement object position acquisition unit 40 causes at least two sensors 26 out of the plurality of sensors 26 to detect the position information of the same measurement object B (measurement object position information). The inter-sensor position information acquisition unit 42 acquires inter-sensor position information, which is information on the relative positions and relative orientations of the sensors 26 based on the position information of the same measurement object B. FIG. The reference sensor position information acquisition unit 44 is based on the design information of the moving body 10, and is information on the position and orientation of the reference sensor, which is the reference sensor among the plurality of sensors 26, with respect to the reference position P of the moving body 10. Acquire reference sensor position information. The sensor position information acquisition unit 46 acquires sensor position information, which is information on the position and orientation of each sensor 26 with respect to the reference position P of the moving body 10, based on the inter-sensor position information and the reference sensor position information. According to this device, the burden on the operator can be reduced.

A calibration method according to the present disclosure is a method for calibrating the sensor 26 provided in the moving body 10 to detect the surroundings. This calibration method includes steps of fixing the position and orientation of the reference position P of the moving body 10 with respect to the reference position PJ of the jig 10, Based on the step of detecting the known position information of the measurement object B, the detection result of the position information of the measurement object B, and the position information of the measurement object B with respect to the reference position of the mobile object 10, and obtaining sensor position information, which is information of the position and orientation of the sensor 26 with respect to the reference position P. According to this method, the burden on the worker can be reduced.

In addition, the moving body 10 includes a plurality of sensors 26, and the calibration method according to the present disclosure includes the step of causing at least two sensors 26 out of the plurality of sensors 26 to detect the position information of the same object; and obtaining inter-sensor position information, which is information on the relative positions and relative orientations of the sensors 26 based on the position information of the same object. Further, in the step of acquiring the sensor position information, based on the detection result of the position information of the measurement object B, the inter-sensor position information, and the position information of the measurement object B with respect to the reference position P of the moving body 10, acquires sensor position information about the sensor 26 of . According to this method, the burden on the worker can be reduced.

Although the embodiment of the present invention has been described above, the embodiment is not limited by the contents of this embodiment. In addition, the components described above include those that can be easily assumed by those skilled in the art, those that are substantially the same, and those within the so-called equivalent range. Furthermore, the components described above can be combined as appropriate. Furthermore, various omissions, replacements, or modifications of components can be made without departing from the gist of the above-described embodiments.

REFERENCE SIGNS LIST 10 moving body 26, 26A, 26B, 26C, 26D sensor 30 calibration device 40 measurement object position acquisition unit 42 inter-sensor position information acquisition unit 44 reference sensor position information acquisition unit 46 sensor position information acquisition unit B measurement object P reference position

Claims (10)

A method for calibrating a plurality of sensors provided on a moving body for detecting the surroundings,
causing at least two of the plurality of sensors to detect position information of the same measurement object;
obtaining inter-sensor position information, which is information on relative positions and relative orientations of the sensors, based on the position information of the same object to be measured;
Based on design information of the moving body, a position of a reference sensor among the plurality of sensors, which is a reference sensor, with respect to a reference position of the moving body, and a direction in which the reference sensor faces with respect to a reference direction of the moving body. obtaining reference sensor position information, which is information about the angle of
a step of acquiring sensor position information, which is information on the position and orientation of each of the sensors with respect to the reference position of the moving object, based on the inter-sensor position information and the reference sensor position information;
including
In the step of detecting position information of the object to be measured, the sensor is caused to detect light reflected by the object to be measured, and based on the information of the light detected by the sensor, the measurement is performed in the coordinate system of the sensor. calculating coordinates of a reflection position, which is a portion of an object where the light is reflected, and calculating position information of the measurement object based on the coordinates of the reflection position;
In the step of acquiring the inter-sensor position information,
selecting the coordinates of the reflection position used to calculate the inter-sensor position information;
First converted coordinates obtained by converting the first coordinates of the reflection position in the coordinate system of the first sensor detected by the first sensor into the coordinate system of the second sensor, and the second sensor detected by the second sensor. setting the amount of conversion from the coordinate system of the first sensor to the coordinate system of the second sensor such that the difference between the second coordinate of the reflection position in the coordinate system of is equal to or less than a predetermined value;
calibration method. 2. The calibration method according to claim 1, wherein in the step of acquiring the reference sensor position information, the reference sensor position information is also acquired based on the inter-sensor position information between the reference sensor and the other sensor. The measurement object has a known shape,
The calibration method according to claim 1 , wherein in the step of acquiring the inter-sensor position information, the coordinates of the reflection position used for calculating the inter-sensor position information are selected based on the shape of the measurement object. . In the step of detecting the position information of the measurement object, the first sensor and the second sensor are caused to detect the coordinates of the reflection position a plurality of times,
2. The calibration method according to claim 1 , wherein in the step of acquiring the inter-sensor position information, the conversion amount is set based on a difference between each of the first conversion coordinates and the second coordinates. In the step of acquiring the inter-sensor position information, an arbitrary position and orientation in the coordinate system of the first sensor is obtained by using a transformation amount from the coordinate system of the first sensor to the coordinate system of the second sensor. transforming the position and orientation in the coordinate system of the second sensor, and using the amount of transformation from the coordinate system of the second sensor to the coordinate system of the first sensor, transforming the position and orientation in the coordinate system of the second sensor into the When converted to the position and orientation in the coordinate system of the first sensor, the difference between the position and orientation in the coordinate system of the first sensor before conversion and the position and orientation in the coordinate system of the first sensor after conversion is 5. The calibration method according to claim 1 , wherein the amount of transformation from the coordinate system of said first sensor to the coordinate system of said second sensor is set so as to be equal to or less than a predetermined value. In the step of detecting the position information of the object to be measured, causing the sensor to detect light along the predetermined direction so that the point group indicating the coordinates of the reflection position is aligned in the predetermined direction;
In the step of acquiring the inter-sensor position information, the inter-sensor position information is calculated based on the coordinates of the reflection positions other than the coordinates of the reflection positions of a predetermined number at both ends in the predetermined direction. Item 6. The calibration method according to any one of items 5 . In the step of acquiring the inter-sensor position information,
The difference between the shape of the measurement object specified based on the point group before the coordinates of both ends are removed and the shape of the measurement object specified based on the point group after the coordinates of both ends are removed , to determine if it is greater than or equal to the threshold,
When the difference is equal to or greater than the threshold, the point cloud after removing the coordinates at both ends is further removed until the difference in the shape of the measurement object specified before and after the coordinates at both ends are removed is less than the threshold. 7. The calibration method according to claim 6 , wherein the process of removing coordinates is repeated, and the inter-sensor position information is calculated based on points remaining when the difference is less than a threshold. A program for causing a computer to execute a method of calibrating a plurality of sensors provided in a mobile body for detecting the surroundings,
causing at least two of the plurality of sensors to detect position information of the same measurement object;
obtaining inter-sensor position information, which is information on relative positions and relative orientations of the sensors, based on the position information of the same object to be measured;
Based on design information of the moving body, a position of a reference sensor among the plurality of sensors, which is a reference sensor, with respect to a reference position of the moving body, and a direction in which the reference sensor faces with respect to a reference direction of the moving body. obtaining reference sensor position information, which is information about the angle of
a step of acquiring sensor position information, which is information on the position and orientation of each of the sensors with respect to the reference position of the moving object, based on the inter-sensor position information and the reference sensor position information;
including
In the step of detecting position information of the object to be measured, the sensor is caused to detect light reflected by the object to be measured, and based on the information of the light detected by the sensor, the measurement is performed in the coordinate system of the sensor. calculating coordinates of a reflection position, which is a portion of an object where the light is reflected, and calculating position information of the measurement object based on the coordinates of the reflection position;
In the step of acquiring the inter-sensor position information,
selecting the coordinates of the reflection position used to calculate the inter-sensor position information;
First converted coordinates obtained by converting the first coordinates of the reflection position in the coordinate system of the first sensor detected by the first sensor into the coordinate system of the second sensor, and the second sensor detected by the second sensor. setting the amount of conversion from the coordinate system of the first sensor to the coordinate system of the second sensor such that the difference between the second coordinate of the reflection position in the coordinate system of is equal to or less than a predetermined value; make the computer run
program. A calibration device for a plurality of sensors provided in a moving body to detect the surroundings,
a measurement object position acquisition unit that causes at least two sensors among the plurality of sensors to detect position information of the same measurement object;
an inter-sensor position information acquisition unit that acquires inter-sensor position information, which is information on the relative positions and relative orientations of the sensors, based on the position information of the same measurement object;
Based on design information of the moving body, a position of a reference sensor among the plurality of sensors, which is a reference sensor, with respect to a reference position of the moving body, and a direction in which the reference sensor faces with respect to a reference direction of the moving body. A reference sensor position information acquisition unit that acquires reference sensor position information that is information about the angle of
a sensor position information acquisition unit that acquires sensor position information, which is information on the position and orientation of each of the sensors with respect to the reference position of the moving object, based on the inter-sensor position information and the reference sensor position information;
including
The measurement object position information acquisition unit causes the sensor to detect the light reflected by the measurement object, and based on the information of the light detected by the sensor, the measurement object position information acquisition unit in the coordinate system of the sensor. calculating the coordinates of the reflection position where the light is reflected, and calculating the position information of the measurement object based on the coordinates of the reflection position;
The inter-sensor position information acquisition unit
selecting the coordinates of the reflection position used to calculate the inter-sensor position information;
First converted coordinates obtained by converting the first coordinates of the reflection position in the coordinate system of the first sensor detected by the first sensor into the coordinate system of the second sensor, and the second sensor detected by the second sensor. setting the amount of conversion from the coordinate system of the first sensor to the coordinate system of the second sensor such that the difference between the second coordinate of the reflection position in the coordinate system of is equal to or less than a predetermined value;
calibration device. A method for calibrating a sensor provided in a moving body for detecting the surroundings,
a step of fixing the position and orientation of the reference position of the moving body with respect to the reference position of the jig;
a step of causing the sensor to detect position information of an object whose position and orientation with respect to the reference position of the jig are known;
A sensor position, which is information on the position and orientation of the sensor with respect to the reference position of the mobile object, based on the detection result of the position information of the measurement object and the position information of the measurement object with respect to the reference position of the mobile object. obtaining information;
A calibration method according to any one of claims 1 to 6 , comprising:

Images