JPWO2020148899A1 - Radar device and position / orientation measurement method - Google Patents

Abstract

The first estimated surface including the point cloud corresponding to the surface portion of the shape model (40) is calculated to judge whether it is correct or not, and the boundary of the second estimated surface determined to be correct is calculated and expanded outward to the outside. A second correspondence data showing the correspondence relationship between the point cloud included in the surface determined by the boundary expanded to and the surface portion of the shape model (40) is calculated, and the shape model and the point cloud are calculated based on the second correspondence data. Perform the fitting.

Description

The present invention relates to a radar device for observing a target with a plurality of antennas and a position / orientation measuring method for measuring the position and orientation of each of the plurality of antennas.

A distributed radar device including a plurality of antennas generates a composite beam pattern in which the beam patterns of the plurality of antennas are combined in consideration of the position and orientation of each antenna. Therefore, if there is an error in the position or orientation of each antenna, an error also occurs in the composite beam pattern, and the performance of the radar device deteriorates. Therefore, a distributed radar device requires the exact position and orientation of each antenna.

For example, the position / orientation measuring device described in Patent Document 1 fits three-dimensional point cloud data on the surface of an object whose position and orientation are to be measured to a known three-dimensional shape of the object represented by the shape model. .. By fitting the 3D point cloud data measured from the actual object to the known 3D shape of the object, it is possible to measure the position and the posture of the actual object.

Japanese Unexamined Patent Publication No. 2016-170050

In order to fit the shape represented by the shape model and the point cloud measured by the actual object, it is necessary to accurately determine which part of the shape the point cloud corresponds to. However, in the position / orientation measuring device described in Patent Document 1, how to obtain the correspondence between the known shape of the object represented by the shape model and the point cloud measured by the actual object is fully considered. It may not be possible to fit these.

The present invention solves the above problems, and an object of the present invention is to obtain a radar device and a position / orientation measurement method capable of accurately obtaining the correspondence between the shape of the antenna represented by the shape model and the point cloud measured by the antenna. And.

The radar device according to the present invention includes a plurality of antennas and a control device for controlling each beam pattern of the plurality of antennas, and the control device is a plurality of filter units provided corresponding to each of the plurality of antennas. It also has a plurality of arithmetic units and a shape model input unit for inputting a shape model representing the shape of the antenna corresponding to each filter unit for each of the plurality of filter units, and each of the plurality of filter units has a shape model input unit. , The point cloud measured in the area including the shape model input by the shape model input unit is extracted from the point cloud which is the measurement point of a large number of three-dimensional coordinates, and each of the plurality of calculation units is represented by the shape model. For the surface portion constituting the shape of the antenna, the first correspondence data showing the correspondence relationship between the point cloud and the surface portion is calculated based on the perpendicular line drawn from the point cloud extracted by the filter unit corresponding to the antenna. Based on the first correspondence data, the estimated surface including the point cloud corresponding to the surface portion is calculated to judge the correctness, the boundary of the estimated surface determined to be correct is calculated and expanded to the outside, and expanded to the outside. A second correspondence data showing the correspondence relationship between the point cloud and the surface portion included in the surface determined by the boundary is calculated, and the shape model and the point cloud are fitted based on the second correspondence data.

According to the present invention, an estimated surface including a point cloud corresponding to a surface portion of the shape represented by the shape model is calculated to determine whether it is correct or not, and the boundary of the estimated surface determined to be correct is calculated and expanded outward to the outside. Correspondence data showing the correspondence between the point cloud and the surface portion included in the surface determined by the boundary expanded to is calculated, and the shape model and the point cloud are fitted based on the correspondence data. As a result, the correspondence between the shape of the antenna represented by the shape model and the point cloud measured by the antenna can be accurately obtained.

It is a block diagram which shows the structure of the radar apparatus which concerns on Embodiment 1. FIG. It is a block diagram which shows the structure of the antenna in Embodiment 1. FIG. It is a block diagram which shows the structure of the control device in Embodiment 1. FIG. It is a figure which shows the example of the shape model. It is a flowchart which shows the position | posture measurement method which concerns on Embodiment 1. It is a figure which shows the point group, the shape model, and the perpendicular line drawn from the measurement point of a point group with respect to the surface part which constitutes the shape shown by a shape model. It is a figure which shows the example of the 1st correspondence data. It is a figure which shows the face part and the point cloud which make up the shape which the shape model which has a correct correspondence relationship shows. It is a figure which shows the result of fitting the point group in the correct correspondence relationship with respect to the surface part which constitutes the shape shown by a shape model. It is a figure which shows the boundary of the 2nd estimation plane, the point cloud in this boundary, and the shape model. It is a figure which shows the boundary enlarged to the outside, the point cloud in this boundary, and the shape model. It is a figure which shows the correspondence relationship between the point cloud included in the face part determined by the boundary enlarged to the outside, and the face part which constitutes the shape shown by a shape model. It is a figure which shows the result of fitting the measurement point in an erroneous correspondence with respect to the face part which constitutes the shape shown by a shape model. It is a block diagram which shows the structure of the modification of the antenna provided in the radar apparatus which concerns on Embodiment 1. FIG. FIG. 15A is a block diagram showing a hardware configuration that realizes the function of the radar device according to the first embodiment. FIG. 15B is a block diagram showing a hardware configuration for executing software that realizes the functions of the radar device according to the first embodiment. It is a block diagram which shows the structure of the control device provided in the radar device which concerns on Embodiment 2. FIG. It is a block diagram which shows the structure of the control device provided in the radar device which concerns on Embodiment 3. FIG. It is a block diagram which shows the structure of the control device provided in the radar device which concerns on Embodiment 4. FIG.

Embodiment 1.
FIG. 1 is a block diagram showing a configuration of a radar device according to the first embodiment. FIG. 2 is a block diagram showing the configuration of the antenna according to the first embodiment. FIG. 3 is a block diagram showing a configuration of the control device according to the first embodiment. The radar device shown in FIG. 1 includes antennas 1-1, 1-2, ..., 1-N, a control device 2, and a coordinate measuring device 3. N is the number of antennas. The position of each antenna is, for example, the position of the center of the antenna, and the posture of each antenna is, for example, the azimuth and elevation of the antenna with respect to the position of the antenna.

Antennas 1-1, 1-2, ..., 1-N are a group of distributed antennas, and each antenna is connected to the control device 2. The control device 2 controls each beam pattern of the antennas 1-1, 1-2, ..., 1-N. The coordinate measuring device 3 is a device that measures three-dimensional coordinates of a region including each of the antennas 1-1, 1-2, ..., 1-N at multiple points, and measures a large number of objects from the object in the region. Measure a point cloud that is a point. The coordinate measuring device 3 is, for example, a laser scanner.

As shown in FIG. 2, each of the antennas 1-1, 1-2, ..., 1-N includes a transmitting unit 10, a receiving unit 11, and a driving unit 12. The transmission unit 10 converts an electric signal (transmission signal) input from the control device 2 into radio waves, and radiates the converted radio waves into space. The receiving unit 11 converts the radio wave received from the space into an electric signal (received signal), and outputs the converted electric signal to the control device 2. The drive unit 12 drives the antenna to change the position and orientation of the antenna. For example, the drive unit 12 changes the position and orientation of the antenna instructed by the control device 2 by driving the antenna with the drive amount instructed by the control device 2.

As shown in FIG. 3, the control device 2 includes a position / orientation setting unit 20, a shape model input unit 21, a filter unit 22-1,22-2, ..., 22-N, and a calculation unit 23-1,23-. 2, ..., 23-N and a drive amount calculation unit 24 are provided. Each of the filter units 22-1,22-2, ..., 22-N is provided corresponding to the antennas 1-1, 1-2, ..., 1-N, respectively, and the arithmetic unit 23-1. , 23-2, ..., 23-N are provided corresponding to the antennas 1-1, 1-2, ..., 1-N, respectively.

The position / orientation setting unit 20 sets a rough position and orientation of the antenna corresponding to each filter unit for each of the filter units 22-1, 22, 2, ..., 22-N. For example, the approximate position and orientation values of antennas 1-1, 1-2, ..., 1-N are obtained by the user using a measure, a laser rangefinder, or a protractor. To. Then, the position / orientation setting unit 20 sets the rough position and orientation of the antenna acquired by the user in the filter unit corresponding to the antenna.

The shape model input unit 21 inputs a shape model representing a known shape of the antenna corresponding to each filter unit to each of the filter units 22-1, 22, 2, ..., 22-N. The shape of the antenna represented by the shape model indicates the ideal position and orientation of the antenna so that the beam pattern of the antenna becomes a desired pattern. The ideal position and orientation of the antenna is calculated using, for example, electromagnetic field simulation. Further, in the shape model, the three-dimensional shape of the antenna is expressed by the shape in which a plurality of face portions are combined.

The shape model may be, for example, a file described in the IGES format of CAD data. The shape model input unit 21 reads a file of the shape model corresponding to each of the antennas 1-1, 1-2, ..., 1-N from an external device, and reads the read file into the filter units 22-1, 22, 22. Output to -2, ..., 22-N, respectively.

FIG. 4 is a diagram showing a shape model 40. The shape model 40 shown in FIG. 4 is a rectangular parallelepiped composed of six face portions (1) to (6) and represents the shape of the antenna. The shape model in the first embodiment can express a more complicated shape than the rectangular parallelepiped, but in the following, for the sake of simplicity, the three-dimensional shape of the antenna 1-k (1 ≦ k ≦ N) is the rectangular parallelepiped shown in FIG. It is assumed that the shape model of the antenna 1-k is the shape model 40.

Each of the filter units 22-1, 22, 2, ..., 22-N is input by the shape model input unit 21 from a point cloud which is a large number of measurement points of three-dimensional coordinates measured by the coordinate measuring device 3. The point cloud measured in the region containing the shape model is extracted. For example, the filter unit 22-1 selects a point cloud measured in a region including the shape model corresponding to the antenna 1-1 input by the shape model input unit 21 from the point cloud measured by the coordinate measuring device 3. Extract.

Each of the calculation units 23-1,23-2, ..., 23-N subtracts from the point cloud extracted by the filter unit corresponding to the antenna with respect to the surface portion constituting the shape of the antenna represented by the shape model. Based on the vertical line, the first correspondence data showing the correspondence relationship between the point cloud and the surface portion is calculated. Subsequently, each of the calculation units 23-1, 23-2, ..., 23-N calculates a first estimated surface including a point cloud corresponding to the surface unit based on the first corresponding data. The correctness is determined, and the boundary of the second estimated surface is calculated and expanded outward, using the first estimated surface determined to be correct as the second estimated surface. After that, each of the calculation units 23-1, 23-2, ..., 23-N is the second correspondence data showing the correspondence relationship between the surface portion determined by the outwardly expanded boundary and the measurement point of the point cloud. Is calculated, and the shape model and the point cloud are fitted based on the second corresponding data.

The drive amount calculation unit 24 calculates the drive amount of each antenna based on the fitting results of each of the calculation units 23-1, 23-2, ..., 23-N, and calculates the drive amount of each antenna. It is set in the drive unit 12 of the antenna of. For example, the drive amount calculation unit 24 has a known shape (a shape indicating the ideal position and orientation of the antenna 1-k) indicated by the shape model of the antenna 1-k and an antenna as a fitting result by the calculation unit 23-k. The current accurate position and orientation of 1-k are input, and the drive amount of the antenna 1-k is calculated so that there is no deviation between the two. Then, the drive amount calculation unit 24 sets the calculated drive amount in the drive unit 12 of the antenna 1-k. The drive unit 12 drives the antenna 1-k with the drive amount set by the drive amount calculation unit 24, so that the antenna 1-k has the position and orientation of the antenna 1-k corresponding to the shape represented by the shape model. Become. As a result, the beam pattern of the antenna 1-k becomes a desired pattern.

Next, the operation of the radar device according to the first embodiment will be described.
The position / orientation setting unit 20 sets a rough position and orientation of the antenna corresponding to each filter unit for each of the filter units 22-1, 22, 2, ..., 22-N. Further, the shape model input unit 21 inputs the shape model of the antenna corresponding to each filter unit to each of the filter units 22-1, 22, 2, ..., 22-N. Either the rough position and orientation setting of the antenna by the position / orientation setting unit 20 and the input of the shape model by the shape model input unit 21 may be performed first, or may be performed in parallel.

Each of the filter units 22-1, 22, 2, ..., 22-N is measured from the point cloud measured by the coordinate measuring device 3 in the region including the shape model input by the shape model input unit 21. Extract the point cloud. For example, when the shape of the antenna 1-1 is the rectangular parallelepiped shown in FIG. 4, the maximum value of the x-coordinates of the vertices of the rectangular parallelepiped shown by the shape model of the antenna 1-1 is x- _max, and the minimum value of the x-coordinates of the vertices is set to xmax. Let x _min , the maximum value of the y coordinate be y _max , the minimum value of the y coordinate of the vertex be y _min , the maximum value of the z coordinate be z _max, and the minimum value of the z coordinate of the vertex be z _min . _{The filter unit 22-1 has a range of x min} ≤ x ≤ x _max, a range of y _min ≤ y ≤ y _max , and a range of z _min ≤ z ≤ z _max from the point cloud measured by the coordinate measuring device 3. The point cloud included in the three-dimensional coordinate area with reference to is extracted. A certain margin may be provided in the above-mentioned range of the x-coordinate, the above-mentioned range of the y-coordinate, and the above-mentioned range of the z-coordinate that determines the three-dimensional coordinate region.

Further, each of the filter units 22-1, 22, 2, ..., 22-N is input from the coordinate measuring device 3 based on the rough position and orientation of the antenna set by the position / orientation setting unit 20. From the point group which is the measurement point of a large number of three-dimensional position coordinates, the point group of the measurement points measured in the region including the shape model input by the shape model input unit 21 may be extracted. For example, when the measurement error of the rough position and orientation of the antenna 1-k is delta, the filter unit 22-k has x _min − delta ≦ x ≦ x _{max + delta from the point cloud measured by the coordinate measuring device 3.} The point cloud included in the three-dimensional coordinate region determined by the range of, y _{min −} delta ≦ y ≦ y _max + delta and the range of z _{min −} delta ≦ z ≦ z _{max + delta may be extracted.} In this way, by giving a margin of delta to the above-mentioned range of the x-coordinate, the above-mentioned range of the y-coordinate, and the above-mentioned range of the z-coordinate that determines the three-dimensional coordinate region, the point cloud is extracted more than necessary. It can be prevented from being done.

In the point cloud extraction process by each of the filter units 22-1, 22, 2, ..., 22-N, as described above, the rough position and orientation of the antenna set from the position / orientation setting unit 20 are used. It may not be used. In this case, the control device 2 does not have to include the position / orientation setting unit 20.

Point clouds extracted by the filter unit corresponding to each calculation unit are input to each of the calculation units 23-1, 23-2, ..., 23-N, and are formed by the input point groups. Calculate the deviation between the shape of the antenna and the shape of the antenna represented by the shape model. For example, the calculation unit 23-k (1 ≦ k ≦ N) obtains the correspondence between the point cloud extracted by the filter unit 22-k and the surface portion forming the shape represented by the shape model of the antenna 1-k. The point cloud and the shape model are fitted based on this correspondence. As a result, if there is no installation error in the antenna 1-k, the shape formed by the point cloud matches the shape represented by the shape model, but if there is an installation error in the antenna 1-k, the shape represented by the shape model will be obtained. On the other hand, the shape formed by the point cloud shifts.

The fitting results of each of the calculation units 23-1, 23-2, ..., 23-N are input to the drive amount calculation unit 24. The drive amount calculation unit 24 corresponds to, for example, the ideal position and orientation of the antenna 1-k corresponding to the shape represented by the shape model and the shape formed by the point cloud as the fitting result by the calculation unit 23-k. Input the deviation from the position and orientation of the antenna 1-k. The drive amount calculation unit 24 calculates the drive amount of the antenna 1-k in which the antenna 1-k is in the ideal position and orientation based on this deviation, and the antenna 1-k calculates the calculated antenna drive amount. It is set in the drive unit 12 to have. The drive unit 12 drives the antenna 1-k with the drive amount set by the drive amount calculation unit 24, so that the antenna 1-k is adjusted to an ideal position and posture.

Next, the detailed operation of the calculation unit, that is, the position / orientation measurement method according to the first embodiment will be described. FIG. 5 is a flowchart showing a position / posture measurement method according to the first embodiment.
The calculation unit 23-k is a perpendicular line drawn from each measurement point of the point cloud extracted by the filter unit 22-k to a surface portion (hereinafter, simply referred to as a surface portion of the shape model) constituting the shape indicated by the shape model. Is calculated (step ST1). The position of the antenna indicated by the shape model is represented by local coordinates in step ST1, but is converted into global coordinates based on the position and orientation of the antenna updated in step ST10 described later.

FIG. 6 is a diagram showing perpendicular lines drawn from the measurement points of the point group with respect to the point cloud, the shape model 40, and the surface portion (1) of the shape model 40. As described above, the known three-dimensional shape of the antenna 1-k represented by the shape model 40 is a shape showing an ideal position and orientation in which the beam pattern of the antenna 1-k becomes a desired pattern. The actual position and orientation of the antenna 1-k are often deviated due to the installation error of the antenna. Therefore, the calculation unit 23-k calculates a perpendicular line drawn from the measurement point of the point cloud with respect to the surface portion (1) of the shape model 40. When a plurality of perpendicular lines are calculated from one measurement point, the perpendicular line having the shortest length among the calculated perpendicular lines is defined as the perpendicular line drawn from the measurement point to the surface portion.

For example, as shown in FIG. 6, the calculation unit 23-k can calculate a perpendicular line p1 drawn on the surface portion (1) from each of many measurement points d1 included in the same plane.
However, as shown by reference numeral A, although the measurement point d1 is not included in the plane where the measurement point d1 exists, the perpendicular line p2 may be erroneously calculated from the measurement point d2 on the surface portion (1). In this case, it may be erroneously determined that the measurement point d2 that does not correspond to the surface portion (1) has a correspondence relationship with the surface portion (1). Further, when the shape formed by the point cloud is deviated from the rectangular parallelepiped represented by the shape model 40 and there is no surface portion (1) directly below the measurement point d1, as shown by reference numeral B, the surface portion is from the measurement point d1. I can't draw a perpendicular line to (1). In this case, although the measurement point d1 corresponds to the surface portion (1), it may be erroneously determined that the measurement point d1 has no corresponding relationship with the surface portion (1).

The calculation unit 23-k calculates the first correspondence data indicating the correspondence relationship between the point cloud and the surface portion based on the perpendicular line drawn from the point cloud to the surface portion (step ST2). For example, when the calculation unit 23-k can connect the measurement point of the point cloud and the surface portion with one perpendicular line, the calculation unit 23-k determines that the point cloud and the surface portion have a corresponding relationship, and corresponds to these. The first corresponding data showing the relationship is calculated. FIG. 7 is a diagram showing an example of the first corresponding data. The first correspondence data shown in FIG. 7 shows the correspondence between the point cloud having 1 to N measurement points and the surface portions (1) to (6) of the rectangular parallelepiped shown by the shape model 40.

FIG. 8 is a diagram showing a face portion and a point cloud of the shape model 40 having a correct correspondence relationship. Further, FIG. 9 is a diagram showing the result of fitting the point cloud with the correct correspondence to the surface portion of the shape model 40. As shown in FIG. 8, the point group 100 corresponds to the surface portion (1) of the shape model 40, the point group 101 corresponds to the surface portion (3) of the shape model 40, and the point group 102 corresponds to the surface portion (3) of the shape model 40. It corresponds to the surface part (6). When the correct correspondence between the point cloud and the surface portion is required, the point cloud is correctly fitted to the surface portions (1) to (6) of the shape model 40 as shown in FIG.

Next, the calculation unit 23-k calculates the first estimated surface including the point cloud corresponding to the surface unit (i) based on the first corresponding data (step ST3). For example, the calculation unit 23-k randomly selects the coordinates of three measurement points from the point cloud determined to correspond to the surface unit (i) based on the first correspondence data, and the selected three measurement points. The first estimated surface is obtained from. By selecting the three measurement points in this way, all the parameters of the plane that determines the surface portion can be obtained.

Subsequently, the calculation unit 23-k determines whether or not the first estimation surface is correct (step ST4). For example, in the calculation unit 23-k, among the measurement points of the point cloud determined to correspond to the first estimation surface, the number of measurement points whose distance to the first estimation surface is equal to or less than the threshold value is determined from step ST3. When the distance obtained by the previous repetition of the process up to step ST6 is larger than the number of measurement points below the threshold value, it is determined that the first estimation surface is correct. It should be noted that the first estimation surface obtained at the first repetition is determined to be correct. When the measurement error of the measurement point of the point group is σ, the threshold value is set to a value of about 2σ.

When it is determined that the first estimation surface is correct (step ST4; YES), the calculation unit 23-k sets the first estimation surface as the second estimation surface (step ST5). On the other hand, when it is determined that the first estimation surface is incorrect (step ST4; NO), the calculation unit 23-k determines whether or not the end conditions for the processes from step ST3 to step ST6 are satisfied (step). ST6). For example, the end condition is whether or not the processes from step ST3 to step ST6 are repeated a certain number of times.

For the processing from step ST3 to step ST6, an algorithm robust to outliers such as the RANSAC (RANdom Sample Consensus) algorithm described in the following reference may be adopted.
(Reference) M. Armstrong and A. Zisserman, “Robust object tracking” in Proceedings of Asian Convention on Computer Vision, pp. 58-62, 1995.

If the above termination condition is not satisfied (step ST6; NO), the calculation unit 23-k repeatedly executes the process from step ST3. On the other hand, when the above termination condition is satisfied (step ST6; YES), the calculation unit 23-k calculates the boundary of the second estimation surface (step ST7). FIG. 10 is a diagram showing a boundary 201 of the second estimated surface 200, a point cloud 100a within the boundary 201, and a shape model 40. For example, the calculation unit 23-k calculates the boundary 201 by defining the boundary 201 as a closed curve surrounding the point cloud 100a using the alpha shape algorithm.

Subsequently, the calculation unit 23-k expands the boundary 201 of the second estimation surface 200 to the outside. FIG. 11 is a diagram showing an outwardly enlarged boundary 201a, a point cloud 100a within the boundary 201a, and a shape model 40. For example, the calculation unit 23-k could not calculate the first correspondence data as shown by the reference numeral B in FIG. 6, but the measurement point originally having a correspondence relationship with the surface portion (1) is the shape model 40. The boundary 201a is calculated by enlarging the boundary 201 outward so as to be included in the point cloud corresponding to the surface portion (1) of the above.

Next, the calculation unit 23-k calculates a second correspondence data showing the correspondence relationship between the point cloud and the surface portion (i) included in the surface determined by the outwardly expanded boundary 201a (step ST8). FIG. 12 is a diagram showing a correspondence relationship between the point cloud 100 included in the surface determined by the outwardly enlarged boundary 201a and the surface portion (1) of the shape model 40. For example, the calculation unit 23-k selects the point group 100 of the measurement points whose distance from the boundary 201a is equal to or less than the threshold value among the point groups extracted by the filter unit 22-k. In FIG. 12, the point cloud 101 and the point cloud 102 of the measurement points exceeding the distance from the boundary 201a are not selected. Also, the measurement point d2 on a plane different from the point cloud 100 is not selected. The calculation unit 23-k calculates a second correspondence data showing the correspondence relationship between the selected point cloud 100 and the surface portion (1). When the measurement error of the measurement points of the point group is σ, the threshold value is set to a value of about 2σ.

After that, the calculation unit 23-k determines whether or not the processes from step ST2 to step ST8 have been executed for all the surface portions of the shape model 40 (step ST9). If there is a surface portion on which the above processing has not been executed (step ST9; NO), the calculation unit 23-k adds 1 to i (step ST10) and then returns to the processing of step ST2. For example, when the above process is executed for the face portion (1) and the face portions (2) to (6) are not processed, the above process is subsequently executed for the face portion (2).

When the above processing is executed for all the surface portions (step ST9; YES), the calculation unit 23-k performs fitting between the shape model 40 and the point cloud based on the second corresponding data, so that the shape model 40 A point cloud corresponding to this is applied to the surface portion of the rectangular parallelepiped represented by, and the current position and orientation of the antenna 1-k are updated (step ST11). For example, for fitting, a method of minimizing the distance between the measurement point and the surface portion by using the nonlinear least squares method can be adopted.

The calculation unit 23-k determines whether or not the fitting end condition is satisfied (step ST12). If the end condition is not satisfied (step ST12; NO), the process returns to step ST1 and a series of processes from step ST1 are repeated. On the other hand, if the end condition is satisfied (step ST12; YES), the process of FIG. 5 is terminated. As a result, the repetition of fitting is completed. For example, if the error of the fitting executed this time is larger than X% of the residual of the fitting executed last time for M times, the repetition of the fitting is terminated. For example, let X be 90 and M be 5. By repeating this fitting, the accurate position and orientation of the antenna 1-k can be obtained.

FIG. 13 is a diagram showing the result of fitting the measurement points with respect to the surface portion of the shape model 40 in an erroneous correspondence relationship. When the calculation unit 23-k calculates the correspondence data between the point cloud and the surface unit without performing the processes of step ST1, step ST2 and steps ST7 to ST9 shown in FIG. 5, the correct position of the antenna 1-k and I can't ask for a posture. For example, when the processes of step ST1, step ST2, and steps ST7 to ST9 are not executed, as shown in FIG. 13, the correspondence between the point group 100 and the surface portion (1), the point group 101 and the surface portion (3). ), And the correspondence between the point cloud 102 and the surface portion (6) cannot be obtained correctly. Therefore, as shown by the arrow in FIG. 13, the point cloud is fitted in a state where the point cloud is deviated from the shape model.

Next, a modified example of the antennas 1-1, 1-2, ..., 1-N will be described.
FIG. 14 is a block diagram showing a configuration of a modified example of the antennas 1-1, 1-2, ..., 1-N included in the radar device according to the first embodiment. Each of the antennas 1-1, 1-2, ..., 1-N shown in FIG. 14 includes a transmitting unit 10, a receiving unit 11, a driving unit 12, a posture measuring unit 13, and a position measuring unit 14. The transmitting unit 10, the receiving unit 11, and the driving unit 12 function in the same manner as those shown in FIG.

The posture measuring unit 13 measures the rough posture of the antenna. For example, the attitude measuring unit 13 measures the azimuth and elevation angles of the antenna on the order of 0.1 degrees. The position measuring unit 14 measures the rough position of the antenna. For example, the position measuring unit 14 may be a system that measures a rough position of an antenna on the order of several centimeters, such as GPS (Global Positioning System).

The position / orientation setting unit 20 inputs the rough posture of the antenna 1-k from the posture measuring unit 13 included in the antenna 1-k, and determines the rough position of the antenna 1-k from the position measuring unit 14 included in the antenna 1-k. Input and set the rough position and orientation of the antenna in the filter units 22-1,22-2, ..., 22-N.

Next, the hardware configuration that realizes the function of the control device 2 will be described.
The functions of the position / orientation setting unit 20, the shape model input unit 21, the filter units 22-1 to 22-N, the calculation unit 23-1 to 23-N, and the drive amount calculation unit 24 in the control device 2 shown in FIG. 3 are It is realized by the processing circuit. That is, the control device 2 includes a processing circuit for executing the processing from step ST1 to step ST12 shown in FIG. The processing circuit may be dedicated hardware, or may be a CPU (Central Processing Unit) that executes a program stored in the memory.

FIG. 15A is a block diagram showing a hardware configuration that realizes the function of the radar device according to the first embodiment. FIG. 15B is a block diagram showing a hardware configuration for executing software that realizes the functions of the radar device according to the first embodiment. In FIGS. 15A and 15B, the first interface 300 is an interface that relays the exchange of signals between each of the antennas 1-1 to 1-N and the control device 2. The second interface 301 is an interface that relays the exchange of data between the control device 2 and the coordinate measuring device 3.

When the processing circuit is the processing circuit 302 of the dedicated hardware shown in FIG. 15A, the processing circuit 302 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, or an ASIC (Application Specific Integrated Circuit). ), FPGA (Field-Programmable Gate Array), or a combination thereof. The functions of the position / orientation setting unit 20, the shape model input unit 21, the filter unit 22-1 to 22-N, the calculation unit 23-1 to 23-N, and the drive amount calculation unit 24 in the control device 2 are realized by separate processing circuits. Alternatively, these functions may be collectively realized by one processing circuit.

When the processing circuit is the processor 303 shown in FIG. 15B, the position / orientation setting unit 20, the shape model input unit 21, the filter unit 22-1 to 22-N, the calculation unit 23-1 to 23-N, and the drive in the control device 2 The function of the quantity calculation unit 24 is realized by software, firmware, or a combination of software and firmware. The software or firmware is described as a program and stored in the memory 304.

By reading and executing the program stored in the memory 304, the processor 303 reads and executes the position / orientation setting unit 20, the shape model input unit 21, the filter units 22-1 to 22-N, and the calculation unit 23-in the control device 2. The functions of 1 to 23-N and the drive amount calculation unit 24 are realized. For example, the control device 2 includes a memory 304 for storing a program in which the processes from steps ST1 to ST12 in the flowchart shown in FIG. 5 are executed as a result when executed by the processor 303. These programs cause the computer to execute the procedure or method of the arithmetic units 23-1 to 23-N. The memory 304 may be a computer-readable storage medium in which a program for operating the computer as a calculation unit 23-1 to 23-N is stored.

The memory 304 is, for example, a non-volatile semiconductor such as a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), an EEPROM (Electrolytically Magnetic), or an EPROM (Electrically-EPROM). This includes discs, flexible discs, optical discs, compact discs, mini discs, DVDs, and the like.

A part of the functions of the position / orientation setting unit 20, the shape model input unit 21, the filter unit 22-1 to 22-N, the calculation unit 23-1 to 23-N, and the drive amount calculation unit 24 in the control device 2 is dedicated. It may be implemented in hardware and partly in software or firmware. For example, the position / orientation setting unit 20, the shape model input unit 21, and the drive amount calculation unit 24 realize the functions by the processing circuit 302, which is dedicated hardware, and the filter units 22-1 to 22-N and the calculation unit 23- The functions of 1 to 23-N are realized by the processor 303 reading and executing the program stored in the memory 304. In this way, the processing circuit can realize the above-mentioned functions by hardware, software, firmware, or a combination thereof.

As described above, in the radar device according to the first embodiment, the first estimated surface including the point cloud corresponding to the surface portion of the shape model 40 is calculated, the correctness is determined, and the second estimation determined to be correct. The boundary of the surface is calculated and expanded outward, and the second correspondence data showing the correspondence relationship between the point cloud included in the surface determined by the boundary expanded outward and the surface portion of the shape model 40 is calculated, and the second correspondence data is calculated. Fitting between the shape model 40 and the point cloud is performed based on the corresponding data. As a result, the correspondence between the shape of the antenna represented by the shape model 40 and the point cloud measured from the antenna can be accurately obtained.

The shape of the antenna represented by the shape model indicates the ideal position and orientation of the antenna determined by the electromagnetic field simulation so that the radar device satisfies the desired performance. On the other hand, the position and orientation of the antenna in the real environment are set to approximate values due to the installation error of the antenna. In order to correct the antenna in the real environment to the ideal position and orientation, it is necessary to measure the accurate position and orientation of the antenna in the real environment. As described above, since the radar device according to the first embodiment can measure the accurate position and orientation of the antenna, the antenna is adjusted to the ideal position and orientation even if there is an installation error of the antenna. be able to.

Although FIG. 1 shows a radar device including the coordinate measuring device 3, the coordinate measuring device 3 may be a device provided separately from the radar device. That is, if the radar device according to the first embodiment includes antennas 1-1, 1-2, ..., 1-N, and a control device 2 capable of acquiring point cloud data from the coordinate measuring device 3. Good.

Embodiment 2.
FIG. 16 is a block diagram showing a configuration of a control device 2A included in the radar device according to the second embodiment. In FIG. 16, the same components as those in FIG. 3 are designated by the same reference numerals, and the description thereof will be omitted. The control device 2A includes a position / orientation setting unit 20, a shape model input unit 21, a filter unit 22-1 to 22-N, a calculation unit 23-1 to 23-N, a received signal correction unit 25, and a radar signal processing unit 26. ..

The received signal correction unit 25 corrects the radar signal received by each antenna based on the fitting result of the shape model and the point cloud by each of the calculation units 23-1 to 23-N. For example, the reception signal correction unit 25 demodulates a plurality of radar signals received by the reception units 11 of each of the antennas 1-1 to 1-N, and the amplitude of the radar signal according to the deviation of the point group with respect to the shape model. And the phase is corrected, and a synthesis process of a plurality of corrected radar signals is performed. As a result, the received beam pattern of the antenna can be adjusted to the desired pattern.

The radar signal processing unit 26 observes the target based on the radar signal corrected by the received signal correction unit 25. For example, the radar signal processing unit 26 calculates the target position or speed based on the radar signal corrected by the received signal correction unit 25.

As described above, the radar device according to the second embodiment includes a received signal correction unit 25 and a radar signal processing unit 26. By having these configurations, it is possible to adjust the reception beam pattern of the receiving unit 11 to a desired pattern without mechanically driving the antennas 1-1 to 1-N. As a result, deterioration of radar reception performance can be suppressed.

Embodiment 3.
FIG. 17 is a block diagram showing a configuration of a control device 2B included in the radar device according to the third embodiment. In FIG. 17, the same components as those in FIG. 3 are designated by the same reference numerals, and the description thereof will be omitted. The control device 2B includes a position / orientation setting unit 20, a shape model input unit 21, a filter unit 22-1 to 22-N, a calculation unit 23-1 to 23-N, a transmission signal correction unit 27, and a transmission signal generation unit 28. ..

The transmission signal correction unit 27 corrects the radar signal transmitted by each antenna based on the fitting result of the shape model and the point cloud by the calculation units 23-1 to 23-N. For example, the transmission signal correction unit 27 can adjust the transmission beam pattern of the antenna to a target pattern by correcting the amplitude and phase of the transmission signal generated by the transmission signal generation unit 28, and thus the radar transmits. Deterioration of performance can be suppressed.

The transmission signal generation unit 28 generates a radar signal transmitted by each of the antennas 1-1 to 1-N and outputs the radar signal to the transmission signal correction unit 27. For example, the transmission signal generation unit 28 assumes that each of the antennas 1-1 to 1-N has an ideal position and orientation, and generates a radar signal having a desired frequency, waveform, and phase as the transmission signal. ..

As described above, the radar device according to the third embodiment includes a transmission signal correction unit 27 and a transmission signal generation unit 28. By having these configurations, it is possible to adjust the transmission beam pattern of the transmission unit 10 to a desired pattern without mechanically driving the antennas 1-1 to 1-N. As a result, deterioration of radar transmission performance can be suppressed.

Embodiment 4.
FIG. 18 is a block diagram showing a configuration of a control device 2C included in the radar device according to the fourth embodiment. In FIG. 18, the same components as those in FIG. 3 are designated by the same reference numerals, and the description thereof will be omitted. The control device 2C includes a position / orientation setting unit 20, a shape model input unit 21, a filter unit 22-1 to 22-N, a calculation unit 23-1 to 23-N, a drive amount calculation unit 24, a received signal correction unit 25, and a radar. It includes a signal processing unit 26, a transmission signal correction unit 27, and a transmission signal generation unit 28.

The received signal correction unit 25 corrects the radar signal received by each antenna based on the fitting result of the shape model and the point cloud by each of the calculation units 23-1 to 23-N. The radar signal processing unit 26 observes the target based on the radar signal corrected by the received signal correction unit 25. The transmission signal correction unit 27 corrects the radar signal transmitted by each antenna based on the fitting result of the shape model and the point cloud by the calculation units 23-1 to 23-N. The transmission signal generation unit 28 generates a radar signal transmitted by each of the antennas 1-1 to 1-N and outputs the radar signal to the transmission signal correction unit 27.

As described above, the radar device according to the fourth embodiment includes a drive amount calculation unit 24, a reception signal correction unit 25, a radar signal processing unit 26, a transmission signal correction unit 27, and a transmission signal generation unit 28. By having these configurations, the transmission beam pattern of the transmission unit 10 can be adjusted to a desired pattern without mechanically driving the antennas 1-1 to 1-N, and the reception beam pattern of the reception unit 11 can be adjusted. Can be adjusted to the desired pattern. As a result, deterioration of radar reception performance and transmission performance can be suppressed.

It should be noted that the present invention is not limited to the above-described embodiment, and within the scope of the present invention, any combination of the embodiments or any component of the embodiment may be modified or the embodiment. Any component can be omitted in each of the above.

The radar device according to the present invention can be used for target observation because it can accurately determine the correspondence between the shape of the antenna represented by the shape model and the point cloud measured by the antenna.

1-11-N, 1-k interface, 2,2A, 2B, 2C control device, 3 coordinate measurement device, 10 transmitter, 11 receiver, 12 drive unit, 13 attitude measurement unit, 14 position measurement unit, 20 Position / orientation setting unit, 21 shape model input unit, 22-1 to 22-N, 22-k filter unit, 23-1 to 23-N, 23-k calculation unit, 24 drive amount calculation unit, 25 received signal correction Unit, 26 Radar signal processing unit, 27 Transmission signal correction unit, 28 Transmission signal generation unit, 40 shape model, 100, 100a, 101, 102 point group, 200 Second estimation surface, 201, 201a boundary, 300 First Interface, 301 second interface, 302 processing circuit, 303 processor, 304 memory.

Claims (8)

With multiple antennas
A control device for controlling each beam pattern of the plurality of antennas is provided.
The control device is
A plurality of filter units and a plurality of arithmetic units provided corresponding to each of the plurality of antennas,
Each of the plurality of filter units has a shape model input unit for inputting a shape model representing the shape of the antenna corresponding to each of the filter units.
Each of the plurality of filter units
From the point cloud which is the measurement point of a large number of three-dimensional coordinates, the point cloud measured in the region including the shape model input by the shape model input unit is extracted.
Each of the plurality of arithmetic units
A first method showing the correspondence between the point cloud and the surface portion based on the perpendicular line drawn from the point cloud extracted by the filter unit corresponding to the antenna with respect to the surface portion forming the shape of the antenna represented by the shape model. Calculate the corresponding data and
Based on the first correspondence data, an estimated surface including a point cloud corresponding to the surface portion is calculated to determine the correctness.
Calculate the boundary of the estimated surface determined to be correct and expand it outward.
A second correspondence data showing the correspondence relationship between the point cloud and the face portion included in the face determined by the boundary enlarged to the outside is calculated, and the second correspondence data is calculated.
A radar device characterized by fitting a shape model and a point cloud based on the second corresponding data. Each of the plurality of antennas has a drive unit that drives the antenna to change its position and orientation.
A drive amount calculation unit is provided that calculates the drive amount of the antenna corresponding to the calculation unit based on the fitting result of the shape model and the point cloud by the calculation unit and sets the calculated drive amount in the drive unit. The radar device according to claim 1, wherein the radar device is characterized by the above. A position / orientation setting unit for setting a rough position and orientation of the antenna corresponding to the filter unit is provided for each of the plurality of filter units.
Each of the plurality of filter units
Based on the rough position and orientation of the antenna set by the position / orientation setting unit, in a region including a shape model input from the shape model input unit from a point cloud which is a measurement point of a large number of three-dimensional coordinates. The radar device according to claim 1, wherein the measured point cloud is extracted. Each of the plurality of antennas
A position measuring unit that measures the rough position of the antenna,
It has a posture measuring unit that measures the rough posture of the antenna.
The position / orientation setting unit inputs the rough position of the antenna measured by the position measuring unit, inputs the rough posture of the antenna measured by the posture measuring unit, and inputs the rough posture of the antenna. The radar device according to claim 3, wherein a different position and posture are set in the filter unit. A reception signal correction unit that corrects the reception signal received by the antenna based on the fitting result of the shape model and the point cloud by the calculation unit.
A radar signal processing unit that observes a target based on the received signal corrected by the received signal correction unit, and
The radar device according to claim 1, wherein the radar device is provided. A transmission signal generator that generates a transmission signal transmitted by the antenna,
A transmission signal correction unit that corrects the transmission signal generated by the transmission signal generation unit based on the fitting result of the shape model and the point cloud by the calculation unit.
The radar device according to claim 1, wherein the radar device is provided. A reception signal correction unit that corrects the reception signal received by the antenna based on the fitting result of the shape model and the point cloud by the calculation unit.
A radar signal processing unit that observes a target based on the received signal corrected by the received signal correction unit, and
A transmission signal generator that generates a transmission signal transmitted by the antenna,
A transmission signal correction unit that corrects the transmission signal generated by the transmission signal generation unit based on the fitting result of the shape model and the point cloud by the calculation unit.
The radar device according to claim 1, wherein the radar device is provided. With multiple antennas
A control device for controlling each beam pattern of the plurality of antennas is provided.
The control device
A plurality of filter units and a plurality of arithmetic units provided corresponding to each of the plurality of antennas,
Each of the plurality of filter units has a shape model input unit for inputting a shape model representing the shape of the antenna corresponding to each of the filter units.
Each of the plurality of filter units
It is a position / orientation measurement method of a radar device that extracts a point cloud measured in a region including a shape model input by the shape model input unit from a point cloud which is a measurement point of a large number of three-dimensional coordinates.
Each of the plurality of arithmetic units
A first method showing the correspondence between the point cloud and the surface portion based on the perpendicular line drawn from the point cloud extracted by the filter unit corresponding to the antenna with respect to the surface portion forming the shape of the antenna represented by the shape model. Steps to calculate the corresponding data and
Based on the first correspondence data, a step of calculating an estimated surface including a point cloud corresponding to the surface portion and determining correctness
The step of calculating the boundary of the estimated surface determined to be correct and expanding it to the outside,
A step of calculating a second correspondence data showing the correspondence relationship between the point cloud and the face portion included in the face determined by the boundary enlarged to the outside, and
The step of fitting the shape model and the point cloud based on the second corresponding data, and
A position / posture measurement method characterized by being equipped with.

Images