Radar correction system and radar correction method

Abstract

This invention discloses a radar correction system and a radar correction method. The method includes: acquiring a first point cloud using radar and generating a first fitting plane based on the first point cloud; acquiring a second point cloud using radar and generating a second fitting plane based on the second point cloud; and correcting the radar coordinate system based on the angle between the first fitting plane and the second fitting plane.

Description

Technical Field

[0001] This invention relates to a radar measurement technology, and more particularly to a radar correction system and a radar correction method. Background Technology

[0002] With the increasing number of elderly people living alone, the demand for home care is also rising. Many home care systems use cameras or wearable devices to monitor the status of those being cared for. However, using video surveillance may have drawbacks such as invading privacy or causing leakage of personal information.

[0003] Furthermore, most people dislike being restricted by wearable devices or often forget to wear them. To address these issues, some home care systems use radar to monitor those being cared for. However, radar can tilt due to improper installation or external forces, leading to deviations in the coordinate system of the point cloud data acquired by the radar. Therefore, correctly calibrating the radar's coordinate system is one of the important issues in this field. Summary of the Invention

[0004] This invention provides a radar correction system and a radar correction method that can correct the coordinate system of a tilted radar.

[0005] An embodiment of the present invention provides a radar correction system comprising a radar and a controller. The controller is communicatively connected to the radar, wherein the controller is configured to perform: acquiring a first point cloud via the radar and generating a first fitting plane based on the first point cloud; acquiring a second point cloud via the radar and generating a second fitting plane based on the second point cloud; and correcting the radar's coordinate system based on the angle between the first fitting plane and the second fitting plane.

[0006] An embodiment of the present invention provides a radar correction method, comprising: acquiring a first point cloud via radar and generating a first fitting plane based on the first point cloud; acquiring a second point cloud via radar and generating a second fitting plane based on the second point cloud; and correcting the radar coordinate system based on the angle between the first fitting plane and the second fitting plane.

[0007] Based on the above, the radar correction system of the present invention can generate a fitting plane based on the point cloud collected by the radar, and determine whether the radar is tilted based on the fitting plane. If the radar is tilted, the radar correction system can correct the radar's coordinate system without using additional sensors by using a rotation matrix or control actuators. Attached Figure Description

[0008] Figure 1 A schematic diagram of a radar correction system is shown for one embodiment of the present invention;

[0009] Figure 2A top view of the field is shown for one embodiment of the present invention;

[0010] Figure 3 A side view of the field is shown for one embodiment of the present invention;

[0011] Figure 4 A front view of the field is shown for one embodiment of the present invention;

[0012] Figure 5 A flowchart illustrating radar correction is shown for one embodiment of the present invention;

[0013] Figure 6 A top view of a radar and multiple reflectors is shown for one embodiment of the present invention;

[0014] Figure 7 A flowchart illustrating a radar correction method is shown for one embodiment of the present invention.

[0015] Symbol Explanation

[0016] 10: Radar Correction System

[0017] 100: Controller

[0018] 110,210: Processor

[0019] 120, 220: Storage media

[0020] 130, 230: Transceiver

[0021] 20: Field

[0022] 21: Wall

[0023] 200: Radar

[0024] 240: Actuator

[0025] 250: Antenna

[0026] 260: Inertial Measurement Unit

[0027] 300:Object

[0028] A, B, C, D: Reflectors

[0029] S501, S502, S503, S504, S505, S506, S507, S508, S701, S702, S703: Steps Detailed Implementation

[0030] To make the content of this invention more readily apparent, the following specific embodiments are provided as examples on which this invention can indeed be practiced. Furthermore, wherever possible, elements / components / steps referred to by the same reference numerals in the drawings and embodiments represent the same or similar parts.

[0031] Figure 1 A schematic diagram of a radar correction system 10 according to an embodiment of the present invention is shown. The radar correction system 10 may include a controller 100 and a radar 200, wherein the controller 100 is communicatively connected to the radar 200 and components in the radar 200. In one embodiment, the controller 100 may be embedded in the radar 200.

[0032] Controller 100 may include processor 110, storage medium 120, and transceiver 130. Processor 110 may be, for example, a central processing unit (CPU), or other programmable general-purpose or special-purpose microcontroller (MCU), microprocessor, digital signal processor (DSP), programmable controller, application-specific integrated circuit (ASIC), graphics processing unit (GPU), image signal processor (ISP), image processing unit (IPU), arithmetic logic unit (ALU), complex programmable logic device (CPLD), field-programmable gate array (FPGA), or other similar elements or combinations thereof. Processor 110 may be coupled to storage medium 120 and transceiver 130, and access and execute multiple modules and various applications stored in storage medium 120.

[0033] Storage medium 120 may be any type of fixed or removable random access memory (RAM), read-only memory (ROM), flash memory, hard disk drive (HDD), solid state drive (SSD), or similar elements or combinations thereof, for storing multiple modules or various applications that can be executed by processor 110.

[0034] Transceiver 130 transmits or receives signals wirelessly or via a wired connection. Transceiver 130 may also perform operations such as low-noise amplification, impedance matching, mixing, up or down frequency conversion, filtering, amplification, and similar functions.

[0035] Radar 200 is, for example, a frequency-modulated continuous wave (FMCW) radar. Radar 200 can detect objects in a field within its coverage area to generate point clouds or bounding boxes corresponding to the objects (e.g., walls, ceilings, or obstacles in the field), and obtain the centroid or center of gravity information of the point cloud. Radar 200 may include a processor 210, storage medium 220, transceiver 230, actuator 240, and antenna 250. In one embodiment, radar 200 may further include an inertial measurement unit (IMU) 260.

[0036] Processor 210 may be, for example, a CPU, or other programmable general-purpose or special-purpose MCU, microprocessor, DSP, programmable controller, ASIC, GPU, ISP, IPU, ALU, CPLD, FPGA, or other similar components or combinations thereof. Processor 210 may be coupled to storage medium 220, transceiver 230, actuator 240, and inertial measurement unit 260, and access and execute multiple modules and various applications stored in storage medium 220. When controller 100 is embedded in radar 200, processor 210 and controller 100 (or processor 110) may be the same hardware device. Processor 210 (or controller 100) may analyze the point cloud acquired by radar 200 to determine the movement path, instantaneous speed, average speed, dwell time in a specific area, or human activity information, etc., based on time and location information.

[0037] Storage medium 220 is, for example, any type of fixed or removable RAM, ROM, flash memory, HDD, SSD or similar element or combination thereof, used to store multiple modules or various applications that can be executed by processor 210.

[0038] Transceiver 230 transmits or receives signals wirelessly or via a wired connection. Transceiver 230 can also perform operations such as low-noise amplification, impedance matching, mixing, up- or down-frequency conversion, filtering, amplification, and similar functions. Transceiver 230 can be coupled to antenna 250 and transmit or receive wireless signals through antenna 250, thereby generating point clouds.

[0039] The actuator 240 may include mechanical structures such as a motor. The actuator 240 may be configured to rotate the antenna 250, thereby rotating the coordinate system of the radar 200 with the antenna 250 as a reference point.

[0040] The inertial measurement unit 260 is, for example, an accelerometer, a three-axis sensor, a six-axis sensor, or a nine-axis sensor. The inertial measurement unit 260 can detect the acceleration of the radar 200 in a specific direction. The controller 100 or the processor 210 can determine whether the radar 200 is tilted based on the measurement results of the inertial measurement unit 260.

[0041] Figure 2 A top view of field 20 is shown according to an embodiment of the present invention. When radar 200 is correctly mounted on the wall (i.e., the XZ plane), the three axes of the radar 200's original coordinate system (e.g., Cartesian coordinate system) are the X-axis, Y-axis, and Z-axis, respectively. Object 300 is located to the right front of radar 200, and the coordinates of object 300 in the original coordinate system are (X... t ,Y t When radar 200 rotates clockwise by an angle θ due to improper installation or external forces, the three axes of radar 200's coordinate system become the X', Y', and Z' axes. The coordinates of object 300 will then transform into (X'... t ,Y' t This caused object 300 to be mistakenly identified as being directly in front of radar 200.

[0042] Figure 3 A side view of the field 20 is shown according to an embodiment of the present invention. When the radar 200 is correctly mounted on the wall (i.e., the XZ plane), the three axes of the original coordinate system of the radar 200 are the X-axis, Y-axis, and Z-axis. When the radar 200 tilts downward due to improper installation or external forces, the three axes of the radar 200's coordinate system become the X'-axis, Y'-axis, and Z'-axis, where α is the angle between the Y'-axis and the Y-axis. The coordinates of the object 300 in the coordinate system will change, causing the object 300 to be misjudged as being in a higher position.

[0043] Figure 4A front view of the field 20 is shown according to an embodiment of the present invention. When the radar 200 is correctly mounted on the wall (i.e., the XZ plane), the three axes of the original coordinate system of the radar 200 are the X-axis, Y-axis, and Z-axis. The object 300 is located to the upper right of the radar 200, and the coordinates of the object 300 in the original coordinate system are (X... t Z t When radar 200 rotates clockwise by an angle β due to improper installation or external forces, the three axes of radar 200's coordinate system become the X', Y', and Z' axes. The coordinates of object 300 will then transform into (X'... t ,Z' t This caused object 300 to be mistakenly identified as being directly above radar 200.

[0044] When radar 200 tilts, the point cloud data of fixed objects around radar 200 will change. For example, the point cloud corresponding to a vertical wall will no longer appear as a vertical plane, but as a tilted plane. Assume radar 200 is horizontally mounted and facing a vertical wall. When radar 200 is not tilted, the plane equation of the vertical wall is x = 0. When radar 200 tilts, causing its coordinate system to rotate along the Y-axis, the plane equation of the vertical wall changes to x = az + b, where a is the slope of the vertical wall. Controller 100 can calculate the tilt angle of radar 200 as θ = arctan(a) based on the change in the plane equation.

[0045] Assume that radar 200 is tilted around the Y-axis of the original coordinate system. After determining θ, controller 100 can transform the coordinate system of radar 200 back to the original coordinate system when radar 200 is not tilted according to the rotation matrix Ry(θ) as shown in formula (1).

[0046]

[0047] The controller 100 can use the rotation matrix Ry(θ) to convert the coordinates (x,y,z) of the point cloud detected by the tilted radar 200 into untilted coordinates (x',y',z'), as shown in formula (2).

[0048]

[0049] On the other hand, the controller 100 can rotate the antenna 250 of the radar 200 by the brake 240, thereby adjusting the beam direction of the radar 200.

[0050] Figure 5 A flowchart of radar correction is illustrated according to an embodiment of the present invention, wherein the steps of the flowchart can be derived from, for example... Figure 1 The radar correction system 10 shown is implemented.

[0051] In step S501, the controller 100 can obtain the first point cloud corresponding to the first time period (e.g., the time period when the radar 200 has not yet tilted) through the radar 200.

[0052] In step S502, controller 100 may generate a first fitting plane based on the first point cloud. In one embodiment, controller 110 may filter dynamic point clouds (e.g., point clouds of people or pets) and retain static point clouds (e.g., point clouds of walls or floors) in the first point cloud to update the first point cloud. After updating the first point cloud, controller 100 generates the first fitting plane based on the updated first point cloud. For example, controller 100 may generate a first fitting plane x = 0 corresponding to the wall facing radar 200 based on the first point cloud.

[0053] In one embodiment, processor 110 may perform a random sample consensus (RANSAC) algorithm on the point cloud to generate a fitting plane.

[0054] In step S503, the controller 100 can obtain a second point cloud corresponding to a second time period (e.g., a time period in which the radar 200 has tilted) through the radar 200, wherein the second time period is later than the first time period.

[0055] In one embodiment, the controller 100 can periodically control the radar 200 to scan and obtain a second point cloud, and then determine whether the radar 200 is tilted based on the second point cloud. In another embodiment, the controller 100 can determine whether the radar 200 is tilted based on the measurement results of the inertial measurement unit 260. If the controller 100 determines that the radar 200 is tilted, the controller 100 can obtain the second point cloud through the radar 200 based on the triggering of the measurement results.

[0056] In step S504, the controller 100 may generate a second fitting plane based on the second point cloud. In one embodiment, the controller 110 may filter out dynamic point clouds in the second point cloud and retain static point clouds to update the second point cloud. After updating the second point cloud, the controller 100 generates a second fitting plane based on the updated second point cloud.

[0057] In step S505, the controller 100 can determine the angle θ between the first fitting plane and the second fitting plane.

[0058] In step S506, the controller 100 can determine whether the included angle θ is greater than a threshold. If the included angle θ is greater than the threshold, the controller 100 can execute step S508 to adjust the coordinate system of the radar 200 to a greater extent. If the included angle θ is less than or equal to the threshold, the controller 100 can execute step S507 to adjust the coordinate system of the radar 200 to a lesser extent.

[0059] In step S507, the controller 100 generates a rotation matrix based on the included angle θ and corrects the coordinate system of the radar 200 based on the rotation matrix. After correction, the point cloud output by the radar 200 has correct coordinates.

[0060] In step S508, the controller 100 can control the actuator 240 to rotate the antenna 250 according to the included angle θ, thereby reducing the actual tilt angle of the radar 200. In one embodiment, after completing step S508, the controller 100 can re-execute steps S503 to S507 to fine-tune the coordinate system of the radar 200. After the calibration is completed, the point cloud output by the radar 200 can have the correct coordinates.

[0061] For example, after tilting, the points captured by radar 200 from the wall facing radar 200 (e.g., the wall corresponding to the first fitting plane x = 0) are shown in Table 1. The points in Table 1 indicate that the wall is no longer a vertical plane (i.e., the plane x = 0), but a tilted plane. Controller 110 can generate a tilted plane equation x = 0.1z + 0.1 based on these points as a second fitting plane. Controller 110 can calculate the angle θ = arctan(0.1) ≈ 5.71° between the first fitting plane x = 0 and the second fitting plane equation x = 0.1z + 0.1. Controller 110 can perform coordinate transformation on the point cloud using the rotation matrix Rx (5.71°). For example, for the point cloud coordinates (0.3,3,2) obtained after the radar 200 is tilted, the controller 100 can use the rotation matrix Rx (5.71°) to convert the coordinates (0.3,3,2) into the coordinates (0.1,3,2.02) corresponding to the untilted radar 200.

[0062] Table 1

[0063] Point index X coordinate Y coordinate Z-coordinate 1 0.1 2 0 2 0.2 2.5 1 3 0.3 3 2 4 0.4 3.5 3

[0064] In one embodiment, the radar correction system 100 may further include one or more reflectors disposed within the coverage area of ​​the radar 200, wherein the reflectors may include, but are not limited to, corner reflectors. The reflectors are positioned at fixed locations in the field. When the radar 200 scans a reflector, the reflection point corresponding to the reflector has higher energy. This point with higher energy can assist the controller 100 or the radar 200 in positioning (e.g., locating the reflector or its placement).

[0065] In one embodiment, at least four reflectors are provided within the coverage area of ​​the radar 200. Figure 6A top view of radar 200 and a plurality of reflectors (reflectors A, B, C, and D) is illustrated according to an embodiment of the present invention. Controller 110 can locate surrounding objects or their fitted planes based on multiple positions of the plurality of reflectors. For example, controller 110 can locate the viewing direction (e.g., wall 21) facing radar 200 based on multiple positions of reflectors A, B, C, and D.

[0066] Controller 110 can calculate multiple distances between reflectors based on multiple positions of reflectors A, B, C, and D, including... and After reflector D is moved, the distance corresponding to reflector D is... and This can change. Controller 110 can correspond to the unchanged distance (e.g.: or Other reflectors (e.g., reflectors A, B, or C) can be used as reference points to position surrounding objects or their fitted planes, or to correct the position of reflector D.

[0067] For example, suppose reflectors A, B, C, and D are located in the original coordinate system at positions A = (1,3,0), B = (1,1,0), C = (-1,3,0), and D = (-1,1,0), respectively. Controller 110 can calculate multiple distances between reflectors based on the multiple positions of reflectors A, B, C, and D. and Suppose that reflector D is moved in the opposite direction (1) to radar 200, causing the coordinates of reflector D to change to D = (-1, 2, 0). Multiple distances between reflectors can then be changed as follows: and Controller 100 can determine distance and The reference point has changed, so reflector D can no longer be used as a reference point.

[0068] On the other hand, the controller 100 can determine the distance. and Since nothing has changed, reflectors A, B, and C can still be used as reference points. Controller 100 can use the stationary reflectors A, B, and C as reference points to locate surrounding objects or their fitted planes, for example, to locate the viewing angle direction faced by radar 200.

[0069] In one embodiment, the controller 100 can use the stationary reflectors A, B, and C as reference points to calculate the tilt angle to correct the coordinate system, thereby calculating the new coordinates of reflector D. For example, the controller 100 can perform triangulation based on the coordinates of each reflector A, B, and C and the distance between each reflector A, B, and C and reflector D to calculate the coordinates of reflector D.

[0070] Figure 7 A flowchart of a radar correction method is illustrated according to an embodiment of the present invention, wherein the radar correction method may be performed by, for example... Figure 1 The radar correction system 10 shown is implemented. In step S701, a first point cloud is acquired by the radar, and a first fitting plane is generated based on the first point cloud. In step S702, a second point cloud is acquired by the radar, and a second fitting plane is generated based on the second point cloud. In step S703, the radar coordinate system is corrected based on the angle between the first fitting plane and the second fitting plane.

[0071] In summary, the radar correction system of the present invention can determine whether the radar is installed at an angle on a wall, corner, or ceiling based on the point cloud collected by the radar. If the radar is tilted, the radar correction system can correct the radar's coordinate system based on the tilt angle. When the tilt angle is small, the radar correction system can use a rotation matrix to correct the coordinate system. When the tilt angle is too large, the radar correction system can control the radar's actuator to rotate the radar's antenna, thereby significantly reducing the radar's actual tilt angle. Furthermore, when a reflector placed within the radar's coverage area to assist in radar positioning is moved, the radar correction system can correct the position information of the moved reflector based on the distances between multiple reflectors and the radar.

Claims

1. A radar correction system, comprising: radar; as well as A controller, communicatively connected to the radar, wherein the controller is configured to perform: The radar acquires a first point cloud, and a first fitting plane is generated based on the first point cloud; A second point cloud is obtained through the radar, and a second fitting plane is generated based on the second point cloud; and The radar coordinate system is corrected based on the angle between the first fitting plane and the second fitting plane.

2. The radar correction system of claim 1, wherein the controller is configured to perform: Determine whether the included angle is greater than a threshold; In response to the included angle being less than or equal to one of the thresholds, a rotation matrix is ​​generated based on the included angle; and The coordinate system is corrected according to the rotation matrix.

3. The radar correction system of claim 1, wherein the radar includes an actuator and an antenna, wherein the controller is communicatively connected to the actuator and configured to further perform: Determine whether the included angle is greater than a threshold; and In response to the included angle being greater than the threshold, the actuator is controlled to rotate the antenna according to the included angle.

4. The radar correction system as claimed in claim 1, further comprising: A reflector is positioned within the coverage area of ​​the radar, wherein The controller locates the reflector based on the energy of the reflection point corresponding to the reflector in the first point cloud.

5. The radar correction system as claimed in claim 1, further comprising: At least four reflectors are respectively positioned within the coverage area of ​​the radar, wherein The controller positions the first fitting plane based on multiple positions of the at least four reflectors.

6. The radar correction system of claim 5, wherein the controller is configured to further perform: Calculate multiple distances between the at least four reflectors based on the multiple locations; Determine whether the first distance corresponding to the first reflector among the plurality of distances has changed, and determine whether the second distance corresponding to the second reflector among the plurality of distances has changed; and In response to a change in the first distance but no change in the second distance, the first position of the first reflector is corrected using the second reflector as a reference point.

7. The radar correction system of claim 5, wherein the controller is configured to further perform: Calculate multiple distances between the at least four reflectors based on the multiple locations, wherein the at least four reflectors include a first reflector, a second reflector, a third reflector, and a fourth reflector; Determine whether the set of distances corresponding to the first reflector among the plurality of distances has changed, and determine whether the other distances among the plurality of distances besides the set of distances have changed; as well as In response to a change in the set of distances but no change in the other distances, the first fitting plane is located using the second reflector, the third reflector, and the fourth reflector as reference points.

8. The radar correction system of claim 1, wherein the first point cloud corresponds to a first time period, and the second point cloud corresponds to a second time period later than the first time period.

9. The radar correction system of claim 1, wherein the radar comprises: An inertial measurement unit (IMU) is communicatively connected to the controller and generates measurement results, wherein... The measurement result triggers the controller to acquire the second point cloud.

10. The radar correction system of claim 1, wherein the controller is configured to further perform: Filter the dynamic point cloud in the first point cloud to update the first point cloud; Filtering dynamic point clouds in the second point cloud to update the second point cloud; and The angle is determined by the updated first point cloud and the updated second point cloud.

11. A radar correction method, comprising: The first point cloud is obtained by radar, and a first fitting plane is generated based on the first point cloud; A second point cloud is obtained through the radar, and a second fitting plane is generated based on the second point cloud; and The radar coordinate system is corrected based on the angle between the first fitting plane and the second fitting plane.

12. The radar correction method of claim 11, wherein the step of correcting the radar's coordinate system based on the angle between the first fitting plane and the second fitting plane includes: Determine whether the included angle is greater than a threshold; In response to the included angle being less than or equal to one of the thresholds, a rotation matrix is ​​generated based on the included angle; as well as The coordinate system is corrected according to the rotation matrix.

13. The radar correction method of claim 11, wherein the radar includes an actuator and an antenna, wherein the step of correcting the coordinate system of the radar according to the angle between the first fitting plane and the second fitting plane includes: Determine whether the included angle is greater than a threshold; as well as In response to the included angle being greater than the threshold, the actuator is controlled to rotate the antenna according to the included angle.

14. The radar correction method as described in claim 11, further comprising: The reflector is located based on the energy of the reflection point corresponding to the reflector in the first point cloud, wherein the reflector is positioned within the coverage area of ​​the radar.

15. The radar correction method of claim 11, wherein the step of generating the first fitting plane based on the first point cloud includes: The first fitting plane is located based on multiple positions of at least four reflectors, wherein the at least four reflectors are respectively positioned within the coverage area of ​​the radar.

16. The radar correction method as described in claim 15, further comprising: Calculate multiple distances between the at least four reflectors based on the multiple locations; Determine whether the first distance corresponding to the first reflector among the plurality of distances has changed, and determine whether the second distance corresponding to the second reflector among the plurality of distances has changed; as well as In response to a change in the first distance but no change in the second distance, the first position of the first reflector is corrected using the second reflector as a reference point.

17. The radar correction method as described in claim 15, further comprising: Calculate multiple distances between the at least four reflectors based on the multiple locations, wherein the at least four reflectors include a first reflector, a second reflector, a third reflector, and a fourth reflector; Determine whether the set of distances corresponding to the first reflector among the plurality of distances has changed, and determine whether the other distances among the plurality of distances besides the set of distances have changed; as well as In response to a change in the set of distances but no change in the other distances, the first fitting plane is located using the second reflector, the third reflector, and the fourth reflector as reference points.

18. The radar correction method of claim 11, wherein the first point cloud corresponds to a first time period, and the second point cloud corresponds to a second time period later than the first time period.

19. The radar correction method of claim 11, wherein the step of obtaining the second point cloud via the radar includes: Measurement results are generated through an inertial measurement unit; as well as The second point cloud is obtained in response to the triggering of the measurement result.

20. The radar correction method as described in claim 11, further comprising: Filter the dynamic point cloud in the first point cloud to update the first point cloud; Filter the dynamic point cloud in the second point cloud to update the second point cloud; as well as The angle is determined by the updated first point cloud and the updated second point cloud.

Images