Radar detection system and radar field of view direction adjustment method

Abstract

A radar detection system and a radar field of view direction adjustment method are disclosed. A processing unit of the radar detection system performs, in a scan round: obtaining object clusters of a current frame and a current state and a next state of each object cluster based on digital feedback signals; and determining a displacement amount and judging whether a tracking cluster in the object clusters is located at an edge of a current field of view of a radar unit according to the current state of the tracking cluster and the displacement amount, and in response to the tracking cluster being located at the edge of the current field of view, issuing a control signal to adjust the field of view direction based on the next state of the tracking cluster.

Description

Technical Field

[0001] This invention relates to the field of radar detection, and in particular to techniques for using frequency-modulated continuous wave signals in radar detection. Background Technology

[0002] In recent years, radar detection technology has developed rapidly, and related technologies have been gradually applied to patient care, long-term care for the elderly, and infant care. For infants, bedridden patients, or the elderly, radar only needs to detect the area on the bed, and radar with a fixed pointing area can be used. However, for caregivers who are mobile and can move, the radar loses its detection signal when they leave the radar's field of view, making it impossible to detect the caregiver's physiological information. Summary of the Invention

[0003] In view of this, some embodiments of the present invention provide a radar detection system, a radar field of view adjustment method, a computer-readable recording medium with stored programs, and a non-transitory computer program product to improve the problems of the prior art.

[0004] Some embodiments of the present invention provide a radar detection system. The radar detection system includes a radar unit, a pointing control unit, and a processing unit. The radar unit is configured to generate a radio frequency signal. The radar unit radiates the radio frequency signal and receives a feedback signal, and demodulates and digitizes the feedback signal to obtain a digital feedback signal. The pointing control unit is configured to adjust the field of view of the radar unit according to a control signal. The processing unit is configured to perform the following steps within a scan round: obtaining the object clusters of the current frame and the current state and next state of each object cluster based on the digital feedback signal; determining the displacement amount, and judging whether the tracking cluster is located at the edge of the current field of view of the radar unit based on the current state and displacement amount of the tracking cluster; and issuing a control signal to adjust the field of view direction based on the next state of the tracking cluster in response to the tracking cluster being located at the edge of the current field of view.

[0005] Some embodiments of the present invention provide a radar field-of-view adjustment method, applicable to a radar detection system, executed by the processing unit of the radar detection system. The radar detection system includes a radar unit, a pointing control unit, and a processing unit. The radar unit is configured to generate a radio frequency signal. The radar unit radiates the radio frequency signal and receives a feedback signal, and demodulates and digitizes the feedback signal to obtain a digital feedback signal. The pointing control unit is configured to adjust the field-of-view direction of the radar unit according to a control signal. The radar field-of-view adjustment method includes execution within a scan round: obtaining the object clusters of the current frame and the current state and next state of each object cluster based on the digital feedback signal; determining the displacement amount, and judging whether the tracking cluster is located at the edge of the current field of view of the radar unit based on the current state and displacement amount of the tracking clusters in the object clusters; in response to the tracking cluster being located at the edge of the current field of view, issuing a control signal based on the next state of the tracking cluster to adjust the field-of-view direction.

[0006] Some embodiments of the present invention provide a computer-readable medium with built-in storage and a non-transitory computer program product, which, when loaded and executed by a processor, can perform the aforementioned radar field of view adjustment method.

[0007] Based on the above, some embodiments of the present invention integrate radar detection, radar tracking, and adjustment and control of the radar's field of view, breaking through the limitations of a fixed position. Attached Figure Description

[0008] Figure 1 A block diagram of a radar detection system illustrated in an embodiment of the present invention;

[0009] Figure 2 This is a schematic diagram of a radar detection system illustrated in an embodiment of the present invention;

[0010] Figure 3 This is a schematic diagram of a radar detection system illustrated in an embodiment of the present invention;

[0011] Figure 4-1 This is a schematic diagram illustrating the adjustment of the field of view according to an embodiment of the present invention;

[0012] Figure 4-2 This is a schematic diagram illustrating the adjustment of the field of view according to an embodiment of the present invention;

[0013] Figure 5-1 These are schematic diagrams illustrating the operation of a radar detection system according to some embodiments of the present invention;

[0014] Figure 5-2 These are schematic diagrams illustrating the operation of a radar detection system according to some embodiments of the present invention;

[0015] Figure 5-3 These are schematic diagrams illustrating the operation of a radar detection system according to some embodiments of the present invention;

[0016] Figure 6 A schematic diagram illustrating the current field of view and image coordinates for an embodiment of the present invention;

[0017] Figure 7 This is a schematic diagram illustrating the current field of view, the total field of view, and the image coordinates according to an embodiment of the present invention.

[0018] Figure 8 This is a schematic diagram illustrating the current field of view, the total field of view, and the image coordinates according to an embodiment of the present invention.

[0019] Figure 9 This is a schematic diagram illustrating the current field of view, the total field of view, and the image coordinates according to an embodiment of the present invention.

[0020] Figure 10This is a schematic diagram illustrating the current field of view, the total field of view, and the image coordinates according to an embodiment of the present invention.

[0021] Figure 11 A schematic diagram illustrating the structure of a processing unit according to some embodiments of the present invention;

[0022] Figure 12-1 This is a flowchart illustrating a radar field of view adjustment method according to some embodiments of the present invention;

[0023] Figure 12-2 This is a flowchart illustrating a radar field of view adjustment method according to some embodiments of the present invention;

[0024] Figure 13 A flowchart illustrating the point cloud map generation process of an embodiment of the present invention;

[0025] Figure 14 This is a flowchart illustrating a radar field of view adjustment method according to some embodiments of the present invention;

[0026] Figure 15 This is a flowchart illustrating a radar field of view adjustment method according to some embodiments of the present invention;

[0027] Figure 16 This is a flowchart illustrating a radar field of view adjustment method according to some embodiments of the present invention;

[0028] Figure 17 This is a flowchart illustrating a radar field of view adjustment method according to some embodiments of the present invention;

[0029] Figure 18 This is a flowchart illustrating a radar field of view adjustment method according to some embodiments of the present invention.

[0030] Symbol Explanation

[0031] 100: Radar Detection System

[0032] 101: Antenna Unit

[0033] 102: Front-end unit

[0034] 103: Processing Unit

[0035] 104: Points to the control unit

[0036] 105: Radar Unit

[0037] 201: Transmitting antenna unit

[0038] 202: Receiving antenna unit

[0039] 203: Launching Unit

[0040] 204: Signal Generator

[0041] 205: Receiving Unit

[0042] 206: Demodulation Unit

[0043] 207: Analog-to-digital converter

[0044] 208-1~208-K: Transmitting antennas

[0045] 209-1~209-N, 210-1~210-M: Receiving antennas

[0046] K, N, M: Positive integers

[0047] 401-1~401-5: Antenna Group

[0048] 500: Point cloud map

[0049] 501: point

[0050] 502, 503: Clusters

[0051] 504: Detection Area Diagram

[0052] 601: The Plane of the Present Vision

[0053] 602, 603: Objects

[0054] 602', 603': Object projection

[0055] 701: Global View Coordinate System

[0056] 702, 804: View Direction

[0057] 703: The center point of the plane of the current field of view

[0058] 801: Center point of the global horizon coordinate system

[0059] 802: The center point of the plane of the current field of view

[0060] 803-1~803-7: Trajectory Points

[0061] 1100: Processing Unit

[0062] 1101: Processor

[0063] 1102: Internal Memory

[0064] 1103: Non-volatile memory

[0065] S1101~S1105, S1201~S1204, S1301~S1302, S1401, S1501~S1502, S1601~S1602, S1701~S1702: Steps Detailed Implementation

[0066] The foregoing and other technical contents, features, and effects of the present invention will be clearly presented in the following detailed description of embodiments with reference to the accompanying drawings. The thickness or dimensions of the elements in the drawings are exaggerated, omitted, or generalized for the understanding and reading of those skilled in the art. The dimensions of each element are not exactly their actual dimensions and are not intended to limit the implementation of the invention. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in proportions, or adjustments in size, without affecting the effects and objectives achieved by the invention, should still fall within the scope of the technical content disclosed in the present invention. The same reference numerals will be used to denote the same or similar elements in all the drawings. The terms "coupled" or "connected" as used in the following embodiments can refer to any direct or indirect, wired or wireless connection means.

[0067] Figure 1 This is a block diagram of a radar detection system drawn according to an embodiment of the present invention. Please refer to... Figure 1 The radar detection system 100 includes a radar unit 105, a processing unit 103, and a pointing control unit 104. The radar unit 105 includes an antenna unit 101 and a front-end unit 102. The antenna unit 101 is configured to radiate radio frequency signals into free space, where the radio frequency signals reflect feedback signals upon impact with objects. The antenna unit 101 receives the feedback signals from the radio frequency signals. The front-end unit 102 is configured to generate the aforementioned radio frequency signals and demodulate and digitize the aforementioned feedback signals to obtain digital feedback signals. The pointing control unit 104 is configured to adjust the field of view (FOV) direction of the radar unit 105 according to control signals. The processing unit 103 is configured to receive the aforementioned digital feedback signals and send control signals to the pointing control unit 104. The field of view direction of the radar unit 105 refers to the center direction of the detectable range of the radar unit 105.

[0068] In some embodiments of the present invention, the aforementioned radio frequency signal is a continuous wave (CW) signal. In some embodiments of the present invention, the aforementioned radio frequency signal is a frequency-modulated continuous wave (FMCW) signal. In some embodiments of the present invention, the aforementioned radio frequency signal is a continuous wave signal with a fixed frequency.

[0069] Figure 2This is a schematic diagram of a radar detection system according to an embodiment of the present invention. Please refer to... Figure 2 The aforementioned antenna unit 101 further includes a transmitting antenna unit 201 and a receiving antenna unit 202. The transmitting antenna unit 201 includes multiple transmitting antennas 208-1 to 208-K. The transmitting antennas 208-1 to 208-K radiate the aforementioned radio frequency signal into free space. The receiving antenna unit 202 includes multiple receiving antennas 209-1 to 209-N and 210-1 to 210-M to receive feedback signals. Here, K, N, and M are positive integers representing the number of transmitting antennas 208-1 to 208-K and receiving antennas 209-1 to 209-N and 210-1 to 210-M configured. The actual number depends on the requirements of the radar detection system 100, and this invention does not impose any limitations on this number.

[0070] The design of a typical transmitting antenna needs to consider the frequency of the transmitted signal, the field of view, and the purpose to determine the antenna design. Antenna designs can employ lens antennas, patch antennas, or waveguide leaky-wave antennas. In some embodiments of this invention, transmitting antennas 208-1 to 208-K are patch antennas.

[0071] The design of a typical receiving antenna must consider the frequency of the received signal. To determine the direction of an object, multiple receiving antennas are needed. Receiving antennas typically contain several beams to receive echoes from objects at different azimuth angles, thereby determining the object's location. The design of the receiving antenna must consider the frequency range of the received radio frequency signal and whether it is necessary to determine the direction of the object under test. If direction determination is required, a design of a single-input multiple-output (SIMO) antenna or a multiple-input multiple-output (MIMO) antenna must be considered. In some embodiments of this invention, receiving antennas 209-1 to 209-N and 210-1 to 210-M are patch antennas, implemented using a printed circuit board.

[0072] The front-end unit 102 includes a signal generator 204, a transmitting unit 203, a receiving unit 205, a demodulation unit 206, and an analog-to-digital converter 207. The signal generator 204 generates a radio frequency (RF) signal and simultaneously transmits it to both the transmitting unit 203 and the demodulation unit 206. The transmitting unit 203 includes a power amplifier (PA) to amplify the RF signal and transmits the amplified RF signal to the transmitting antenna unit 201 to radiate the frequency-modulated signal into free space.

[0073] The receiving unit 205 includes a signal amplifier and a filter to receive the feedback signal received by the antenna unit 101, and to amplify and filter the received feedback signal. The demodulation unit 206 is coupled to the signal generator 204 and the receiving unit 205. The demodulation unit 206 receives the radio frequency signal generated by the signal generator 204 and the feedback signal amplified and filtered by the receiving unit 205, and demodulates the amplified and filtered feedback signal based on the radio frequency signal. The analog-to-digital converter 207 converts the demodulated feedback signal into a digital feedback signal and transmits it to the processing unit 103 for subsequent signal processing.

[0074] In some embodiments of the present invention, signal generator 204 generates a linearly modulated frequency signal with a start frequency of 77 GHz, an end frequency of 81 GHz, and a time period Tc of 40 μs. In some embodiments of the present invention, signal generator 204 generates a linearly modulated frequency signal with a start frequency of 24 GHz, an end frequency of 28 GHz, and a time period Tc of 40 μs. The linearly modulated frequency signal, after appropriate Fast Fourier Transform (FFT) signal processing, can be used to detect the position and velocity of an object, or even the object's breathing, heartbeat, etc. In this embodiment, demodulation unit 206 mixes and combines the frequency-modulated signal generated by signal generator 204 with the feedback signal amplified and filtered by receiving unit 205 to generate an intermediate frequency (IF) signal. Analog-to-digital converter 207 converts the IF signal into a digital feedback signal and transmits it to processing unit 103 for subsequent signal processing to obtain the information contained in the feedback signal.

[0075] In some embodiments of the present invention, the signal generator 204 generates a continuous wave signal of a single frequency. The single-frequency continuous wave signal, after being subjected to a Doppler Fast Fourier Transform, can be used to detect phenomena such as breathing and heartbeat. Simultaneously, the aforementioned plurality of transmitting antennas 208-1 to 208-K and plurality of receiving antennas 209-1 to 209-N and 210-1 to 210-M can be used to detect the direction of the object.

[0076] In some embodiments of the present invention, the pointing control unit 104 includes a horizontal direction adjustment drive circuit, a vertical direction adjustment drive circuit, a horizontal direction adjustment mechanism, a vertical direction adjustment mechanism, and a pointing sensor. The pointing control unit 104, according to the instructions of the processing unit 103, controls the field of view of the radar unit 105, performing horizontal and / or vertical direction adjustments. The pointing control unit 104 drives the horizontal direction adjustment mechanism and the vertical direction adjustment mechanism through the horizontal and vertical direction adjustment drive circuits to control the field of view of the radar unit 105. The horizontal and vertical direction adjustment mechanisms include motors and related mechanisms and electromechanical components. The aforementioned motors may be stepper motors or servo motors, and the present invention is not limited thereto. The pointing sensor reports the field of view of the radar unit 105. In this embodiment, the pointing sensor represents the field of view of the radar unit 105 using a set of horizontal and vertical angles. The pointing control unit 104 confirms whether the pointing instruction from the processing unit 103 has been completed based on the aforementioned set of horizontal and vertical angles. It is also worth noting that if the field of view of radar unit 105 is ±60° horizontally and ±60° vertically, then if the field of view of radar unit 105 is to be expanded to ±90° horizontally and ±90° vertically, then the pointing control unit 104 needs to be able to adjust the field of view direction of radar unit 105 by ±30° horizontally and ±30° vertically.

[0077] It should be noted that since the pointing control unit 104 can control the field of view of the radar unit 105, this increases the detectable range of the radar unit 105. In this invention, the current field of view of the radar unit 105 is called the current field of view (FOV). The detectable range achievable by controlling the field of view direction of the radar unit 105 is called the total field of view (FOV). Taking the aforementioned field of view of the radar unit 105 as ±60° horizontally and ±60° vertically, and expanding the field of view of the radar unit 105 to ±90° horizontally and ±90° vertically by controlling the field of view direction of the radar unit 105 as an example, the total field of view of the radar unit 105 is ±90° horizontally and ±90° vertically. One current field of view of the radar unit 105 is ±60° horizontally and ±60° vertically. Another current field of view of the radar unit 105 is ±60° horizontally and ±30° vertically.

[0078] Figure 3 This is a schematic diagram of a radar detection system according to an embodiment of the present invention. Figure 4-1 and Figure 4-2 This is a schematic diagram illustrating the adjustment of the field of view according to an embodiment of the present invention. Please also refer to... Figure 3 , Figure 4-1 and Figure 4-2 .like Figure 3As shown, radar element 105 is a multi-faceted antenna group (such as four-faced, six-faced, or eight-faced), and antenna element 101 includes multiple antenna groups arranged in a ring (such as...). Figure 3 (As shown). In Figure 3 In the illustrated structure, the pointing control unit 104 selects and switches to the antenna group corresponding to the desired field of view, thereby adjusting the field of view direction of the radar unit 105. Used in... Figure 3 The illustrated structure allows for rapid adjustment of the field of view of radar unit 105 without the need for an electromechanical drive mechanism. For example... Figure 4-1 and Figure 4-2 As shown, when the pointing control unit 104 selects antenna groups 401-1 to 401-5, the field of view of the radar unit 105 is field of view direction 1. When the pointing control unit 104 selects antenna groups 401-2 to 401-5, the field of view of the radar unit 105 is field of view direction 2.

[0079] Figure 6 This is a schematic diagram illustrating the current field of view and image coordinates according to an embodiment of the present invention. Figure 7 This is a schematic diagram illustrating the current field of view, the total field of view, and the image coordinates according to an embodiment of the present invention. Figure 8 This is a schematic diagram illustrating the current field of view, the total field of view, and image coordinates according to an embodiment of the present invention. Please also refer to... Figure 6 , Figure 7 and Figure 8 By projecting the object (object 602) detected within the current field of view of the radar unit 105 of the radar detection system 100 onto the plane 601 corresponding to the current field of view (object projection 602'), a one- or two-dimensional coordinate system can be given to the object detected within the current field of view of the radar unit 105, which is called the current field of view coordinate system. Figure 7 For example, plane 601 corresponding to the current field of view can be defined as a two-dimensional coordinate system using angles. The first dimension is the horizontal angle, ranging from ±50°. The second dimension is the vertical angle, ranging from ±40°. The field of view direction 702 corresponds to the origin 703 of the current field of view coordinate system.

[0080] like Figure 7 As illustrated, the total field of view obtained by controlling the field of view direction of radar unit 105 can also be defined by an angle into a corresponding total field of view coordinate system 701. The first dimension of the total field of view coordinate system 701 is the horizontal angle, ranging from ±90°. The second dimension is the vertical angle, also ranging from ±90°. Knowing the coordinates of the center point 703 of the plane 601 corresponding to the current field of view within the total field of view coordinate system 701 allows for easy conversion between the current field of view coordinates and the total field of view coordinate system 701.

[0081] The following describes in detail, with reference to the accompanying drawings, some embodiments of the radar field of view adjustment method of the present invention and how the various modules of the radar detection system 100 work together.

[0082] Figure 5-1 , Figure 5-2 and Figure 5-3 This is a schematic diagram illustrating the operation of a radar detection system 100 according to some embodiments of the present invention. Figure 12-1 The above is a flowchart illustrating a radar field of view adjustment method according to some embodiments of the present invention. Please also refer to... Figure 5-1 , Figure 5-2 , Figure 5-3 and Figure 12-1 The radar detection system 100 periodically radiates radio frequency signals to detect the area ( Figure 5-1 The detection area (schematic 504) is scanned and a feedback signal is obtained, wherein the radio frequency signal is the aforementioned linearly modulated frequency signal. The processing unit 103, corresponding to the aforementioned periodic scan, executes steps S1101 to S1102 within one scan cycle. In step S1101, the processing unit 103 obtains at least one object cluster in the current frame and the current state and next state of each object cluster based on the digital feedback signal. In step S1102, the processing unit 103 determines a displacement and, based on the current state and displacement of the tracking cluster corresponding to the target to be tracked in the current frame, determines whether the tracking cluster is located at the edge of the current field of view of the radar unit 105. In response to the tracking cluster being located at the edge of the current field of view, the processing unit 103 issues a control signal to the pointing control unit 104 based on the next state of the tracking cluster to adjust the field of view direction of the radar unit 105.

[0083] Figure 12-2 This is a flowchart illustrating a radar field-of-view adjustment method according to some embodiments of the present invention. Please refer to... Figure 12-2 ,exist Figure 12-2 In the illustrated embodiment, step S1101 further includes steps S1103 to S1105. In step S1103, the processing unit 103 obtains the current frame signal based on the aforementioned digital feedback signal and generates a point cloud map 500 based on the current frame signal. The point cloud map 500 includes a plurality of points 501. In step S1104, the processing unit 103 performs cluster analysis on the point cloud map 500 according to a clustering algorithm to classify the plurality of points 501 into at least one cluster corresponding to the current frame. Figures 5-1 to 5-3 In the illustrated embodiment, the processing unit 103 classifies the aforementioned plurality of points 501 into clusters 502 and 503 corresponding to the current frame. The processing unit 103 also obtains the current state of each cluster based on the current information of the current frame.

[0084] To identify a target, processing unit 103 maintains and tracks each cluster of objects in each frame. Starting from the first frame, processing unit 103 assigns an identification tag (which can be a numerical number, such as 1, 2, 3, etc.) to each cluster of objects detected in the first frame and calculates the current state of each cluster. Each cluster of objects detected in the first frame is recorded as a tracking object. Processing unit 103 also determines the tracking cluster corresponding to the target to be tracked (such as a caregiver) based on the current state of the first frame.

[0085] After the second frame is received, the processing unit 103 obtains the object clusters and their states for the second frame. Based on the past information of each tracked object (i.e., each object cluster in the first frame) in the first frame and the current state of each object cluster in the second frame, the processing unit 103 attempts to associate each object cluster in the second frame with each tracked object in the first frame that has an identification tag. If an object cluster in the second frame can be associated with a tracked object in the first frame, the identification tag of the tracked object in the first frame associated with it is used as the identification tag of the corresponding object cluster in the second frame. If an object cluster in the second frame cannot be associated with any tracked object in the first frame, the processing unit 103 assigns a new number to this object cluster in the second frame as the identification tag of this object cluster, and records this object cluster in the second frame as a new tracked object. And so on.

[0086] If no tracked object with a certain identification mark is detected in a preset D consecutive frames, the processing unit 103 will mark the cluster of objects with this identification mark as inactive and will no longer attempt to match the cluster of objects in the newly received frame.

[0087] Therefore, in step S1105, processing unit 103 uses a calculation program, past information from past frames, and the current state of each cluster in the current frame to confirm the identification marker of each cluster in the current frame. Processing unit 103 then predicts the next state of each cluster in the current frame based on the current state of each cluster. In some embodiments of the present invention, the next state of each cluster includes the next position of each cluster, such as... Figure 5-3 drawn.

[0088] Figure 13 This is a flowchart illustrating the point cloud generation process according to an embodiment of the present invention. In this embodiment, the radio frequency signal is the aforementioned linear frequency modulated continuous wave signal, and as... Figure 13As shown, the aforementioned step S1103 also includes steps S1201 to S1204. In step S1201, the processing unit 103 performs range processing and Doppler processing on the original data block formed by the digital feedback signal to obtain a processed data block. The range processing includes Range Fast Fourier Transform (Range FFT), and the Doppler processing includes Doppler Fast Fourier Transform (Doppler FFT). As previously mentioned, the demodulation unit 206 mixes and combines the frequency-modulated signal generated by the signal generator 204 with the feedback signal amplified and filtered by the receiving unit 205 to generate an intermediate frequency (IF) signal. To detect objects within different ranges (distances), we perform Fast Fourier Transform processing on the IF signal. Each peak after processing represents an object at a corresponding distance. This is called Range Fast Fourier Transform. For a target of interest, we can repeat the Range Fast Fourier Transform until we have enough data to perform a second-stage Fast Fourier Transform. The result of this second-stage Fast Fourier Transform is a two-dimensional multivalued matrix, whose peaks correspond to the Doppler frequency displacement of the moving target. This method is called the Doppler Fast Fourier Transform.

[0089] In step S1202, processing unit 103 performs moving target indication (MTI) on the processed data block to remove stationary points in the processed data block. In step S1203, after the aforementioned step S1202, processing unit 103 uses a detection algorithm to remove points generated by noise background in the processed data block.

[0090] It is worth noting that in some embodiments of the present invention, the aforementioned detection algorithm is a Constant False Alarm Rate (CFAR) algorithm. In other embodiments of the present invention, the detection algorithm is selected from one of the following groups: Cell-averaging CFAR (CA-CFAR), Greatest-Of-Cell-Average CFAR (GOCA-CFAR), Smallest-Of-Cell-Average CFAR (SOCA-CFAR), and Ordered Statistic CFAR (OS-CFAR).

[0091] In step S1204, after the aforementioned step S1203, the processed data blocks undergo angle processing to produce the following: Figure 5-1 The point cloud diagram 500 is shown. Angle processing includes Angle Fast Fourier Transform (Angle FFT). When two objects are at equal distances and have the same velocity relative to the radar detection system 100, range FFT and Doppler FFT are ineffective. Therefore, the angle of arrival (AoA) needs to be estimated. Since the distance from the object to each antenna is different, the estimation of the angle of arrival is based on the phasor change of the range FFT or Doppler FFT peaks, which requires at least two receiving antennas. Similarly, the angle estimation problem can be solved by performing a Fast Fourier Transform on the phasor sequence corresponding to the peaks of the two-dimensional Fast Fourier Transforms (range FFT and Doppler FFT). This method is called Angle Fast Fourier Transform.

[0092] In some embodiments of the present invention, the clustering algorithm in step S1104 is a density-based spatial clustering of applications with noise (DBSCAN) algorithm. In some embodiments of the present invention, the metric of the aforementioned density-based spatial clustering algorithm is as described in Equation 1 below, to reduce the contribution of the vertical z-axis (vertical axis) in the clustering:

[0093] D(p,q)=(p x -q x ) 2 +(p y -q y ) 2 +a×(p z -q z ) 2 , ………(Equation 1)

[0094] Where α is a real value less than 1, p x q x The x-coordinates of points p and q are respectively. y q y Let be the y-coordinates of points p and q, respectively, and p z q z Let p and q be the z-coordinates of points p and q, respectively.

[0095] In some embodiments of the present invention, α is selected as 0.05, and in some embodiments of the present invention, α is selected as 0.25.

[0096] In some embodiments of the present invention, the clustering algorithm is selected as peak grouping, modified density-based spatial clustering of applications with noise (modified DBSCAN), or hierarchical density-based spatial clustering of applications with noise (HDBSCAN).

[0097] In some embodiments of the present invention, the current state of each cluster of objects in the current frame described in step S1104 includes the position of the cluster. The position of each cluster is determined by the physical center of that cluster. In some embodiments of the present invention, the physical center of the cluster is the centroid of the plurality of corresponding point clouds of that cluster. In some embodiments of the present invention, such as Figure 5-2 As shown, each cluster includes an outer frame, and the processing unit 103 uses the center point of the outer frame of each cluster as the physical center of each cluster to determine the position of the current state of each cluster.

[0098] In some embodiments of the present invention, the aforementioned outer frame is rectangular. In some embodiments of the present invention, the aforementioned outer frame is circular. In some embodiments of the present invention, the aforementioned outer frame is elliptical.

[0099] It is worth noting that in the aforementioned embodiment, the radar detection system 100 periodically radiates a linearly modulated frequency signal to detect the area ( Figure 5-1 The radar detection system 100 can also use a fixed-frequency continuous wave signal to scan the detection area (see schematic diagram 504) and obtain a feedback signal. Figure 5-1 The detection area diagram 504 in the image is scanned to obtain a feedback signal. At this time, the processing unit 103 performs Doppler processing and angle processing on the original data block formed by the digital feedback signal to generate a 2D point cloud map. The processing unit 103 then performs cluster analysis on the aforementioned 2D point cloud map according to a clustering algorithm to classify the points in the aforementioned 2D point cloud map into at least one cluster corresponding to the current frame.

[0100] Figure 14 For a flowchart illustrating the radar field of view adjustment method according to some embodiments of the present invention, please refer to [link / reference]. Figure 14 As illustrated in the embodiment shown in Figure 12 above, the processing unit 103 maintains a set of tracking objects based on the incoming frame, each tracking object having an identification tag. Figure 14In the illustrated embodiment, the calculation procedure described in step S1105 includes an association algorithm and a tracking algorithm. The past information of the aforementioned past frames includes the tracking object status of each tracked object. The current status of each cluster in the current frame includes the position of each cluster, and the tracking object status of each tracked object includes the position of the individual tracked object. Step S1105 includes steps S1301 and S1302. In step S1301, the processing unit 103 uses the aforementioned association algorithm to establish an association between each cluster and each tracked object based on the position of each tracked object and the position of each cluster to confirm the identification mark of each cluster.

[0101] In some embodiments of the present invention, the aforementioned association algorithm is the Hungarian algorithm. Processing unit 103 uses the Hungarian algorithm to establish associations between each object cluster and each tracked object in the current frame. Processing unit 103 establishes a cost matrix for each object cluster and each tracked object in the current frame, and uses minimizing the combined distance loss to allocate each object cluster in the current frame to each tracked object. It should be noted that the cost matrix can be non-square, because the number of tracked objects and the number of object clusters in the current frame may be different. The elements M of the cost matrix... i,j Set to the distance between the tracked object i and the cluster j in the current frame. If M i,j If the value exceeds a threshold, the processing unit 103 is set to a large real number L to avoid correlation.

[0102] After the processing unit 103 establishes an association between each cluster of objects and each tracked object in the current frame, in step S1302, the processing unit 103 uses a tracking algorithm to predict the next position of each cluster of objects based on the position of each cluster of objects. In this embodiment, the tracking algorithm is a Kalman filter.

[0103] Figure 15 This is a flowchart illustrating a radar field-of-view adjustment method according to some embodiments of the present invention. Please refer to... Figure 15 Following step S1302, step S1105 further includes step S1401. In step S1401, if there is an unassociated tracked object among the tracked objects that has not been associated with any of the object clusters in the current frame, the processing unit 103 uses the aforementioned Kalman filter to predict the next position of each unassociated tracked object based on its position to provide a position prediction for each unassociated tracked object.

[0104] It is worth noting that in the foregoing embodiments, the processing unit 103 uses a Kalman filter as the tracking algorithm. In some embodiments of the present invention, the foregoing tracking algorithm may also use an extended Kalman filter (EKF), an unscented Kalman filter, an interactive multiple model (IMM) algorithm, or a probability hypothesis density filter, and the present invention is not limited thereto.

[0105] It is also worth noting that in the foregoing embodiments, the processing unit 103 uses the Hungarian algorithm as the association algorithm. In some embodiments of the present invention, the aforementioned association algorithm may also use global nearest neighbor (GNN), probabilistic data association filter, or joint probabilistic data association filter, and the present invention is not limited thereto.

[0106] Figure 16 This is a flowchart illustrating a radar field-of-view adjustment method according to some embodiments of the present invention. Please refer to... Figure 16 The calculation procedure described in step S1105 includes a probability calculation procedure and a tracking algorithm. The current state of each cluster contains a state component, and the past information of past frames contains the state component related to the current state of each cluster and the probability distribution associated with the identification tag. In this embodiment, the aforementioned state component is the position of the cluster. The past information of past frames contains the probability P of each tracked object appearing at its current coordinates. i P1(s) represents the probability that the tracked object with identification mark 1 appears at each current coordinate, and P2(s) represents the probability that the tracked object with identification mark 2 appears at each current coordinate.

[0107] The aforementioned probability calculation procedure includes calculating the probability that the identification marker of each cluster is a number based on the aforementioned probability distribution and the position of each cluster. The aforementioned step S1105 includes steps S1501 and S1502. In step S1501, the processing unit 103 confirms the identification marker of each cluster based on the probability that the identification markers of all clusters are a number. For example, if the position of a cluster A is x, the processing unit 103 can estimate the probability that cluster A is a tracking object marked 1 (P1(x)) and the probability that cluster A is a tracking object marked 2 (P2(x)). The processing unit 103 compares the magnitudes of P1(x) and P2(x). If P1(x) is greater than or equal to P2(x), the processing unit 103 determines that cluster A is a tracking object marked 1; otherwise, the processing unit 103 determines that cluster A is a tracking object marked 2, and so on.

[0108] In step S1502, the processing unit 103 uses the aforementioned Kalman filter as a tracking algorithm to predict the next state of each cluster based on the current state (including position) of each cluster, and continues to update the aforementioned probability distribution.

[0109] As mentioned above, the processing unit 103 may also use an extended Kalman filter, an unscented Kalman filter, an interactive multiple model algorithm, or a probability assumption density filter as the tracking algorithm, and the present invention does not limit it.

[0110] One embodiment of the present invention utilizes the location and physiological characteristics of clusters to determine the identification markers of clusters. Figure 17 This is a flowchart illustrating a radar field-of-view adjustment method according to some embodiments of the present invention. Please refer to... Figure 17 The calculation procedure described in step S1105 includes a probability calculation procedure and a tracking algorithm. The current state of each cluster includes its location and physiological characteristics, where the physiological characteristic is heart rate. Past information from past frames includes a first probability distribution related to the location of each cluster and the identification tag, and a second probability distribution related to the heart rate of each cluster and the identification tag.

[0111] In this embodiment, the past information of past frames includes the probability PP of each tracked object appearing at each current coordinate. i (s), the probability that each tracked object is at its current heart rate u is PH. i PH1(u) represents the probability that the tracked object with identification mark 1 appears at each coordinate, where u is the numerical value of the heart rate and i is the numerical code of the identification mark. For example, PP1(s) represents the probability that the tracked object with identification mark 1 appears at each coordinate, and PP2(s) represents the probability that the tracked object with identification mark 2 appears at each coordinate. PH1(u) represents the probability that the tracked object with identification mark 1 has a current heart rate of u.

[0112] In this embodiment, the aforementioned probability calculation procedure includes calculating a first probability that the identification marker of each cluster is a specific number based on a first probability distribution and the position of each cluster, and calculating a second probability that the identification marker of each cluster is the aforementioned specific number based on a second probability distribution and the heart rate of each cluster. The aforementioned step S1105 includes steps S1601 and S1602. In step S1601, the processing unit 103 confirms the identification marker of each cluster based on the first probability and the second probability that the identification marker of all clusters is a specific number. For example, if the position of a cluster A is x and its heart rate is y, then the processing unit 103 can estimate the probability that cluster A is a tracking object marked as 1 (PP1(x)) and the probability that cluster A is a tracking object marked as 2 (PP2(x)) based on its position; the processing unit 103 can also estimate the probability that cluster A is a tracking object marked as 1 (PH1(y)) and the probability that cluster A is a tracking object marked as 2 (PH2(y)) based on its heart rate. In this embodiment, the processing unit 103 compares the magnitudes of aPP1(x)+bPH1(y) and aPP2(x)+bPH2(y), where a and b are two predetermined positive real numbers.

[0113] If aPP1(x) + bPH1(y) is greater than or equal to aPP2(x) + bPH2(y), then processing unit 103 determines that cluster A is a tracking object marked as 1; otherwise, processing unit 103 determines that cluster A is a tracking object marked as 2, and so on.

[0114] In step S1602, the processing unit 103 uses the aforementioned Kalman filter as a tracking algorithm to predict the next state of each cluster based on the current state of each cluster (including position and heart rate), and continues to update the aforementioned probability distribution.

[0115] It is worth noting that in the foregoing embodiments, the physiological characteristic selected is heart rate, while in some embodiments of the present invention, the physiological characteristic selected is respiratory rate. The selected physiological characteristic is only required to be detectable by the radar unit 105 and to represent the cluster of objects in the current frame; the present invention does not limit it.

[0116] Please refer to this again. Figure 6 and Figure 7 As previously mentioned, by projecting the detected object (object 602) within the current field of view of the radar unit 105 of the radar detection system 100 onto the plane 601 corresponding to the current field of view (object projection 602'), a current field of view coordinate system defined by angle can be given to the current field of view of the radar unit 105. The total field of view obtained by controlling the field of view direction of the radar unit 105 can also be defined by angle into a corresponding total field of view coordinate system 701. Please also refer to... Figure 6 , Figure 7 and Figure 8Using the aforementioned current field-of-view coordinate system and global field-of-view coordinate system 701, the processing unit 103 can convert the movement coordinates of the object detected by the radar unit 105. For example, Figure 6 Object 602 moves to become object 603, and object 603 is projected onto plane 601 corresponding to the current field of view as object projection 603'. Processing unit 103 can, as follows: Figure 8 As illustrated, two-dimensional coordinates of object projections 602' and 603' are assigned based on their positions on plane 601 corresponding to the current field of view. Processing unit 103 can determine whether object 602 will move outside the current field of view of radar unit 105 based on the two-dimensional coordinates of object projections 602' and 603'.

[0117] Figure 9 This is a schematic diagram illustrating the current field of view, the total field of view, and image coordinates according to an embodiment of the present invention. Please also refer to... Figures 6 to 9 In some embodiments of the present invention, the displacement amount mentioned in step S1104 is a predetermined fixed amount, which includes a first fixed component and a second fixed component. For example, the first fixed component is 15° and the second fixed component is 10°. Furthermore, the current state of each cluster of objects includes the current coordinate position in the current view coordinate system. Step S1104 further includes a processing unit 103 determining the edge of the current view based on the first fixed component and the second fixed component. Taking the first fixed component as 15° and the second fixed component as an example, the edge of the current view is... Figure 7 In plane 601, there are regions with horizontal angles ranging from 35° to 50° and from -50° to -35°, and vertical angles ranging from 30° to 40° and from -40° to -30°. Processing unit 103 determines whether the tracking cluster is located at the edge of the current field of view of radar unit 105 based on the current coordinate position of the tracking cluster in the current field of view coordinate system.

[0118] Figure 18 The above is a flowchart illustrating a radar field of view adjustment method according to some embodiments of the present invention. Please also refer to... Figures 6 to 9 and Figure 18 ,exist Figure 18In the illustrated embodiment, the current state of each cluster includes its current coordinate position in the current field of view coordinate system and its current velocity obtained with reference to the current field of view coordinate system. The aforementioned step S1102 includes steps S1701 and S1702. In step S1701, the processing unit 103 determines the first fixed component and the second fixed component of the aforementioned displacement amount based on the reaction time required to adjust the field of view direction of the radar unit 105 and the current velocity of the tracking cluster. It also determines the edge of the current field of view based on the first fixed component and the second fixed component of the displacement amount. For example, if the processing unit 103 determines the reaction time required to adjust the field of view direction of the radar unit 105 as T, and the current velocity of the tracking cluster has a first component V1 and a second component V2 with reference to the current field of view coordinate system, then the first fixed component is set to V1T, and the second fixed component is set to V2T.

[0119] In step S1702, the processing unit 103 determines whether the tracking cluster is located at the edge of the current field of view of the radar unit 105 based on the current coordinate position of the tracking cluster in the current field of view coordinate system.

[0120] In this embodiment, in response to the tracking cluster being located at the edge of the current field of view, the processing unit 103, based on the estimated next position of the tracking cluster, issues a control signal to adjust the field of view direction of the radar unit 105 so that the origin of the current field of view coordinate system is, for example, Figure 9 The origin of the current view coordinate system is 802, which is close to the projection of the tracking cluster, such as the object projection 603'.

[0121] In the foregoing embodiments, since the current field of view coordinate system related to the field of view of the radar unit 105 is used, the processing unit 103 can quickly calculate the control signal to adjust the field of view of the radar unit 105.

[0122] Figure 10 The diagram illustrating the current field of view, total field of view, and image coordinates according to an embodiment of the present invention is shown below. Please also refer to... Figures 6 to 10 .exist Figure 10 In the diagram, trajectory points 803-1 to 803-7 represent the movement trajectory of the object's projection 603'. The field of view of radar unit 105 is determined by... Figure 8 The field of view direction 702 moves as Figure 9 The view direction is 804. At this time, the center point 802 of the current view plane 601 after the movement does not overlap with the center point 801 of the total view coordinate system 701. Because the processing unit 103 stores the coordinate position of the center point 802 of the current view plane 601 after the movement in the total view coordinate system 701, the processing unit 103 can represent the positions of trajectory points 803-1 to 803-7 in the coordinates of the total view coordinate system 701 and store them in the processing unit 103.

[0123] Figure 11 This is a schematic diagram illustrating the structure of a processing unit 1100 according to some embodiments of the present invention. Figure 11 As shown, at the hardware level, the processing unit 1100 includes a processor 1101, internal memory 1102, and non-volatile memory 1103. Internal memory 1102 is, for example, random-access memory (RAM). Of course, the processing unit 1100 may also include other hardware required for its functions.

[0124] Internal memory 1102 and non-volatile memory 1103 are used to store programs, which may include program code, including computer operation instructions. Internal memory 1102 and non-volatile memory 1103 provide instructions and data to processor 1101. Processor 1101 reads the corresponding computer program from non-volatile memory 1103 into internal memory 1102 and then runs it. Processor 1101 is specifically used to execute the instructions shown in Figure 12. Figure 18 The steps described herein.

[0125] Processor 1101 may be an integrated circuit chip with signal processing capabilities. In implementation, the methods and steps disclosed in the foregoing embodiments can be performed by the integrated logic circuitry in the hardware of processor 501 or by instructions in software form. Processor 1101 may be a general-purpose processor, including a Central Processing Unit (CPU), a Tensor Processing Unit (TPU), a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, capable of implementing or executing the methods and steps disclosed in the foregoing embodiments.

[0126] In some embodiments of the present invention, a computer-readable recording medium storing a program is also provided. The computer-readable recording medium stores at least one instruction, which, when executed by the processor 1101 of the processing unit 1100, enables the processor 1101 of the processing unit 1100 to perform the aforementioned... Figures 12-1 to 18 The steps described.

[0127] Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random-access memory (SRAM), dynamic random-access memory (DRAM), other types of random-access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other internal memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable media that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient media, such as modulated data signals and carrier waves.

[0128] Although the present invention has been disclosed in conjunction with the above embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications and refinements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be defined by the appended claims.

Claims

1. A radar detection system, comprising: The radar unit is configured to generate a radio frequency signal, radiate the radio frequency signal and receive a feedback signal, and demodulate and digitize the feedback signal to obtain a digital feedback signal. The pointing control unit is configured to adjust the field of view of the radar unit according to control signals; as well as The processing unit is configured to perform the following steps during a scan round: (a) Based on the digital feedback signal, obtain at least one cluster of objects in the current frame and the current state and next state of each of the at least one cluster of objects; as well as (b) Determine the displacement and, based on that displacement, determine the edge of the current horizon; Based on the current state of the tracking cluster within the at least one object cluster, it is determined whether the tracking cluster is located at the edge of the current field of view of the radar unit; in response to the tracking cluster being located at the edge of the current field of view, a control signal is issued based on the next state of the tracking cluster to adjust the direction of the field of view. The current state of each of the at least one cluster of objects includes the current coordinate position and the current velocity. The aforementioned step (b) includes: the processing unit determining the displacement based on the reaction time required to adjust the field of view of the radar unit and the current velocity, and determining whether the tracking cluster is located at the edge of the current field of view of the radar unit based on the current coordinate position.

2. The radar detection system as described in claim 1, wherein, The processing unit is configured to perform step (a) during the scan round, which includes: (a1) Generate a point cloud map based on the digital feedback signal; (a2) Perform cluster analysis on the point cloud map according to the clustering algorithm to obtain the at least one cluster of objects in the current frame and obtain the current state of each at least one cluster of objects based on the current information of the current frame; as well as (a3) Using the calculation program, past information of past frames and the current state of each of the at least one cluster, the identification mark of each of the at least one cluster is confirmed, and the next state of each of the at least one cluster is predicted based on the current state of each of the at least one cluster.

3. The radar detection system as described in claim 2, wherein, The displacement is a predetermined fixed amount, and the current state of each of the at least one cluster of objects includes the current coordinate position. The aforementioned step (b) includes the processing unit determining the edge of the current field of view based on the displacement, and determining whether the tracking cluster is located at the edge of the current field of view of the radar unit based on the current coordinate position of the tracking cluster.

4. The radar detection system as claimed in claim 2, wherein the calculation program includes an association algorithm and a tracking algorithm, the past information of the past frame includes the tracking object state of at least one tracking object, and the aforementioned step (a3) ​​includes: the processing unit using the association algorithm to establish an association between the at least one object cluster and the at least one tracking object based on the tracking object state of each of the at least one tracking object and the current state of each of the at least one object cluster to confirm the identification mark of each of the at least one object cluster, and the processing unit using the tracking algorithm to predict the next state of each of the at least one object cluster based on the current state of each of the at least one object cluster.

5. The radar detection system of claim 4, wherein the calculation program includes the association algorithm and the tracking algorithm, the past information of the past frame includes the tracking object state of the at least one tracking object, and the aforementioned step (a3) ​​further includes: in response to the existence of at least one unassociated tracking object among the at least one tracking object that has not been associated with the at least one cluster of objects, the processing unit uses the tracking algorithm to predict the next tracking object state of each of the at least one unassociated tracking object based on the tracking object state of each of the at least one unassociated tracking object to provide a state prediction for each of the at least one unassociated tracking object.

6. The radar detection system as described in claim 4 or 5, wherein the correlation algorithm is selected from one of the group consisting of the free Hungarian algorithm, the global nearest neighbor algorithm, the probability data correlation filter, and the joint probability data correlation filter.

7. The radar detection system of claim 2, wherein the calculation program includes a probability calculation program and a tracking algorithm, the current state of each of the at least one clusters includes a state component, the past information of the past frame includes at least one probability distribution related to the state component of the current state of each of the at least one clusters and the identification mark, the probability calculation program includes calculating the probability that the identification mark of each of the at least one clusters is an object in a category based on the at least one probability distribution and the state component of the current state of each of the at least one clusters; the aforementioned step (a3) ​​includes: confirming the identification mark of each of the at least one clusters based on the probability that the identification mark of each of the at least one clusters is an object in the category; and the processing unit uses the tracking algorithm to predict the next state of each of the at least one clusters based on the current state of each of the at least one clusters, wherein the state component of the current state is one of the group consisting of a free position and physiological characteristics.

8. The radar detection system as described in claim 2, wherein, The radio frequency signal is a frequency-modulated continuous wave signal, and the aforementioned step (a1) includes: (a11) Perform range processing and Doppler processing on the original data block formed by the digital feedback signal to obtain the processed data block; (a12) Provide an active target indication to the processed data block to remove stationary points in the processed data block; (a13) Following step (a12) above, a detection algorithm is used to remove points generated by the noisy background from the processed data block; and (a14) After the aforementioned step (a13), the angle of the processed data block is processed to generate the point cloud map.

9. The radar detection system as described in claim 2, wherein, The clustering algorithm is selected from one of the following algorithm groups: free peak clustering algorithm, density space-based clustering algorithm, improved density space-based clustering algorithm, and hierarchical density space-based clustering algorithm; and the processing unit determines the position of the current state of the cluster for each of the at least one cluster based on the physical center of the cluster, wherein the physical center is selected from one of the following groups: the centroids of multiple corresponding point clouds of the cluster and the center point of the outer frame of the cluster.

10. The radar detection system of claim 2, wherein the radar unit comprises an antenna unit and a front-end unit, the antenna unit being configured to radiate the radio frequency signal into free space and receive the feedback signal, and the front-end unit being configured to generate the radio frequency signal and demodulate and digitize the feedback signal to obtain the digital feedback signal.

11. A radar field-of-view direction adjustment method, applicable to a radar detection system, executed by a processing unit, the radar detection system comprising: The radar unit is configured to generate a radio frequency signal, radiate the radio frequency signal and receive a feedback signal, and demodulate and digitize the feedback signal to obtain a digital feedback signal. The pointing control unit is configured to adjust the field of view of the radar unit according to control signals; and This processing unit; The radar field of view adjustment method includes: During the scanning round: (a) Based on the digital feedback signal, obtain at least one cluster of objects in the current frame and the current state and next state of each of the at least one cluster of objects; as well as (b) Determine the displacement and, based on that displacement, determine the edge of the current horizon; Based on the current state of the tracking cluster within the at least one object cluster, it is determined whether the tracking cluster is located at the edge of the current field of view of the radar unit; in response to the tracking cluster being located at the edge of the current field of view, a control signal is issued based on the next state of the tracking cluster to adjust the field of view direction of the radar unit. The current state of each of the at least one cluster of objects includes the current coordinate position and the current velocity. The aforementioned step (b) includes: the processing unit determining the displacement based on the reaction time required to adjust the field of view of the radar unit and the current velocity, and determining whether the tracking cluster is located at the edge of the current field of view of the radar unit based on the current coordinate position.

12. The radar field of view adjustment method as described in claim 11, wherein, Step (a) includes: (a1) Generate a point cloud map based on the digital feedback signal; (a2) Perform cluster analysis on the point cloud map according to the clustering algorithm to obtain the at least one cluster of objects in the current frame and obtain the current state of each at least one cluster of objects based on the current information of the current frame; as well as (a3) Using the calculation program, past information of past frames and the current state of each of the at least one cluster, the identification mark of each of the at least one cluster is confirmed, and the next state of each of the at least one cluster is predicted based on the current state of each of the at least one cluster.

13. The radar field of view adjustment method as described in claim 12, wherein, The displacement is a predetermined fixed amount, and the current state of each of the at least one cluster of objects includes the current coordinate position. The aforementioned step (b) includes the processing unit determining the edge of the current field of view based on the displacement, and determining whether the tracking cluster is located at the edge of the current field of view of the radar unit based on the current coordinate position of the tracking cluster.

14. The radar field of view adjustment method as described in claim 12, wherein the calculation program includes an association algorithm and a tracking algorithm, the past information of the past frame includes the tracking object state of at least one tracking object, and the aforementioned step (a3) ​​includes: the processing unit uses the association algorithm to establish an association between the at least one object cluster and the at least one tracking object based on the tracking object state of each of the at least one tracking object and the current state of each of the at least one object cluster to confirm the identification mark of each of the at least one object cluster, and the processing unit uses the tracking algorithm to predict the next state of each of the at least one object cluster based on the current state of each of the at least one object cluster.

15. The radar field of view adjustment method as claimed in claim 14, wherein the calculation program includes the association algorithm and the tracking algorithm, the past information of the past frame includes the tracking object state of the at least one tracking object, and the aforementioned step (a3) ​​further includes: in response to the existence of at least one unassociated tracking object among the at least one tracking object that has not been associated with the at least one object cluster, the processing unit uses the tracking algorithm to predict the next tracking object state of each of the at least one unassociated tracking objects based on the tracking object state of each of the at least one unassociated tracking objects to provide a state prediction for each of the at least one unassociated tracking objects.

16. The radar field of view adjustment method as described in claim 14 or 15, wherein the correlation algorithm is selected from one of the group consisting of the free Hungarian algorithm, the global nearest neighbor algorithm, the probability data correlation filter, and the joint probability data correlation filter.

17. The radar field of view adjustment method as claimed in claim 12, wherein the calculation program includes a probability calculation program and a tracking algorithm, the current state of each of the at least one clusters includes a state component, the past information of the past frame includes at least one probability distribution related to the state component of the current state of each of the at least one clusters and the identification mark, the probability calculation program includes calculating the probability that the identification mark of each of the at least one clusters is a number based on the at least one probability distribution and the state component of the current state of each of the at least one clusters; the aforementioned step (a3) ​​includes: confirming the identification mark of each of the at least one clusters based on the probability that the identification mark of each of the at least one clusters is the number; and the processing unit uses the tracking algorithm to predict the next state of each of the at least one clusters based on the current state of each of the at least one clusters.

18. The radar field of view adjustment method as described in claim 17, wherein the state component of the current state is one of the group consisting of a free position and physiological characteristics.

19. The radar field of view adjustment method as claimed in claim 12, wherein the calculation program includes a probability calculation program and a tracking algorithm, the current state of each of the at least one clusters includes position and physiological characteristics, the past information of the past frame includes at least one first probability distribution related to the position and the identification mark of the current state of each of the at least one clusters, and the past information of the past frame includes at least one second probability distribution related to the physiological characteristics and the identification mark of the current state of each of the at least one clusters, and the probability calculation program includes calculation based on the at least one first probability distribution and each of the at least one clusters. The processing unit calculates a first probability that the identification mark of each of the at least one clusters is the number based on the current state and the position of the current state, and calculates a second probability that the identification mark of each of the at least one clusters is the number based on the at least one second probability distribution and the physiological characteristics of the current state of each of the at least one clusters; the aforementioned step (a3) ​​includes: confirming the identification mark of each of the at least one clusters based on the first probability and the second probability that the identification mark of each of the at least one clusters is the number; and the processing unit uses the tracking algorithm to predict the next state of each of the at least one clusters based on the current state of each of the at least one clusters.

20. The radar field of view adjustment method as described in any one of claims 14, 15, 17 or 18, wherein the tracking algorithm is selected from the group consisting of a free extended Kalman filter, an unscented Kalman filter, an interactive multiple model algorithm, and a probability assumption density filter.

21. The radar field of view adjustment method as described in claim 12, wherein, The radio frequency signal is a frequency-modulated continuous wave signal, and the aforementioned step (a1) includes: (a11) Perform range processing and Doppler processing on the original data block formed by the digital feedback signal to obtain the processed data block; (a12) Provide an active target indication to the processed data block to remove stationary points in the processed data block; (a13) Following step (a12) above, a detection algorithm is used to remove points generated by the noisy background from the processed data block; and (a14) After the aforementioned step (a13), the angle of the processed data block is processed to generate the point cloud map.

22. The radar field of view adjustment method as described in claim 21, wherein, The detection algorithm selects one of the following groups: free fixed false alarm rate, average fixed false alarm rate, maximum average fixed false alarm rate, minimum average fixed false alarm rate, and ordered statistic fixed false alarm rate.

23. The radar field of view adjustment method as described in claim 12, wherein, This clustering algorithm is selected from one of the algorithm groups consisting of the free peak clustering algorithm, the density space-based clustering algorithm, the improved density space-based clustering algorithm, and the hierarchical density space-based clustering algorithm.

24. The radar field of view adjustment method as claimed in claim 23, wherein the processing unit determines the position of the current state of the at least one cluster of objects based on the physical center of the cluster, wherein, The physical center is selected from one of the groups consisting of the centroids of multiple corresponding point clouds of the free cluster and the center point of the outer frame of the cluster.

Images