WO2021248472A1 - 基于超宽带雷达的目标跟踪方法、装置、设备及存储介质 - Google Patents

基于超宽带雷达的目标跟踪方法、装置、设备及存储介质 Download PDF

Info

Publication number
WO2021248472A1
WO2021248472A1 PCT/CN2020/095889 CN2020095889W WO2021248472A1 WO 2021248472 A1 WO2021248472 A1 WO 2021248472A1 CN 2020095889 W CN2020095889 W CN 2020095889W WO 2021248472 A1 WO2021248472 A1 WO 2021248472A1
Authority
WO
WIPO (PCT)
Prior art keywords
measurement
radar
radar frame
frame
ultra
Prior art date
Application number
PCT/CN2020/095889
Other languages
English (en)
French (fr)
Inventor
周金海
周世镒
常阳
吴耿俊
王依川
石聪
杨宁
唐海
Original Assignee
浙江大学
Oppo广东移动通信有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 浙江大学, Oppo广东移动通信有限公司 filed Critical 浙江大学
Priority to CN202080001113.7A priority Critical patent/CN114080549A/zh
Priority to PCT/CN2020/095889 priority patent/WO2021248472A1/zh
Publication of WO2021248472A1 publication Critical patent/WO2021248472A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/66—Radar-tracking systems; Analogous systems
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/66—Radar-tracking systems; Analogous systems
    • G01S13/70—Radar-tracking systems; Analogous systems for range tracking only

Definitions

  • the embodiments of the present application relate to the technical field of environmental assisted living, and in particular, to a target tracking method, device, equipment, and storage medium based on ultra-wideband radar.
  • the Ambient Assisted Living (AAL) system refers to the use of modern inductive transmission devices to connect all kinds of home appliances on an expandable intelligent technology platform, thereby constructing an environment that can reflect in real time. Analyze the status of the householder, make judgments and reflect immediately, etc.
  • AAL Ambient Assisted Living
  • target tracking is a very important link. Through target tracking, the position and speed of the target object can be calculated, and then combined with a certain prior knowledge of the environment, the analysis of the amount of activity of the target object and the estimation of the degree of disability can be carried out.
  • target tracking is mainly achieved through cameras and/or smart wearable devices.
  • the camera can collect the image and video information of the target object, and the smart wearable device can collect information such as the movement state and physical status of the target object. After the camera and/or smart wearable device collects the corresponding information, the information can be transmitted to the platform where the information is located. Computer equipment for further analysis and processing.
  • the camera has certain requirements on the brightness of the collection environment, and the image or video information collected by the camera is likely to reveal privacy; for the method of target tracking through smart wearable devices, due to smart Wearable devices are invasive, so many people do not have the habit of wearing smart wearable devices.
  • the target tracking in the related technology has many of the above-mentioned defects, how to take into account the use habits of the target object and ensure the privacy of the target object while performing target tracking, still needs further discussion and research.
  • the embodiments of the application provide a target tracking method, device, equipment, and storage medium based on ultra-wideband radar.
  • the technical solution is as follows:
  • an embodiment of the present application provides a target tracking method based on ultra-wideband radar, and the method includes:
  • the position information of at least one target object is determined.
  • an ultra-wideband radar-based target tracking device which includes:
  • the information acquisition module is used to acquire the radar frame obtained by sampling the echo signal by the ultra-wideband radar;
  • the measurement determination module is configured to extract a scattering point based on the radar frame to obtain a measurement corresponding to the radar frame.
  • the measurement refers to the distance unit where the scattering point is located, and the measurement is related to Point-to-point correspondence;
  • An aggregation processing module configured to perform aggregation processing on the measurement corresponding to the radar frame to obtain the aggregation measurement corresponding to the radar frame;
  • the position determining module is configured to determine the position information of at least one target object according to the aggregated measurement corresponding to the radar frame.
  • an embodiment of the present application provides a computer device, the computer device includes a processor and a memory, the memory stores a computer program, and the computer program is loaded and executed by the processor to implement the above-mentioned super Target tracking method of broadband radar.
  • an embodiment of the present application provides a computer-readable storage medium in which a computer program is stored, and the computer program is used to be executed by a processor of a computer device to implement the above-mentioned ultra-wideband radar-based Target tracking method.
  • an embodiment of the present application provides a chip, the chip includes a programmable logic circuit and/or program instructions, when the chip runs on a computer device, it is used to achieve the above-mentioned ultra-wideband radar-based target Tracking method.
  • the embodiments of the present application provide a computer program product, which when the computer program product runs on a computer device, causes the computer device to execute the above-mentioned ultra-wideband radar-based target tracking method.
  • the radar frame obtained by sampling the echo signal by the ultra-wideband radar, and extract the scattering point from the radar frame, and use the distance unit of the scattering point as the measurement to obtain the measurement corresponding to the radar frame, and then aggregate the measurement .
  • the technical solution provided by the embodiments of the present application reduces the number of measurements due to the aggregation processing of the measurements, thereby effectively alleviating the storage pressure of the computer equipment, reducing the computing overhead of the computer equipment, and avoiding waste of resources.
  • the embodiment of the present application applies the ultra-wideband radar to the detection and tracking of the target object, which is sufficient Taking into account the user's usage habits, it helps to protect the user's privacy.
  • FIG. 1 is a schematic diagram of a system architecture provided by an embodiment of the present application.
  • FIG. 2 is a flowchart of a target tracking method based on ultra-wideband radar provided by an embodiment of the present application
  • FIG. 3 is a flowchart of a scatter point extraction process provided by an embodiment of the present application.
  • FIG. 5 is a waveform diagram of different forms of radar frames provided by an embodiment of the present application.
  • FIG. 6 is a flowchart of a target tracking method based on ultra-wideband radar provided by another embodiment of the present application.
  • FIG. 7 is a schematic diagram of a target tracking result provided by an embodiment of the present application.
  • FIG. 8 is a block diagram of an ultra-wideband radar-based target tracking device provided by an embodiment of the present application.
  • FIG. 9 is a block diagram of an ultra-wideband radar-based target tracking device provided by another embodiment of the present application.
  • Fig. 10 is a structural block diagram of a server provided by an embodiment of the present application.
  • FIG. 1 shows the system architecture of the environmental assisted living system provided by an embodiment of the present application.
  • the system architecture includes an ultra-wideband radar 10, a computer device, and a target object 30.
  • the target object 30 refers to the activity object that needs attention in the environmental assisted living system.
  • the target object 30 is a person.
  • the target object 30 refers to the empty-nest elderly.
  • the target object 30 refers to animals; when the technical solution of the present application is applied to attention to robots in office environments, etc., the The target object 30 refers to a robot.
  • the embodiment of the application does not limit the number of target objects 30.
  • the number of target objects 30 is usually one or two.
  • Ultra Wide-Band (UWB) radar 10 refers to a radar with a Fractional Bandwidth (FBW) of a transmitted signal greater than 0.25, which can implement functions such as communication and detection.
  • the ultra-wideband radar 10 can detect the target object 30 and collect echo signals.
  • the ultra-wideband radar 10 in the embodiment of the present application is a pulsed radar sensor.
  • the maintainer terminal 22 in the computer device can control the ultra-wideband radar 10 to intermittently transmit a pulsed periodic signal, and when transmitting The reflected echo signal is received at intervals, that is, the process of transmitting and receiving signals of the ultra-wideband radar 10 is performed at intervals.
  • the pulsed ultra-wideband radar 10 in the embodiment of the present application can avoid the transmission signal Leakage causes interference to the echo signal received by the receiver.
  • the echo signal may be further digitally sampled to obtain a radar frame, and subsequent data processing processes are all analyzed and processed based on the radar frame obtained by the sampling.
  • the embodiment of the present application does not limit the number of ultra-wideband radars 10, and optionally, the number of ultra-wideband radars 10 is one or more.
  • the specific quantity of the ultra-wideband radar 10 can be determined in combination with the space size and spatial distribution of the environment-assisted living system, and the cost of the ultra-wideband radar 10.
  • the number of ultra-wideband radars 10 can be set to four. That is, an ultra-wideband radar 10 is installed in each space in the environmental assisted living system.
  • the embodiment of the application does not limit the location of the ultra-wideband radar 10.
  • the ultra-wideband radar 10 is installed in the corner of each space in the environmental assisted living system; or, the ultra-wideband radar 10 is installed in the environmental assisted living system.
  • the center of each space compared to setting in the center, setting in the corner can avoid obstructing the movement of the target object 30, the arrangement of items in various spaces, and the like.
  • the environmental assisted living system includes a home environment
  • the home environment includes three spaces, namely a bedroom, a living room, and a bathroom
  • an ultra-wideband radar 10 is distributed in each space
  • the ultra-wideband radar 10 can be set in the corner of the space in the horizontal direction, and is directly facing the active area of the target object 30 in the space, so as to avoid obstruction while grasping the overall situation of the space.
  • the movement of the target object 30; the ultra-wideband radar 10 can be set to half the height of the target object 30 in the vertical direction to reduce the number of scattering points extracted in the subsequent processing and reduce the processing overhead of the computer equipment.
  • the ultra-wideband radar 10 is set to half of the highest height of the target object 30 in the vertical direction; alternatively, it is set to half of the average height of the target object 30.
  • Computer equipment refers to equipment with data analysis and processing capabilities.
  • the computer equipment can be further refined into a maintainer terminal 22, a server 24, and a user terminal 26.
  • the maintainer terminal 22 and the user terminal 26 may be terminal devices such as mobile phones, tablet computers, embedded terminals, wearable devices, etc.;
  • the server 24 may be one server or a server cluster composed of multiple servers.
  • the maintainer terminal 22 also has the ability to control the ultra-wideband radar 10.
  • the maintainer terminal 22 can control the ultra-wideband radar 10 to transmit pulse periodic signals intermittently, and receive radar frames sampled by the ultra-wideband radar 10, so as to further send the radar frames to the server 24, and the server 24 Carry out the subsequent data processing process.
  • the user sets the period of the transmission signal of the ultra-wideband radar 10 through the maintainer terminal 22 to achieve the purpose of controlling the ultra-wideband radar 10 to transmit pulse periodic signals intermittently.
  • the maintainer terminal 22 and the ultra-wideband radar 10 are arranged in the same space to facilitate the user's control of the ultra-wideband radar 10.
  • the embodiment of the application does not limit the number of maintainer terminals 22.
  • the number of maintainer terminals 22 is the same as the number of ultra-wideband radar 10; or, the number of maintainer terminals 22 is one.
  • the assisted living system includes a home environment, and the home environment includes multiple spaces, assuming that each space is provided with an ultra-wideband radar 10, the number of maintainer terminals 22 can be the same as the number of ultra-wideband radars 10, that is Each space is provided with a maintainer terminal 22; the number of maintainer terminals 22 can also be one, that is, the home environment includes only one maintainer terminal 22, and the maintainer terminal 22 controls the ultra-wideband in each space Radar 10.
  • the user terminal 26 refers to a terminal device that uses the target tracking result in the environment-assisted living system, and has data viewing, acquiring, analysis and processing capabilities.
  • the user terminal 26 may obtain the target tracking result from the server 24, so as to display the target tracking result on the user interface for the user to view and further analyze and process.
  • the user terminal 26 and the ultra-wideband radar 10 are arranged in different spaces, and the position of the user terminal 26 can be moved, so as to achieve the purpose of viewing the target tracking results anytime and anywhere.
  • the user terminal 26 may be located outside the home environment, so that the user can pay attention to the target object 30 in the home environment outside the home environment, for example, In the case that the target object 30 is an empty nest elderly, the user terminal 26 held by the empty nest elderly’s children is located outside the home environment of the empty nest elderly, which can facilitate the activities of the children living in other places to the empty nest elderly Pay attention.
  • the embodiment of this application does not limit the number of user terminals 26. In actual applications, the number of user terminals 26 can be determined in combination with the number of users associated with the target object 30. For example, in the case where the environmental assisted living system includes a home environment , Assuming that the target object 30 is two empty nest elderly people, the number of user terminals 26 may be the number of children of empty nest elderly people.
  • the server 24 refers to a device that actually processes the radar frames obtained by sampling the ultra-wideband radar 10.
  • the server 24 may receive radar frames sent by the maintainer terminal 22.
  • the server 24 may receive radar frames from the maintainer terminal 22 when there is a need for data analysis; or, the server 24 may receive radar frames from the maintainer terminal 22 every other time.
  • the radar frame is acquired from the maintainer terminal 22 at a preset time, and by acquiring the radar frame every preset time, the server 24 can analyze and process the radar frame in time, and find the abnormal situation of the target object 30 in time.
  • the server 24 may also obtain relevant parameters of the ultra-wideband radar 10 from the maintainer terminal 22, for example, the period of the transmitted signal and the range resolution.
  • the embodiment of the present application does not limit the location of the server 24.
  • the server 24 and the ultra-wideband radar 10 are arranged in the same space; or the server 24 and the ultra-wideband radar 10 are arranged in different spaces, for example, in When the environment-assisted living system is a home environment, the server 24 can be installed outside the home environment. Since the server 24 is usually large in size, it can avoid occupying the space of the home environment if it is installed outside the home environment.
  • the server 24 can receive data from multiple maintainer terminals 10. For example, in the case that the environmental assisted living system includes multiple home environments, the server 24 can receive data from multiple home environments. The data of the maintainer terminal 10 to be set.
  • the server 24 processes the radar frame to obtain the position information of the target object 30, so as to further determine the movement trajectory of the target object 30 according to the position information.
  • the amount of activity and the degree of disability of the target object 30 can be further obtained.
  • the frequency and walking of the target object in and out of the environmental assisted living system can be further obtained.
  • Speed, time in bed, and time spent in a certain space (such as the bathroom) in the environmental assisted living system which can then determine the changes in the activities of the target object 30, and promptly detect the abnormal activities of the target object 30 and give an early warning (such as The abnormal signal is sent to the user terminal 26), etc.
  • communication between the maintainer terminal 22 and the ultra-wideband radar 10, between the server 24 and the maintainer terminal 22, and between the user terminal 26 and the server 24 can be performed through a network.
  • the network It can be a wired network or a wireless network.
  • FIG. 1 only takes the environmental assisted living system including the home environment and the home environment only includes one space as an example for illustration, but this does not limit the technical solution of the present application.
  • the home environment may also include multiple spaces, all of which should fall within the scope of protection of this application.
  • the ultra-wideband radar plays an important role in the target tracking process.
  • the ultra-wideband radar since the scale of the target object that needs to be paid attention to in the environmental assisted living system is usually large, so in the processing process, the number of scattered points corresponding to the extracted target object is large.
  • the distance unit where each scattering point is located is used as the measurement, and the number of measurements that the computer device needs to process is also large. Excessive measurement brings more space storage and processing overhead to the computer device. Great pressure.
  • the environmental assisted living system usually includes other objects. These objects are usually in a static state.
  • the environmental assisted living system when the environmental assisted living system is a home environment, the home environment It also includes furniture 40 (such as refrigerators, air conditioners, wardrobes, etc.), these stationary objects will cause greater clutter interference to the echo signal of the ultra-wideband radar, and affect the detection of target objects.
  • environment-assisted living usually includes multiple target objects, and the motion trajectories of these multiple target objects may also overlap. In this case, the measurement and the target object obtained by the corresponding correlation processing of the computer equipment Challenge.
  • the embodiments of the present application provide a target tracking method based on ultra-wideband radar, which can be used to solve the above technical problems.
  • the technical solution of the present application will be introduced and explained through several exemplary embodiments.
  • FIG. 2 shows a flowchart of an ultra-wideband radar-based target tracking method provided by an embodiment of the present application.
  • the method can be applied to the system architecture shown in FIG. 1, for example, as shown in FIG.
  • the server 24 The method can include the following steps (210-240):
  • Step 210 Obtain a radar frame obtained by sampling the echo signal by the ultra-wideband radar.
  • the user can control the ultra-wideband radar transmission signal through a terminal device (such as the maintainer terminal 22 shown in FIG. 1 above).
  • the system of the ultra-wideband radar is a pulse system
  • the transmitted signal can be a periodic pulse signal.
  • the embodiment of the present application adopts the ultra-wideband radar of the pulse system to prevent the leakage of the transmitted signal from causing interference to the echo signal received by the receiver. .
  • the working principle of the ultra-wideband radar of the pulse system please refer to the above-mentioned embodiment, which will not be repeated here. The following shows the expression of the transmitted signal provided by an embodiment of the present application:
  • Ultra-wideband radar can collect echo signals at the interval of transmitting signals.
  • the echo signal is the signal reflected back after the transmitted signal of the ultra-wideband radar encounters the reflecting object.
  • the ultra-wideband radar also periodically collects echo signals.
  • the echo signal can be further digitally sampled to obtain at least one radar frame.
  • the frame number of the radar frame represents the slow time, and each radar frame is used to indicate the strength of the reflected signal at different radial distance units in the slow time space corresponding to the radar frame, that is, the sampling result of the entire fast time.
  • a terminal device such as the maintainer terminal 22 shown in FIG. 1 above
  • a server such as the server 24 shown in FIG. 1 above
  • the radar frame is stored in the form of a baseband signal, which is not limited in the embodiment of the present application.
  • the center frequency of the radio frequency signal is in the high frequency band and is a real signal, the spectrum utilization rate is low, while the baseband signal is obtained by reducing the frequency of the radio frequency signal.
  • the center frequency of the baseband signal is at zero frequency and is complex.
  • the signal and spectrum utilization rate is relatively high. Therefore, the embodiment of the present application can store the radar frame in the form of baseband signal, so as to reduce the data storage overhead on the premise of ensuring the target detection performance requirement.
  • the above-mentioned reflective object refers to an object that can hinder the propagation of a signal and reflect the signal.
  • the reflective object hinders the propagation of the transmission signal of the ultra-wideband radar, and reflects the echo signal against the transmission signal.
  • the embodiments of this application do not limit the types of reflective objects.
  • the environmental assisted living system includes a home indoor environment
  • the reflective objects Including at least one of the following: animals, robots, home appliances, and furniture.
  • the home appliances 40 in the environmental assisted living system are reflective objects; when the environmental assisted living system includes home outdoor environments, the reflective objects include At least one of the following: animals, vehicles, plants.
  • the server that analyzes and processes the radar frame can obtain the radar frame sampled by the ultra-wideband radar.
  • the server directly obtains the radar frame sampled by the ultra-wideband radar from the ultra-wideband radar; or the server obtains the radar frame from the ultra-wideband radar connection
  • the terminal device obtains the radar frame obtained by sampling the ultra-wideband radar, which is not limited in the embodiment of the present application.
  • the manner in which the server obtains the radar frame can be determined in combination with the system architecture design of the environmental assisted living system.
  • Step 220 Extract the scattering points based on the radar frame to obtain a measurement corresponding to the radar frame.
  • the measurement refers to the distance unit where the scattering point is located, and the measurement corresponds to the scattering point one-to-one.
  • the server can extract the scattering points based on the radar frame.
  • the server may further filter the radar frame after acquiring the radar frame, and then extract the scattering points based on the filtered radar frame.
  • the server may use the CLEAN algorithm to extract the scattering points.
  • the above step 220 includes: obtaining a waveform template, the signal form of the waveform template is the same as the signal form of the radar frame, and the range resolution of the waveform template is the same as the range resolution of the radar frame; and the waveform template is used to extract the radar frame Of scattering points. That is, the server may use a waveform template with the same signal form and range resolution as the radar frame to extract the scattering points of the radar frame. As shown in Figure 3, it shows a flow chart of the scatter point extraction process provided by an embodiment of the present application.
  • the signal form of the waveform template in the flow chart is the same as the signal form of the radar frame, and the distance resolution of the waveform template
  • the rate is also the same as the waveform resolution of the radar frame.
  • the server can extract the scatter points corresponding to the radar frame by performing the scatter point extraction process shown in FIG. 3.
  • the distance unit where the scatter point is located is called a measurement.
  • the server extracts the scatter point corresponding to the radar frame, the measurement corresponding to the radar frame can be obtained. Since the measurement is the distance unit where the scattering point is located, there is a one-to-one correspondence between the scattering point and the measurement.
  • Step 230 Perform aggregation processing on the measurement corresponding to the radar frame to obtain the aggregated measurement corresponding to the radar frame.
  • the number of scatter points extracted by the server may be different, but in general, for each radar frame, the number of scatter points corresponding to the extracted radar frame is concentrated in 1 to 3, and then for each radar frame. For a radar frame, the number of measurements corresponding to the radar frame is also concentrated in 1 to 3.
  • an embodiment of the present application proposes to perform aggregation processing after obtaining the measurement corresponding to the radar frame to obtain the aggregate measurement corresponding to the radar frame. The following describes the process of the polymerization process.
  • the measurements corresponding to the radar frame are arranged in order of size, and the number of measurements corresponding to the radar frame is n, where n is a positive integer; the foregoing step 230 includes: The i-th measurement, to determine whether there are other measurements in the search range corresponding to the i-th measurement, other measurements are greater than the i-th measurement, i is a positive integer less than n; other measurements are included in the search range In this case, the ith measurement and other measurements are averaged to obtain the average measurement.
  • the aggregate measurement corresponding to the radar frame includes the average measurement.
  • the server may sort the measurements corresponding to the radar frame, so that the measurements corresponding to the radar frame are arranged in order of size, which facilitates the subsequent aggregation process.
  • the embodiment of the application does not limit the sorting method.
  • the server sorts the measurements corresponding to the radar frames in ascending order; or, sorts the measurements corresponding to the radar frames in ascending order .
  • the server determines whether to perform aggregation processing in sequence according to the order of the measurements corresponding to the radar frames. For example, if the measurements corresponding to the radar frames are sorted in ascending order, the server starts with the smallest measurement. Determine in sequence whether to perform aggregation processing; in the case where the measurements corresponding to the radar frame are arranged in descending order, the server sequentially determines whether to perform aggregation processing starting from the largest measurement. Exemplarily, for the i-th measurement, the server may determine the search range corresponding to the i-th measurement, and further determine whether other measurements are included in the search range. In the case where other measurements are included, the server determines the need Perform polymerization treatment.
  • the above search range includes the front boundary and the back boundary, the front boundary is the i-th measurement, the back boundary is larger than the front boundary, and the back boundary and the front boundary are separated by r distance units, and r is a positive integer.
  • the embodiment of this application does not limit the specific value of r. In practical applications, the value of r can be determined in combination with the requirements for the degree of aggregation. In the case of fewer measurements, the value of r can be set to a larger value.
  • the server averages the i-th measurement and other measurements included in the corresponding search range to obtain an average measurement, which is included in the aggregation measurement corresponding to the radar frame Measure. It should be noted that after understanding the technical solutions of the present application, those skilled in the art will easily think of other aggregation processing methods, for example, performing other measurements included in the i-th measurement and its corresponding search range Weighted average processing, etc., should all fall within the scope of protection of this application.
  • the search range corresponding to the i-th measurement may not include other measurements.
  • the embodiment of this application also The corresponding processing method is proposed.
  • the above step 230 includes: if no other measurement is included in the search range, retaining the i-th measurement, and the aggregated measurement corresponding to the radar frame includes the i-th measurement.
  • FIG. 4 shows a comparison diagram before and after measurement aggregation corresponding to a radar frame provided by an embodiment of the present application.
  • Figure 4(a) shows the radar frame in the form of baseband signal;
  • Figure 4(b) shows the measurement corresponding to the radar frame obtained by extracting the scattering points based on the radar frame shown in Figure 4(a);
  • Figure 4( c) shows the aggregated measurement corresponding to the radar frame obtained by performing aggregation processing on the measurement shown in Figure 4(b).
  • Step 240 Determine the position information of at least one target object according to the aggregated measurement corresponding to the radar frame.
  • the server may determine the location information of at least one target object in the environmental assisted living system based on the aggregated measurement corresponding to the radar frame obtained in step 230.
  • the embodiment of this application does not limit the type of the target object. In practical applications, the type of the target object is determined based on the object concerned by the environmental assisted living system. For example, when the environmental assisted living system pays attention to the empty nest elderly, the target object is the empty nest Elderly; in the case that the environmental assisted living system pays attention to pets raised, the target object is pets.
  • the embodiment of the present application does not limit the expression form of the position information.
  • the position information is expressed in the form of two-dimensional coordinates.
  • the number of target objects corresponding to the radar frame obtained by the server is equal to the number of target objects corresponding to the position information.
  • the specific determination process of the location information please refer to the following embodiments, which will not be repeated here.
  • the above method further includes: for the j-th object in the target object, according to the sequence of the x radar frames in the time domain, concatenating the position information of the j-th object to obtain the motion trajectory of the j-th object , X is an integer greater than 1, and j is a positive integer.
  • the server can further determine the movement trajectory of the target object according to the position information of the target object in each radar frame, so as to determine the amount of activity and/or disability of the target object. analyze.
  • the serial server can pass the x radar frames according to the sequence of the x radar frames in the time domain Determine the position information, and then obtain the motion trajectory of the j-th object.
  • the embodiment of this application does not limit the specific value of x.
  • the value of x can be determined in combination with the analysis needs of the user. For example, when the user needs to analyze the movement trajectory of the target object in a day, the value of x can be set The value is equal to the number of radar frames obtained by the server in a day.
  • the technical solution provided by the embodiments of the present application obtains radar frames obtained by sampling echo signals by ultra-wideband radar, extracts scattering points based on the radar frames, and uses the distance unit where the scattering points are located as a measurement to Obtain the measurement corresponding to the radar frame, and then aggregate the measurement to obtain the aggregated measurement to reduce the number of measurements and increase the processing speed of the computer equipment, and then further determine the location information of the target object based on the aggregated measurement.
  • the technical solution provided by the embodiments of the present application reduces the number of measurements due to the aggregation processing of the measurements, thereby effectively alleviating the storage pressure of the computer equipment, reducing the computing overhead of the computer equipment, and avoiding waste of resources.
  • the embodiment of the present application applies the ultra-wideband radar to the detection and tracking of the target object, which is sufficient Taking into account the user's usage habits, it helps to protect the user's privacy.
  • the storage and processing of radar frames uses radar frames in the form of baseband signals, because the radar frames in the form of baseband signals are obtained by reducing the frequency of the radar frames in the form of radio frequency signals. Yes, and it is a complex signal with high spectrum utilization, which can ensure target detection performance while reducing the storage overhead of computer equipment.
  • the position information of the target object is connected in series according to the sequence of the radar frame in the time domain to obtain the motion trajectory of the target object.
  • the trajectory can be used for subsequent analysis and processing of the activity amount and disability degree of the target object. Therefore, the technical solution provided by the embodiment of the present application expands the application scenarios of the target tracking method and improves the application potential of the target tracking method.
  • the embodiment of the application proposes Before processing the radar frame corresponding to the echo signal, filter out the clutter signal in the radar frame, as shown below.
  • step 220 the following steps are further included:
  • the ultra-wideband radar since the ultra-wideband radar periodically transmits pulse signals and collects echo signals at intervals of transmitting pulse signals, the ultra-wideband radar collects echo signals periodically.
  • the terminal device that controls the ultra-wideband radar can record the time stamp of the radar frame while acquiring the radar frame, and then each radar frame obtained by the server from the terminal device also contains the time stamp.
  • the server can distinguish the sequence of each radar frame in the time domain, etc.
  • the server may further obtain m continuous radar frames in the time domain, and the m radar frames correspond to different echo signals.
  • the embodiment of the present application does not limit the value of m.
  • the sending and receiving frame rate of the ultra-wideband radar is usually set to more than 100 frames.
  • target tracking is also Set to the same frame rate, that is, more than 100 frames, but in practical applications, the hardware frame rate is set to 5 to 10 to meet the requirements of target tracking. Therefore, in the embodiment of this application, the value range of m can be 10 To 20.
  • the server can form a two-dimensional data matrix based on the m radar frames.
  • the number of rows in the two-dimensional data matrix is used to indicate the total number of range units contained in each radar frame; the number of columns in the two-dimensional data matrix is used to indicate the number of m radar frames, that is, the number of columns is m;
  • the element in the x-th row and the y-th column in the two-dimensional data matrix is used to indicate the echo intensity of the y-th radar frame at the x-th range unit, y is a positive integer less than or equal to m, and x is a positive integer, And x is less than or equal to the total number of distance units contained in each radar frame.
  • the size of this element is related to the power of the periodic pulses emitted by the ultra-wideband radar, target azimuth, control loss, and sampling settings at the receiving end, etc.
  • the server can further decompose the two-dimensional data matrix to extract singular values and singular vectors.
  • the server can use the SVD (Singular Value Decomposition) algorithm to extract the data corresponding to the two-dimensional data matrix. Singular values and singular vectors. Assuming that the total number of distance units contained in each radar frame is z, the size of the two-dimensional data matrix R is (m ⁇ z), and for the two-dimensional data matrix R, the server decomposes to obtain:
  • the server can determine the size of each singular value in the singular vector.
  • the server can sort the singular values in order to facilitate subsequent selection of singular values based on the size.
  • the environmental assisted living system includes a home environment
  • the home environment In addition to the target object, it also includes stationary reflective objects such as furniture and home appliances.
  • the signal reflected by the reflection object other than the target object is called the clutter signal. Since the energy of the clutter signal is usually higher than the signal reflected by the target object, the clutter signal also has a larger singular value.
  • the server can set the largest k singular values in the singular vector to zero.
  • the embodiment of the present application does not limit the specific value of k.
  • the value of k can be determined in combination with the singular value corresponding to the clutter signal, the singular value corresponding to the target object, and the number of other reflective objects.
  • k is 1 or 2, so that while filtering out the clutter, it avoids filtering out the signal reflected by the target object to ensure the accuracy of target detection.
  • the filtered radar frame is obtained, and the filtered radar frame is used to extract the scattering points to obtain the measurement corresponding to the radar frame.
  • the processed two-dimensional data matrix can be obtained. Based on the processed two-dimensional data matrix, the server can further obtain the filtered radar frame, which is It is used to extract the scattered points later to obtain the measurement corresponding to the radar frame.
  • FIG. 5 shows a waveform diagram of different forms of radar frames provided by an embodiment of the present application.
  • Figure 5 (a) refers to a radar frame in the form of radio frequency signals
  • Figure 5 (b) refers to a radar frame in the form of baseband signals
  • Figure 5 (c) refers to a radar frame in the form of baseband signals after filtering clutter.
  • Figure 5 compared to radar frames in the form of radio frequency signals, a radar frame in the form of a baseband signal has a lower center frequency and higher spectrum utilization.
  • the useful signal in the radar frame subjected to clutter filtering is more prominent.
  • the technical solution provided by the embodiments of the present application forms a two-dimensional data matrix by combining multiple consecutive radar frames in the time domain, and sets the largest singular values among the singular vectors obtained by decomposing the two-dimensional data matrix.
  • the largest singular values can be set to zero to achieve filtering
  • the purpose of the clutter signal reflected by the static reflection object is to improve the accuracy of target detection.
  • the embodiment of the present application further proposes that the largest one or two singular values are set to zero, so as to avoid filtering out the signal reflected by the target object, and avoid affecting the accuracy of target detection.
  • the following describes the process of determining the location information of the target object.
  • step 240 includes the following steps:
  • the server may filter the aggregated measurement corresponding to the radar frame to obtain the filtered measurement, and further determine the location information of the target object based on the filtered measurement.
  • the embodiment of the present application does not limit the specific filtering manner.
  • the filtering manner is Kalman filtering, or Wiener filtering, or particle filtering.
  • the server can further select effective measurements from them to improve the accuracy of the target detection results. Since the server filters the aggregated measurement corresponding to the radar frame to obtain the filtered measurement, the number of target objects and the filtered measurement corresponding to each target object can be determined. Furthermore, in the embodiment of the present application, the server selects valid The measurement process is performed for each target object and the filtered measurement corresponding to the target object. The selection process of effective measurement is introduced below.
  • the above step (2) includes: for the j-th object in the target object, obtaining the maximum moving speed of the j-th object and the range resolution of the ultra-wideband radar; according to the maximum moving speed of the j-th object And the distance resolution, set the distance gate corresponding to the j-th object; determine the measurement within the distance gate corresponding to the j-th object in the filtered measurement as a valid measurement.
  • the target object in the environmental assisted living system usually has the characteristics of slow walking and stable walking process
  • a corresponding distance gate can be set, and the distance gate is a constant.
  • the range gate of the j-th object is determined according to the maximum moving speed of the j-th object and the range resolution of the ultra-wideband radar. For example, suppose the maximum moving speed of the j-th object is 1.5 m/s and exceeds The range resolution of the wideband radar is 0.05m, and the time interval between two adjacent frames is 0.1s, then the theoretical value of the range gate corresponding to the j-th object is 1.5*0.1/0.05, which is 3. It should be noted that, considering that the target object is a human body, the human body has multi-point scattering characteristics. Therefore, in order to improve the robustness, the distance gate can be set larger than the theoretical value.
  • the server After the server determines the distance gate corresponding to the jth object, it can select the effective measurement corresponding to the jth object according to the distance gate. Exemplarily, the server compares the measurement corresponding to the j-th object in the filtered measurement with the distance gate of the j-th object, and the measurement located in the distance gate of the j-th object is determined to be a valid measurement .
  • the embodiment of the present application also proposes a corresponding processing method.
  • the above method further includes: determining the filtered measurement as the position information of the j-th object when there is no measurement located within the distance gate corresponding to the j-th object in the filtered measurement.
  • the server compares the measurement corresponding to the j-th object in the filtered measurement of a certain radar frame with the distance gate corresponding to the j-th object, if it is not determined that the j-th object is located For the measurement in the corresponding range gate, the filtered measurement of the radar frame is directly determined as the position information of the j-th object corresponding to the radar frame.
  • the server may further construct a confirmation matrix based on the effective measurement, and determine the measurement value based on the confirmation matrix to further determine the location information of the target object.
  • the value of the element in the p-th row and the q-th column of the confirmation matrix is 0 or 1.
  • the element is used to indicate that the p-th effective measurement is not located in the p-th column.
  • this element is used to indicate that the p-th effective measurement is located within the distance gate corresponding to the q-th target object.
  • the joint event is used to indicate the corresponding relationship between the effective measurement and the target object.
  • JPDA Joint Probabilistic Data Association
  • the server can split the confirmation matrix according to the target criterion to obtain at least one joint event.
  • the joint event is used to indicate the corresponding relationship between the effective measurement and the target object. For example, suppose there are two target objects, namely target 1 and target 2, and also assume that there are 3 effective measurements, which are respectively the measurement 1. Measurement 2 and Measurement 3, then measurement 1 can be discarded, and measurement 2 is assigned to target 1, and measurement 3 is assigned to target 2. The above is to assign these three effective measurements to these two targets A possible situation of the object, that is, the above is a joint event.
  • the target criterion includes a single source criterion and a single measurement criterion.
  • the single source criterion means that the object corresponding to any valid measurement is unique, that is, within a certain radar frame, a valid measurement belongs to only one target object, thus confirming that there is only one "1" in each row of the matrix;
  • the single measurement criterion means that the effective measurement corresponding to any object in the radar frame is unique, that is, in a certain radar frame, a target object can only correspond to one effective measurement, thus confirming that each column of the matrix has at most one "1".
  • the server can further calculate the interconnection probability of the joint event.
  • the expression for the interconnection probability of the joint event ⁇ i (k) is shown below:
  • c represents the normalization constant
  • V represents the volume of the correlation gate
  • P d represents the detection probability
  • m k represents the number of effective measurements
  • T represents the number of target objects
  • N jt [z i (k)] is defined by Given; ⁇ t ( ⁇ i (k)) and ⁇ j ( ⁇ i (k)) both represent binary indicator variables; ⁇ ( ⁇ i (k)) is given by Is given; v j (k) is given by Given that z j (k) is called an effective measurement, Indicates the predicted position of the target t in the current frame, the predicted position is obtained by the server based on the aggregation measurement filtering corresponding to the radar frame, and S t (k) represents the actual error covariance matrix of the t-th target object.
  • the interconnection probability between the effective measurement j and the target object t can be obtained as:
  • the actual state of the t-th target object is updated to obtain:
  • the calculation of the weighted measurement of the j-th target object can refer to the weighted measurement of the target object t
  • the calculation of, I won’t go into details here.
  • the server may further determine the position information of the j-th object based on the weighted measurement of the j-th object.
  • the technical solutions provided by the embodiments of the present application perform filtering processing on aggregated measurements, and then select effective measurements based on the filtered measurements.
  • the radar frame does not contain valid measurements.
  • the filtered measurement is directly used as the position information of the target object. Since the filtered measurement further reduces the error of the aggregate measurement, the filtered measurement is determined as Position information can improve the accuracy of target tracking; when the radar frame contains effective measurements, the weighted measurement is further determined based on the effective measurement, and the position information of the target object is determined based on the weighted measurement. Based on the subsequent measurement, the effective measurement is further determined, and the position information is determined based on the weighted measurement, which can further improve the accuracy of the position information and improve the performance of target tracking.
  • object 1 and object 2 there are two target objects in the environmental assisted living system, namely object 1 and object 2.
  • object 1 starts from the target position and uses the ultra-wideband radar as the terminal, and approaches directly Walk in the direction of the ultra-wideband radar
  • object 2 uses the ultra-wideband radar as the starting point and the target position as the terminal, and walks straight away from the ultra-wideband direction.
  • the starting point and the ending point are constantly exchanged.
  • the target location refers to a position directly in front of the ultra-wideband radar and a certain distance away from the ultra-wideband radar.
  • the frame rate of the ultra-wideband radar is set to 150 frames, that is, the time interval between adjacent radar frames is 1/150 second .
  • the server first obtains the radar frame according to the above settings, and the radar frame is stored in the form of a baseband signal. Due to the slow walking speed of the target object, it is not necessary to use an excessively high frame rate for the process of filtering clutter signals. Therefore, the server arranges every 15 radar frames into a two-dimensional data matrix according to the time sequence, and adopts SVD The algorithm filters out the clutter and takes the average as a frame. Through this operation, the frame rate is reduced to 10 frames per second. Then, for each radar frame, the server uses the CLEAN algorithm to extract the measurement.
  • Table 1 shows the detection comparison of the radar frame in the form of a radio frequency signal and the radar frame in the form of a baseband signal as the input of the CLEAN algorithm.
  • To Radar frames in the form of radio frequency signals Radar frame in the form of baseband signal Average processing time per frame/s 0.038 0.023 Measurement of real target/piece 3-10 1-3
  • the average processing time per frame of the radar frame in the form of baseband signal is shortened by 39.7% compared with the radar frame in the form of radio frequency signal, and the processing speed has been significantly improved.
  • the use of baseband signals in the form of radar frames can significantly reduce the actual target measurement obtained.
  • the range extension target further tends to point targets, which reduces the subsequent processing Confirming the dimensions of the matrix at the intersection of the trajectory of the target object also reduces the number of intersecting wave gates to prevent combinatorial explosions.
  • the range resolution of the radar frame in the form of baseband signal is set to eight times that of the radar frame in the form of radio frequency signal
  • the storage space of the radar frame in the form of baseband signal is compared with that of the radar frame in the form of radio frequency signal. Frame, reduced by one-eighth.
  • Fig. 7 shows the tracking filtering result of the target object obtained through the above-mentioned setting and calculation process.
  • the root mean square error of the tracking filtering result shown in FIG. 7 is calculated by the following formula:
  • z(i) represents the true value of the position of the target object
  • each step of the above method can be executed by the server, or part of it can be executed by the server, and the other part can be executed by the server.
  • the terminal device executes, for example, the step of aggregating the measurements corresponding to the radar frame in the above method steps can be executed by the terminal device (the maintainer terminal 22 in the system architecture shown in FIG. 1), and the other steps are executed by the server .
  • How to implement each step of the foregoing method can be determined in combination with application requirements, system architecture design, etc. The embodiment of the present application does not limit this, but these methods should all fall within the protection scope of the present application.
  • FIG. 8 shows a block diagram of an ultra-wideband radar-based target tracking device provided by an embodiment of the present application.
  • the device has the function of realizing the above method example, and the function can be realized by hardware, or by hardware executing corresponding software.
  • the device may be the server 24 in the system architecture shown in FIG. 1, or may be set in the server 24 in the system architecture shown in FIG. 1.
  • the device 800 may include: an information acquisition module 810, a measurement determination module 820, an aggregation processing module 830, and a location determination module 840.
  • the information acquisition module 810 is configured to acquire radar frames obtained by sampling the echo signal by the ultra-wideband radar.
  • the measurement determination module 820 is configured to extract a scattering point based on the radar frame to obtain a measurement corresponding to the radar frame.
  • the measurement refers to the distance unit where the scattering point is located, and the measurement is consistent with There is a one-to-one correspondence between the scattering points.
  • the aggregation processing module 830 is configured to perform aggregation processing on the measurement corresponding to the radar frame to obtain the aggregate measurement corresponding to the radar frame.
  • the position determining module 840 is configured to determine the position information of at least one target object according to the aggregated measurement corresponding to the radar frame.
  • the measurements corresponding to the radar frame are arranged in order of size, and the number of measurements corresponding to the radar frame is n, where n is a positive integer; the aggregation processing module 830 is configured to: For the i-th measurement among the n measurements, determine whether the search range corresponding to the i-th measurement includes other measurements, and the other measurements are greater than the i-th measurement, the i is a positive integer less than the n; if the other measurement is included in the search range, the i-th measurement and the other measurement are averaged to obtain the average measurement, so The aggregate measurement corresponding to the radar frame includes the average measurement.
  • the aggregation processing module 830 is further configured to: if the other measurement is not included in the search range, retain the i-th measurement, and the aggregation amount corresponding to the radar frame The measurement includes the i-th measurement.
  • the search range includes a front boundary and a back boundary
  • the front boundary is the i-th measurement
  • the back boundary is larger than the front boundary
  • the back boundary and the front boundary are different from each other.
  • the device 800 further includes: a radar frame acquisition module 850, configured to acquire m consecutive radar frames in the time domain; and a matrix determination module 860, configured according to the m
  • the radar frame obtains a two-dimensional data matrix
  • the matrix processing module 870 is used to zero the largest k singular values in the singular vectors obtained by decomposing the two-dimensional data matrix to obtain a processed two-dimensional data matrix, where k is A positive integer
  • the radar frame determination module 880 is used to obtain a filtered radar frame based on the processed two-dimensional data matrix, and the filtered radar frame is used to extract scattering points to obtain the measurement corresponding to the radar frame
  • the element in the x-th row and the y-th column in the two-dimensional data matrix is used to indicate the echo intensity of the y-th radar frame at the x-th distance unit, and the y is a positive value less than or equal to the m
  • the x is a positive integer
  • the x is less than or equal
  • the k is 1 or 2.
  • the value of m ranges from 10 to 20.
  • the measurement determination module 820 is configured to: obtain a waveform template, the signal form of the waveform template is the same as the signal form of the radar frame, and the range resolution of the waveform template is the same as that of the radar frame.
  • the range resolution of the frames is the same; the waveform template is used to extract the scattering points in the radar frame.
  • the position determination module 840 includes: a filtering processing unit 841, configured to perform filtering processing on the aggregated measurement corresponding to the radar frame to obtain a filtered measurement; measurement selection The unit 842 is used to select effective measurements from the filtered measurements; the matrix construction unit 843 is used to construct a confirmation matrix based on the effective measurements; the event determination unit 844 is used to split the confirmation according to the target criterion Matrix to obtain at least one joint event, the joint event is used to indicate the corresponding relationship between the effective measurement and the target object; the measurement weighting unit 845 is used for the jth object in the target object , Based on the interconnection probability of the joint event, determine the weighted measurement of the j-th object, where j is a positive integer; the information determining unit 846 is configured to determine the weighted measurement of the j-th object based on the weighted measurement of the j-th object The position information of the j-th object; wherein the value of the element in the
  • the target criterion includes a single source criterion and a single measurement criterion.
  • the single source criterion means that any object corresponding to any valid measurement is unique
  • the single measurement criterion means that any object is in the The corresponding effective measurement in the radar frame is unique.
  • the measurement selection unit 842 is configured to: for the jth object in the target object, obtain the maximum movement speed of the jth object, and the super The range resolution of the wideband radar; according to the maximum movement speed of the j-th object and the range resolution, set the range gate corresponding to the j-th object; and set the distance gate corresponding to the j-th object in the filtered measurement.
  • the measurement in the distance gate corresponding to the j objects is determined as the effective measurement.
  • the information determining unit 846 is further configured to: in the case that there is no measurement located within the distance gate corresponding to the j-th object in the filtered measurement , Determining the filtered measurement as the location information of the j-th object.
  • the signal form of the radar frame is a baseband signal form.
  • the system of the ultra-wideband radar is a pulse system.
  • the device 800 further includes: a trajectory determination module 890, configured to, for the j-th object in the target object, follow the sequence of x radar frames in the time domain, The position information of the j-th object is connected in series to obtain the motion trajectory of the j-th object, where x is an integer greater than 1, and j is a positive integer.
  • a trajectory determination module 890 configured to, for the j-th object in the target object, follow the sequence of x radar frames in the time domain, The position information of the j-th object is connected in series to obtain the motion trajectory of the j-th object, where x is an integer greater than 1, and j is a positive integer.
  • the technical solution provided by the embodiments of the present application obtains radar frames obtained by sampling echo signals by ultra-wideband radar, extracts scattering points based on the radar frames, and uses the distance unit where the scattering points are located as a measurement to Obtain the measurement corresponding to the radar frame, and then aggregate the measurement to obtain the aggregate measurement to reduce the number of measurements and increase the processing speed of the server, and then further determine the location information of the target object based on the aggregate measurement.
  • the technical solution provided by the embodiment of the present application reduces the number of measurements due to the aggregation processing of the measurements, thereby effectively alleviating the storage pressure of the server, reducing the computing overhead of the server, and avoiding resource waste.
  • the embodiment of the present application applies the ultra-wideband radar to the detection and tracking of the target object, which is sufficient Taking into account the user's usage habits, it helps to protect the user's privacy.
  • FIG. 10 shows a structural block diagram of a computer device provided by an embodiment of the present application.
  • the computer device in the embodiment of the present application may include one or more of the following components: a processor 1010 and a memory 1020.
  • the processor 1010 may include one or more processing cores.
  • the processor 1010 uses various interfaces and lines to connect various parts of the entire computer device, and executes the computer by running or executing instructions, programs, code sets, or instruction sets stored in the memory 1020, and calling data stored in the memory 1020.
  • the processor 1010 may adopt at least one of digital signal processing (Digital Signal Processing, DSP), Field-Programmable Gate Array (Field-Programmable Gate Array, FPGA), and Programmable Logic Array (Programmable Logic Array, PLA).
  • DSP Digital Signal Processing
  • FPGA Field-Programmable Gate Array
  • PLA Programmable Logic Array
  • the processor 1010 may integrate one or a combination of a central processing unit (CPU) and a modem. Among them, the CPU mainly processes the operating system and application programs, etc.; the modem is used to process wireless communications. It is understandable that the above-mentioned modem may not be integrated into the processor 1010,
  • the processor 1010 implements the methods provided in the foregoing method embodiments when executing the program instructions in the memory 1020.
  • the memory 1020 may include random access memory (Random Access Memory, RAM), and may also include read-only memory (Read-Only Memory, ROM).
  • the memory 1020 includes a non-transitory computer-readable storage medium.
  • the memory 1020 may be used to store instructions, programs, codes, code sets or instruction sets.
  • the memory 1020 may include a storage program area and a storage data area, where the storage program area may store instructions for implementing the operating system, instructions for at least one function, instructions for implementing each of the foregoing method embodiments, etc.; the storage data area It can store data created based on the use of computer equipment, etc.
  • the structure of the above-mentioned computer device is only illustrative.
  • the computer device may include more or fewer components, such as a display screen, etc., which is not limited in this embodiment.
  • FIG. 10 does not constitute a limitation on the computer device, and may include more or fewer components than those shown in the figure, or combine certain components, or adopt different component arrangements.
  • the embodiments of the present application also provide a computer-readable storage medium in which a computer program is stored, and the computer program is used to be executed by a processor of a computer device to implement the above-mentioned ultra-wideband radar-based target tracking method .
  • the embodiment of the present application also provides a chip, the chip includes a programmable logic circuit and/or program instructions, when the chip runs on a computer device, it is used to implement the above-mentioned ultra-wideband radar-based target tracking method.
  • the embodiments of the present application also provide a computer program product, which when the computer program product runs on a computer device, causes the computer device to execute the above-mentioned ultra-wideband radar-based target tracking method.
  • the functions described in the embodiments of the present application may be implemented by hardware, software, firmware, or any combination thereof. When implemented by software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on the computer-readable medium.
  • the computer-readable medium includes a computer storage medium and a communication medium, where the communication medium includes any medium that facilitates the transfer of a computer program from one place to another.
  • the storage medium may be any available medium that can be accessed by a general-purpose or special-purpose computer.

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

本申请公开了一种基于超宽带雷达的目标跟踪方法、装置、设备及存储介质，属于环境辅助生活技术领域。所述方法包括：获取超宽带雷达对回波信号进行采样得到的雷达帧；基于雷达帧提取散射点，得到雷达帧对应的量测，量测是指散射点所在的距离单元，且量测与散射点一一对应；对雷达帧对应的量测进行聚合处理，得到雷达帧对应的聚合量测；根据雷达帧对应的聚合量测，确定至少一个目标对象的位置信息。本申请实施例由于对量测进行了聚合处理，降低了量测的数量，有效缓解了计算机设备的存储压力并降低了计算机设备的计算开销。另外，本申请实施例通过将超宽带雷达应用于目标对象的检测和跟踪，充分考虑了用户的使用习惯，有助于保护用户的隐私。

Description

基于超宽带雷达的目标跟踪方法、装置、设备及存储介质 技术领域
本申请实施例涉及环境辅助生活技术领域，特别涉及一种基于超宽带雷达的目标跟踪方法、装置、设备及存储介质。
背景技术
环境辅助生活(Ambient Assisted Living，AAL)系统是指通过现代化的感应传输装置，将家里的各类仪器共同连通在一个具有扩展性的智能技术平台上，从而构建一个可以即时反映的环境，以对家居者的状态等进行分析、立即做出判断与反映等。
在环境辅助生活系统中，目标跟踪是非常重要的一个环节。通过目标跟踪可以计算得到目标对象的位置和速度，进而结合一定的环境先验知识，可以进行目标对象活动量的分析和失能程度的估计。相关技术中，目标跟踪主要通过摄像头和/或智能穿戴设备实现。摄像头可以采集目标对象的图像和视频信息，智能穿戴设备可以采集目标对象的运动状态、体征状态等信息，摄像头和/或智能穿戴设备采集到对应的信息后，可以将该信息传送给所在平台的计算机设备，以进行进一步地分析和处理。然而，针对通过摄像头进行目标跟踪的方式，通常情况下，摄像头对采集环境的亮度具备一定的要求，且摄像头采集图像或视频信息容易泄露隐私；针对通过智能穿戴设备进行目标跟踪的方式，由于智能穿戴设备具备一定的侵入性，从而很多人没有佩戴智能穿戴设备的习惯。
因此，正是由于相关技术中的目标跟踪存在上述诸多的缺陷，如何在进行目标跟踪的同时，顾及目标对象的使用习惯以及确保目标对象的隐私，还需要进一步地讨论和研究。
发明内容
本申请实施例提供了一种基于超宽带雷达的目标跟踪方法、装置、设备及存储介质。所述技术方案如下：
一方面，本申请实施例提供了一种基于超宽带雷达的目标跟踪方法，所述方法包括：
获取超宽带雷达对回波信号进行采样得到的雷达帧；
基于所述雷达帧提取散射点，得到所述雷达帧对应的量测，所述量测是指所述散射点所在的距离单元，且所述量测与所述散射点一一对应；
对所述雷达帧对应的量测进行聚合处理，得到所述雷达帧对应的聚合量测；
根据所述雷达帧对应的聚合量测，确定至少一个目标对象的位置信息。
另一方面，本申请实施例提供了一种基于超宽带雷达的目标跟踪装置，所述装置包括：
信息获取模块，用于获取超宽带雷达对回波信号进行采样得到的雷达帧；
量测确定模块，用于基于所述雷达帧提取散射点，得到所述雷达帧对应的量测，所述量测是指所述散射点所在的距离单元，且所述量测与所述散射点一一对应；
聚合处理模块，用于对所述雷达帧对应的量测进行聚合处理，得到所述雷达帧对应的聚合量测；
位置确定模块，用于根据所述雷达帧对应的聚合量测，确定至少一个目标对象的位置信息。
再一方面，本申请实施例提供了一种计算机设备，所述计算机设备包括处理器和存储器，所述存储器存储有计算机程序，所述计算机程序由所述处理器加载并执行以实现上述基于超 宽带雷达的目标跟踪方法。
又一方面，本申请实施例提供了一种计算机可读存储介质，所述存储介质中存储有计算机程序，所述计算机程序用于被计算机设备的处理器执行，以实现上述基于超宽带雷达的目标跟踪方法。
还一方面，本申请实施例提供了一种芯片，所述芯片包括可编程逻辑电路和/或程序指令，当所述芯片在计算机设备上运行时，用于实现如上述基于超宽带雷达的目标跟踪方法。
还一方面，本申请实施例提供了一种计算机程序产品，当计算机程序产品在计算机设备上运行时，使得计算机设备执行上述基于超宽带雷达的目标跟踪方法。
本申请实施例提供的技术方案可以包括如下有益效果：
通过获取超宽带雷达对回波信号进行采样得到的雷达帧，并雷达帧提取散射点，将散射点所在的距离单元作为量测，以得到雷达帧对应的量测，然后对量测进行聚合处理，得到聚合量测，以降低量测的数量，提升计算机设备的处理速度，之后根据聚合量测进一步确定目标对象的位置信息。本申请实施例提供的技术方案，由于对量测进行了聚合处理，降低了量测的数量，从而有效缓解了计算机设备的存储压力和降低了计算机设备的计算开销，避免资源浪费。另外，由于超宽带雷达采集回波信号的过程中不需要拍摄图像和视频，且超宽带雷达具备非侵入性的特点，本申请实施例通过将超宽带雷达应用于目标对象的检测和跟踪，充分考虑了用户的使用习惯，有助于保护用户的隐私。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本申请一个实施例提供的系统架构的示意图；
图2是本申请一个实施例提供的基于超宽带雷达的目标跟踪方法的流程图；
图3是本申请一个实施例提供的散射点提取过程的流程图；
图4是本申请一个实施例提供的雷达帧对应的量测聚合前后的对比图；
图5是本申请一个实施例提供的不同形式的雷达帧的波形图；
图6是本申请另一个实施例提供的基于超宽带雷达的目标跟踪方法的流程图；
图7是本申请一个实施例提供的目标跟踪结果的示意图；
图8是本申请一个实施例提供的基于超宽带雷达的目标跟踪装置的框图；
图9是本申请另一个实施例提供的基于超宽带雷达的目标跟踪装置的框图；
图10是本申请一个实施例提供的服务器的结构框图。
具体实施方式
为使本申请的目的、技术方案和优点更加清楚，下面将结合附图对本申请实施方式作进一步地详细描述。
请参考图1，其示出了本申请一个实施例提供的环境辅助生活系统的系统架构。该系统架构包括超宽带雷达10、计算机设备和目标对象30。
目标对象30是指环境辅助生活系统中需要关注的活动对象。可选地，目标对象30为人。示例性地，在该环境辅助生活系统应用于家居环境中对空巢老人的关注的情况下，该目标对象30是指空巢老人。需要说明的一点是，本领域技术人员在了解了本申请的技术方案后，将很容易想到将本申请的技术方案应用于其它系统或其它环境，目标对象的类型也因此发生改变，例如，在本申请的技术方案应用于动物商店、动物医院等中对动物的关注的情况下，该目标对象30是指动物；在本申请技术方案应用于办公环境等中对机器人的关注的情况下，该目标对象30是指机器人。本申请实施例对目标对象30的数量不作限定，示例性地，在该环 境辅助生活系统应用于家居环境中对空巢老人的关注的情况下，目标对象30数量通常为一个或两个。
超宽带(Ultra Wide-Band，UWB)雷达10是指发射信号的分数带宽(Fractional Bandwidth，FBW)大于0.25的雷达，其可以实现通信、探测等功能。本申请实施例中，超宽带雷达10可以对目标对象30进行探测，并且采集回波信号。示例性地，超宽带雷达10的发射信号与目标对象30相遇后，返回对应于该发射信号的回波信号，从而超宽带雷达10可以对回波信号进行采集。可选地，本申请实施例中的超宽带雷达10为脉冲体制的雷达传感器，示例性地，计算机设备中的维护者终端22可以控制超宽带雷达10间歇式地发射脉冲周期信号，并且在发射间隔接收反射的回波信号，也即，超宽带雷达10的收发信号的过程间隔进行，相比于连续波体制的雷达传感器，本申请实施例中脉冲体制的超宽带雷达10可以避免发射信号的泄露对接收机接收回波信号造成干扰。本申请实施例中，超宽带雷达10在采集到回波信号后，可以进一步将该回波信号通过数字采样得到雷达帧，后续的数据处理过程均基于采样得到的雷达帧进行分析处理。
本申请实施例对超宽带雷达10的数量不作限定，可选地，超宽带雷达10的数量为一个或者多个。实际应用中，可以结合该环境辅助生活系统的空间大小、空间分布，以及超宽带雷达10的成本等确定超宽带雷达10的具体数量。示例性地，在该环境辅助生活系统包括家居环境的情况下，假设该家居环境包括四个空间，分别为卧室、卫生间、客厅和厨房，那么可以设置超宽带雷达10的数量为四个，也即，在该环境辅助生活系统中的每一个空间中均设置有一个超宽带雷达10。
本申请实施例对超宽带雷达10的位置也不作限定，可选地，超宽带雷达10设置在该环境辅助生活系统中各个空间的角落；或者，超宽带雷达10设置在该环境辅助生活系统中各个空间的中心。其中，设置在角落相比于设置在中心，可以避免阻碍目标对象30的运动、各个空间的物品布置等。示例性地，在该环境辅助生活系统包括家居环境的情况下，假设该家居环境中包括三个空间，分别为卧室、客厅和卫生间，并且假设各个空间中均分布有超宽带雷达10，则在每个空间中，超宽带雷达10在水平方向上可以设置在该空间的角落，并且正对着目标对象30在该空间的活动区域，以在对该空间的全局情况进行掌握的同时，避免阻碍目标对象30的运动；超宽带雷达10在垂直方向上可以设置为目标对象30高度的一半，以减少后续处理过程中提取到的散射点的个数，降低计算机设备的处理开销。可选地，在目标对象30的数量为多个的情况下，超宽带雷达10在垂直方向上设置为目标对象30的最高高度的一半；或者，设置为目标对象30的平均高度的一半。
计算机设备是指具备数据分析和处理能力的设备。本申请实施例中，计算机设备可以进一步细化为维护者终端22、服务器24和使用者终端26。其中，维护者终端22和使用者终端26可以是诸如手机、平板电脑、嵌入式终端、可穿戴设备等终端设备；服务器24可以是一台服务器，也可以是由多台服务器组成的服务器集群。
维护者终端22除了具备数据处理能力之外，还具备控制超宽带雷达10的能力。本申请实施例中，维护者终端22可以控制超宽带雷达10间歇式地发射脉冲周期信号，并接收超宽带雷达10采样得到的雷达帧，以进一步将该雷达帧发送给服务器24，由服务器24进行后续的数据处理过程。可选地，用户通过维护者终端22设置超宽带雷达10的发射信号的周期，以达到控制超宽带雷达10间歇式地发射脉冲周期信号的目的。可选地，维护者终端22与超宽带雷达10设置在相同的空间中，以便于用户对超宽带雷达10的控制。本申请实施例对维护者终端22的数量不作限定，可选地，维护者终端22的数量与超宽带雷达10的数量相同；或者，维护者终端22的数量为一个，示例性地，在环境辅助生活系统包括家居环境，且该家居环境包括多个空间的情况下，假设每个空间均设置有超宽带雷达10，则维护者终端22的数量可以与超宽带雷达10的数量相同，也即，每个空间均设置有维护者终端22；维护者终端22的数量也可以为一个，也即，该家居环境中仅包括一个维护者终端22，该维护者终端22控制各个空间中的超宽带雷达10。
使用者终端26是指使用该环境辅助生活系统中目标跟踪结果的终端设备，其具备数据查看、获取、分析处理能力。本申请实施例中，使用者终端26可以从服务器24获取目标跟踪结果，以将该目标跟踪结果显示在用户界面便于用户的查看，以及进一步地分析处理。可选地，使用者终端26与超宽带雷达10设置在不同的空间，并且，该使用者终端26的位置可以移动，从而可以达到随时随地查看目标跟踪结果的目的。示例性地，在环境辅助生活系统包括家居环境的情况下，使用者终端26可以位于该家居环境之外，以便于用户在家居环境之外对该家居环境中的目标对象30进行关注，例如，在目标对象30为空巢老人的情况下，空巢老人的子女持有的使用者终端26位于空巢老人生活的家居环境之外，可以便于生活在其他地方的子女对空巢老人的活动状态进行关注。本申请实施例对使用者终端26的数量不作限定，实际应用中可以结合与目标对象30相关联的用户的数量确定使用者终端26的数量，例如，在环境辅助生活系统包括家居环境的情况下，假设目标对象30为两个空巢老人，则使用者终端26的数量可以为空巢老人的子女的数量。
服务器24是指对超宽带雷达10采样得到的雷达帧进行实际处理的设备。本申请实施例中，服务器24可以接收维护者终端22发送的雷达帧，可选地，服务器24在有数据分析需求的情况下，从维护者终端22处接收雷达帧；或者，服务器24每隔预设时间从维护者终端22处获取雷达帧，通过每隔预设时间获取雷达帧，可以使得服务器24及时对雷达帧进行分析处理，以及时发现目标对象30的异常情况等。本申请实施例中，服务器24除了从维护者终端22处获取雷达帧之外，还可以从维护者终端22处获取超宽带雷达10的相关参数，例如，发射信号的周期、距离分辨率等。本申请实施例对服务器24的位置设置不作限定，可选地，服务器24与超宽带雷达10设置在相同的空间中；或者，服务器24与超宽带雷达10设置在不同的空间中，例如，在该环境辅助生活系统为家居环境的情况下，服务器24可以设置在家居环境之外，由于服务器24的体积通常较大，设置在家居环境之外可以避免占用家居环境的空间。本申请实施例中，服务器24可以接收来自于多个维护者终端10的数据，示例性地，在环境辅助生活系统包括多个家居环境的情况下，服务器24可以接收来自于多个家居环境中设置的维护者终端10的数据。
本申请实施例中，服务器24对雷达帧进行处理，可以得到目标对象30的位置信息，从而根据该位置信息进一步确定目标对象30的运动轨迹。通过目标对象30的运动轨迹，可以进一步得到目标对象30的活动量和失能程度等，示例性地，利用目标对象30的运动轨迹，可以进一步得到目标对象进出该环境辅助生活系统的频次、行走速度、在床时间、在该环境辅助生活系统中的某一空间(如卫生间)停留的时间，进而可以判断目标对象30活动的变化，以及时发现目标对象30活动的异常并做出预警(如将异常信号发送给使用者终端26)等。
本申请实施例中，维护者终端22与超宽带雷达10之间、服务器24与维护者终端22之间、使用者终端26与服务器24之间均可以通过网络进行通信，可选地，该网络可以为有线网络，也可以为无线网络。
需要说明的一点是，图1仅以环境辅助生活系统包括家居环境，并且该家居环境仅包括一个空间为例进行举例说明，但这并不对本申请的技术方案构成限定。本领域技术人员在了解了本申请的技术方案后，将很容易想到其他的系统架构，例如，将本申请的技术方案应用于其它环境，如办公环境等；在该环境辅助生活系统为家居环境的情况下，该家居环境也可以包括多个空间，这些均应属于本申请的保护范围之内。
根据上述介绍说明，可见超宽带雷达在目标跟踪过程中发挥着重大的作用。然而，在超宽带雷达应用于环境辅助生活系统中时，由于环境辅助生活系统中需要关注的目标对象的尺度通常较大，这样在处理过程中，提取的目标对象对应的散射点的数量较多，后续处理过程中将每个散射点所在的距离单元作为量测，进而计算机设备需要处理的量测的数量也较多，过多的量测给计算机设备的空间存储和处理开销均带来较大的压力。另外，环境辅助生活系统中除了目标对象之外，通常还包括其它的对象，这些对象通常为静止状态，例如，如图1所示，在环境辅助生活系统为家居环境的情况下，该家居环境中还包括家具40(如冰箱、空 调、衣柜等)，这些静止的对象会给超宽带雷达的回波信号造成较大的杂波干扰，影响目标对象的检测。此外，环境辅助生活中通常包括多个目标对象，且这多个目标对象的运动轨迹之间也可能会存在交叉，这种情况下，给计算机设备对应关联处理得到的量测与目标对象带来了挑战。
基于此，本申请实施例提供了一种基于超宽带雷达的目标跟踪方法，可用于解决上述技术问题。下面，通过几个示例性实施例对本申请技术方案进行介绍说明。
请参考图2，其示出了本申请一个实施例提供的基于超宽带雷达的目标跟踪方法的流程图，该方法可应用于图1所示的系统架构中，例如，应用于图1所示的服务器24中。该方法可以包括如下几个步骤(210～240)：
步骤210，获取超宽带雷达对回波信号进行采样得到的雷达帧。
本申请实施例中，用户可以通过终端设备(如上述图1所示的维护者终端22)控制超宽带雷达发射信号。可选地，超宽带雷达的体制为脉冲体制，进而发射信号可以为周期脉冲信号，本申请实施例通过采用脉冲体制的超宽带雷达，可以避免发射信号的泄露对接收机接收回波信号造成干扰。有关脉冲体制的超宽带雷达的工作原理，请参见上述实施例，此处不多赘述。下面示出了本申请一个实施例提供的发射信号的表达式：
Figure PCTCN2020095889-appb-000001
其中，h(t)是指高斯脉冲信号，且h(t)为基带信号；f c是指载波频率，且2πf c＝ω c；τ是指脉宽因子；f(t)为射频形式的发射信号；t为发射信号的发射时刻。
超宽带雷达在发射信号的间隔，可以采集回波信号。回波信号是超宽带雷达的发射信号遇到反射对象后反射回来的信号。本申请实施例中，对应于周期脉冲形式的发射信号，超宽带雷达也是周期性采集回波信号。在超宽带雷达接收到回波信号后，可以进一步将该回波信号通过数字采样得到至少一个雷达帧。其中，雷达帧的帧数表示慢时间，每一个雷达帧用于指示该雷达帧对应的慢时间空间中不同径向距离单元处反射信号的强弱，也即，整个快时间的采样结果。
本申请实施例中，控制超宽带雷达发射信号和采集信号的终端设备(如上述图1所示的维护者终端22)或者服务器(如上述图1所示的服务器24)可以存储雷达帧。可选地，该雷达帧存储为射频信号形式；或者，该雷达帧存储为基带信号形式，本申请实施例对此不作限定。其中，由于射频信号的中心频率在高频频段，且为实信号，频谱利用率较低，而基带信号是通过降低射频信号的频率得到的，基带信号的中心频率在零频上，且为复信号，频谱利用率较高，因此，本申请实施例可以将雷达帧存储为基带信号形式，以在确保目标检测性能要求的前提下，降低数据的存储开销。
另外，上述反射对象是指能够阻碍信号的传播并反射信号的对象，本申请实施例中，反射对象阻碍了超宽带雷达的发射信号的传播，并且针对该发射信号反射回波信号。本申请实施例对反射对象的类型不作限定，实际应用中，针对不同的环境辅助生活系统，可能会有不同类型的反射对象，例如，在环境辅助生活系统包括家居室内环境的情况下，反射对象包括以下至少一项：动物、机器人、家电、家具，如图1所示，该环境辅助生活系统中的家电40即为反射对象；在环境辅助生活系统包括家居户外环境的情况下，反射对象包括以下至少一项：动物、车辆、植物。
对雷达帧进行分析处理的服务器可以获取超宽带雷达采样得到的雷达帧，可选地，服务器直接从超宽带雷达处获取超宽带雷达采样得到的雷达帧；或者，服务器从与超宽带雷达连接的终端设备处获取超宽带雷达采样得到的雷达帧，本申请实施例对此不作限定，实际应用中，服务器获取雷达帧的方式可以结合环境辅助生活系统的系统架构设计来确定。
步骤220，基于雷达帧提取散射点，得到雷达帧对应的量测，量测是指散射点所在的距离单元，且量测与散射点一一对应。
服务器在获取雷达帧后，可以基于该雷达帧提取散射点。可选地，为了提升目标检测的 准确性，服务器可以在获取雷达帧后，进一步对该雷达帧进行滤波处理，然后基于该滤波后的雷达帧提取散射点。有关服务器对雷达帧进行滤波处理的过程的介绍说明，请参见下述实施例，此处不多赘述。
本申请实施例对服务器提取散射点的方式不作限定，可选地，服务器可以采用CLEAN算法提取散射点。在一个示例中，上述步骤220包括：获取波形模板，波形模板的信号形式和雷达帧的信号形式相同，且波形模板的距离分辨率和雷达帧的距离分辨率相同；采用波形模板提取雷达帧中的散射点。也即，服务器可以采用与雷达帧具有相同的信号形式和距离分辨率的波形模板提取雷达帧的散射点。如图3所示，其示出了本申请一个实施例提供的散射点提取过程的流程图，该流程图中的波形模板的信号形式与雷达帧的信号形式相同，且该波形模板的距离分辨率也与雷达帧的波形分辨率相同。服务器通过执行图3所示的散射点提取过程，即可提取雷达帧对应的散射点。
为了便于说明目标跟踪过程，本申请实施例中，将散射点所在的距离单元称为量测，则服务器在提取到雷达帧对应的散射点之后，即可得到雷达帧对应的量测。由于量测是散射点所在的距离单元，因而散射点与量测之间是一一对应的关系。
步骤230，对雷达帧对应的量测进行聚合处理，得到雷达帧对应的聚合量测。
服务器针对每一个雷达帧，提取到的散射点的数量可能会不同，但通常情况下，针对每一个雷达帧，提取到的雷达帧对应的散射点的数量集中在1至3个，进而针对每一个雷达帧，雷达帧对应的量测的数量也集中在1至3个。为了进一步减少雷达帧对应的量测的数量，本申请实施例提出在得到雷达帧对应的量测之后进行聚合处理，得到雷达帧对应的聚合量测。下面对聚合处理的过程进行介绍说明。
在一种可能的实施方式中，雷达帧对应的量测按照大小呈顺序排列，且雷达帧对应的量测的数量为n，n为正整数；上述步骤230包括：对于n个量测中的第i个量测，确定第i个量测对应的搜索范围内是否包含其它量测，其它量测大于第i个量测，i为小于n的正整数；在搜索范围内包含其它量测的情况下，对第i个量测和其它量测进行平均处理，得到平均量测，雷达帧对应的聚合量测包括平均量测。
服务器在聚合处理之前，可以对雷达帧对应的量测进行排序，以使得雷达帧对应的量测按照大小呈顺序排列，便于后续的聚合处理过程。本申请实施例对排序的方式不作限定，可选地，服务器按照从大至小的顺序对雷达帧对应的量测进行排序；或者，按照从小至大的顺序对雷达帧对应的量测进行排序。
聚合处理过程中，服务器按照雷达帧对应的量测的顺序，依次确定是否进行聚合处理，例如，在雷达帧对应的量测按照从小至大的顺序排序的情况下，服务器从最小的量测开始依次确定是否进行聚合处理；在雷达帧对应的量测按照从大至小的顺序排列的情况下，服务器从最大的量测开始依次确定是否进行聚合处理。示例性地，针对第i个量测，服务器可以确定第i个量测对应的搜索范围，并且，进一步确定该搜索范围内是否包含其它量测，在包含其它量测的情况下，服务器确定需要进行聚合处理。
其中，上述搜索范围包括前边界和后边界，前边界为第i个量测，后边界大于前边界，且后边界与前边界之间相隔r个距离单元，r为正整数。本申请实施例对r的具体取值不作限定，实际应用中可以结合对聚合程度的需求来确定r的取值，例如，在对聚合程度要求较高的情况下，也即，在需求得到更少的量测的情况下，可以将r的取值设置得较大。
在确定需要进行聚合处理的情况下，服务器将第i个量测与其对应的搜索范围内包含的其它量测进行平均处理，得到一个平均量测，雷达帧对应的聚合量测中即包含该平均量测。需要说明的一点是，本领域技术人员在了解了本申请的技术方案后，将很容易想到其它的聚合处理方式，例如，对第i个量测与其对应的搜索范围内包含的其它量测进行加权平均处理等，这些均应属于本申请的保护范围之内。
上述是针对需要进行聚合处理的情况，在实际应用中，第i个量测对应的搜索范围内也可能不包含其它量测，针对搜索范围内不包含其它量测的情况，本申请实施例也提出了相应 的处理方式。在一个示例中，上述步骤230包括：在搜索范围内不包含其它量测的情况下，保留第i个量测，雷达帧对应的聚合量测包括第i个量测。
例如，如图4所示，其示出了本申请一个实施例提供的雷达帧对应的量测聚合前后的对比图。图4(a)表示的是基带信号形式的雷达帧；图4(b)表示的是基于图4(a)所示的雷达帧提取散射点，得到的雷达帧对应的量测；图4(c)表示的是对图4(b)所示的量测进行聚合处理得到的雷达帧对应的聚合量测。
步骤240，根据雷达帧对应的聚合量测，确定至少一个目标对象的位置信息。
服务器可以基于步骤230得到的雷达帧对应的聚合量测，确定该环境辅助生活系统中至少一个目标对象的位置信息。本申请实施例对目标对象的类型不作限定，实际应用中结合该环境辅助生活系统关注的对象确定目标对象的类型，例如，在环境辅助生活系统关注空巢老人的情况下，目标对象为空巢老人；在环境辅助生活系统关注饲养的宠物的情况下，目标对象为宠物。本申请实施例对位置信息的表现形式不作限定，可选地，位置信息表现为二维坐标的形式。通常情况下，服务器获取到的雷达帧对应的目标对象的数量，等于位置信息对应的目标对象的数量。有关位置信息的具体确定过程，请参见下述实施例，此处不多赘述。
在一个示例中，上述方法还包括：对于目标对象中的第j个对象，按照x个雷达帧在时域上的先后顺序，串联第j个对象的位置信息，得到第j个对象的运动轨迹，x为大于1的整数，j为正整数。
在得到目标对象在各个雷达帧的位置信息后，服务器可以进一步根据目标对象在各个雷达帧的位置信息，确定目标对象的运动轨迹，以便于对目标对象的活动量和/或失能程度等进行分析。对于环境辅助生活系统中包括多个目标对象的情况下，针对这多个目标对象中的第j个对象，可以按照x个雷达帧在时域上的先后顺序，串联服务器通过这x个雷达帧确定的位置信息，进而得到第j个对象的运动轨迹。本申请实施例对x的具体取值不作限定，实际应用中可以结合用户的分析需求确定x的取值，例如，在用户需求分析一天中目标对象的运动轨迹的情况下，可以设定x的取值等于服务器在一天的时间里获取的雷达帧的帧数。
综上所述，本申请实施例提供的技术方案，通过获取超宽带雷达对回波信号进行采样得到的雷达帧，并基于雷达帧提取散射点，将散射点所在的距离单元作为量测，以得到雷达帧对应的量测，然后对量测进行聚合处理，得到聚合量测，以降低量测的数量，提升计算机设备的处理速度，之后根据聚合量测进一步确定目标对象的位置信息。本申请实施例提供的技术方案，由于对量测进行了聚合处理，降低了量测的数量，从而有效缓解了计算机设备的存储压力和降低了计算机设备的计算开销，避免资源浪费。另外，由于超宽带雷达采集回波信号的过程中不需要拍摄图像和视频，且超宽带雷达具备非侵入性的特点，本申请实施例通过将超宽带雷达应用于目标对象的检测和跟踪，充分考虑了用户的使用习惯，有助于保护用户的隐私。
并且，本申请实施例提供的技术方案，对雷达帧的存储和处理过程，采用的都是基带信号形式的雷达帧，由于基带信号形式的雷达帧是通过降低射频信号形式的雷达帧的频率得到的，且为复信号，频谱利用率较高，可以确保目标检测性能的同时，降低计算机设备的存储开销。
另外，本申请实施例提供的技术方案，在得到目标对象的位置信息后，进一步按照雷达帧在时域上的先后顺序，串联目标对象的位置信息，以得到目标对象的运动轨迹，由于该运动轨迹可以用于后续对目标对象的活动量和失能程度等进行分析处理，从而，本申请实施例提供的技术方案扩展了目标跟踪方法的应用场景，提升了目标跟踪方法的应用潜力。
由于超宽带雷达采集的回波信号除了包括目标对象对应的回波信号之外，还可能包括其他反射对象对应的回波信号，为了提升服务器检测目标对象的准确性，本申请实施例提出在对回波信号对应的雷达帧进行处理之前，滤除雷达帧中的杂波信号，如下所示。
在一种可能的实施方式中，上述步骤220之前，还包括如下几个步骤：
(1)获取在时域上连续的m个雷达帧。
本申请实施例中，由于超宽带雷达是周期性地发射脉冲信号，并且在发射脉冲信号的间隔采集回波信号，因此，超宽带雷达周期性地采集回波信号。可选地，控制超宽带雷达的终端设备在获取雷达帧的同时，可以记录雷达帧的时间戳，进而服务器从该终端设备处获取的各个雷达帧也包含时间戳，在后续的分析处理过程中，服务器可以区分各个雷达帧在时域上的先后顺序等。
服务器获取雷达帧后可以进一步获取在时域上连续的m个雷达帧，并且，该m个雷达帧对应于不同的回波信号。本申请实施例对m的取值不作限定，可选地，由于采用超宽带雷达进行目标检测时，超宽带雷达的收发帧率通常设置为100帧以上，为了在硬件上保持统一，目标跟踪也设置为同样的帧率，即100帧以上，但在实际应用中，硬件帧率设置为5至10即可满足目标跟踪的要求，因此，本申请实施例中，m的取值范围可以为10至20。
(2)根据m个雷达帧得到二维数据矩阵。
服务器在确定了m个雷达帧之后，可以根据这m个雷达帧组成二维数据矩阵。其中，二维数据矩阵的行数用于指示每个雷达帧包含的距离单元的总个数；二维数据矩阵的列数用于指示m个雷达帧的个数，也即，列数即为m；二维数据矩阵中第x行第y列的元素用于指示第y个雷达帧在第x个距离单元处的回波强度，y为小于或等于m的正整数，x为正整数，且x小于或等于每个雷达帧包含的距离单元的总个数，可选地，该元素的大小与超宽带雷达发射的周期性脉冲的功率、目标方位角、控件损耗以及接收端的采样设置等相关。
(3)将二维数据矩阵分解得到的奇异向量中最大的k个奇异值置零，得到处理后的二维数据矩阵，k为正整数。
在组成二维数据矩阵后，服务器可以进一步分解二维数据矩阵，以提取奇异值和奇异向量，可选地，服务器可以采用SVD(Singular Value Decomposition，奇异值分解)算法提取二维数据矩阵对应的奇异值和奇异向量。假设每个雷达帧包含的距离单元的总个数为z，则二维数据矩阵R的尺寸为(m×z)，进而对于该二维数据矩阵R，服务器分解得到：
Figure PCTCN2020095889-appb-000002
其中，U和V分别是尺寸为(m×m)和(z×z)的酉矩阵(Unitary Matrix，又称为幺正矩阵)；S＝diag(σ 1，σ 2，…，σ r)即为奇异向量，奇异向量中的σ 1、σ 2…σ r即为奇异值，且奇异值满足σ 1≥σ 2≥…≥σ r≥0；u i和v i分别为矩阵U和V的列向量。
在分解得到奇异向量后，服务器可以确定奇异向量中各个奇异值的大小，可选地，服务器可以对奇异值进行大小排序，以便于后续根据大小选取奇异值。由于环境辅助生活系统中除了运动的目标对象之外，还存在其他的反射对象，且这些反射对象通常是静止的，示例性地，在环境辅助生活系统包括家居环境的情况下，家居环境中除了包括目标对象之外，还包括家具、家电等静止的反射对象。本申请实施例中，将目标对象之外的反射对象反射的信号称为杂波信号，由于杂波信号的能量通常高于目标对象反射的信号，因而杂波信号也具备较大的奇异值。为了达到滤除杂波的目的，服务器在确定奇异向量之后，可以将奇异向量中最大的k个奇异值置零。本申请实施例对k的具体取值不作限定，实际应用中，可以结合杂波信号对应的奇异值、目标对象对应的奇异值以及其他反射对象的数量等确定k的取值，可选地，k为1或2，这样滤除杂波的同时，避免将目标对象反射的信号也滤除，确保目标检测的 准确性。
(4)基于处理后的二维数据矩阵，得到滤波后的雷达帧，滤波后的雷达帧用于提取散射点得到雷达帧对应的量测。
服务器在将最大的k个奇异值置零后，可以得到处理后的二维数据矩阵，基于该处理后的二维数据矩阵，服务器可以进一步得到滤波后的雷达帧，该滤波后的雷达帧即用于后续提取散射点，以得到雷达帧对应的量测。
请参考图5，其示出了本申请一个实施例提供的不同形式的雷达帧的波形图。其中，图5(a)是指射频信号形式的雷达帧；图5(b)是指基带信号形式的雷达帧；图5(c)是指滤除杂波后的基带信号形式的雷达帧。从图5中可以看出，基带信号形式的雷达帧相比于射频信号形式的雷达帧，中心频率较低，频谱利用率较高。从图5中还可以看出，进行杂波滤除处理的雷达帧中有用信号更为突出。
综上所述，本申请实施例提供的技术方案，通过将时域上连续的多个雷达帧组成二维数据矩阵，并将二维数据矩阵分解得到的奇异向量中最大的几个奇异值置零，由于在环境辅助生活系统中，静态的反射对象往往具有较高的能力，因而其也具有较大的奇异值，本申请实施例通过将最大的几个奇异值置零，可以达到滤除静态的反射对象反射的杂波信号的目的，提升目标检测的准确性。另外，本申请实施例进一步提出，将最大的一至两个奇异值置零，这样可以避免将目标对象反射的信号也滤除，避免影响目标检测的准确性。
下面对确定目标对象的位置信息的过程进行介绍说明。
在一个示例中，上述步骤240包括如下几个步骤：
(1)对雷达帧对应的聚合量测进行滤波处理，得到滤波后量测。
服务器可以对雷达帧对应的聚合量测进行滤波处理，以得到滤波后量测，并基于该滤波后量测进一步确定目标对象的位置信息。本申请实施例对具体的滤波方式不作限定，可选地，该滤波方式为卡尔曼滤波，或者为维纳滤波，或者为粒子滤波。
(2)从滤波后量测中选取有效量测。
基于滤波后量测，服务器可以进一步从中选取有效量测，以提升目标检测结果的准确性。由于在服务器对雷达帧对应的聚合量测进行滤波得到滤波后量测的过程中，可以确定目标对象的数量，以及各个目标对象对应的滤波后量测，进而本申请实施例中，服务器选取有效量测的过程，是针对每个目标对象以及该目标对象对应的滤波后量测进行的。下面对有效量测的选取过程进行介绍说明。
在一个示例中，上述步骤(2)包括：对于目标对象中的第j个对象，获取第j个对象的最大运动速度，以及超宽带雷达的距离分辨率；根据第j个对象的最大运动速度和距离分辨率，设置第j个对象对应的距离波门；将滤波后量测中位于第j个对象对应的距离波门内的量测确定为有效量测。
由于环境辅助生活系统中的目标对象通常具有行走缓慢、行走过程稳定等特征，针对各个目标对象，可以设置对应的距离波门，且该距离波门为常数。可选地，第j个对象的距离波门根据第j个对象的最大运动速度和超宽带雷达的距离分辨率确定，例如，假设第j个对象的最大运动速度为1.5m/s，且超宽带雷达的距离分辨率为0.05m，相邻两帧的时间间隔为0.1s，则第j个对象对应的距离波门的理论值为1.5*0.1/0.05，也即3。需要说明的一点是，考虑到目标对象为人体的情况下，人体具有多点散射特征，因此，为了提升鲁棒性，可以将距离波门设置得比理论值大。
服务器在确定第j个对象对应的距离波门后，即可以根据该距离波门选取第j个对象对应的有效量测。示例性地，服务器将滤波后量测中第j个对象对应的量测与第j个对象的距离波门相比较，位于第j个对象的距离波门内的量测即确定为有效量测。
针对滤波后量测中不包括位于第j个对象对应的距离波门内的量测的情况，本申请实施例也提出了相应的处理方式。可选地，上述方法还包括：在滤波后量测中没有位于第j个对象对应的距离波门内的量测的情况下，将滤波后量测确定为第j个对象的位置信息。也即， 在服务器将某一雷达帧的滤波后量测中第j个对象对应的量测，与第j个对象对应的距离波门进行比较的过程中，若未确定出位于第j个对象对应的距离波门内的量测，则直接将该雷达帧的滤波后量测确定为第j个对象对应于该雷达帧的位置信息。
(3)基于有效量测构建确认矩阵。
针对服务器从滤波后量测中确定出有效量测的情况，服务器可以进一步基于该有效量测构建确认矩阵，并基于该确认矩阵确定量测取值，以进一步确定目标对象的位置信息。本申请实施例中，确认矩阵的第p行第q列的元素的取值为0或1，在取值为0的情况下，该元素用于指示第p个有效量测没有位于所述第q个目标对象对应的距离波门内；在取值为1的情况下，该元素用于指示第p个有效量测位于第q个目标对象对应的距离波门内。
(4)按照目标准则拆分确认矩阵，得到至少一个联合事件，联合事件用于指示有效量测与目标对象之间的对应关系。
考虑到环境辅助生活系统通常应用于家居环境中，且该家居环境中通常包括两个空巢老人，因此，在这种稀疏的目标对象环境下，可以采用联合概率数据关联(Joint Probabilistic Data Association，JPDA)对量测进行分配和关联，也即，在构建确定矩阵之后，服务器可以按照目标准则拆分该确认矩阵，以得到至少一个联合事件。其中，该联合事件用于指示有效量测与目标对象之间的对应关系，例如，假设目标对象有两个，分别为目标1和目标2，还假设有效量测为3个，分别为量测1、量测2和量测3，那么可以舍弃量测1，并将量测2划分给目标1，量测3划分给目标2，上述是将这三个有效量测分配给这两个目标对象的一种可能的情况，也即，上述为一个联合事件。
本申请实施例对目标准则的具体内容不作限定，可选地，目标准则包括单一源准则和单一量测准则。其中，单一源准则是指任一有效量测对应的对象唯一，也即，在某一雷达帧内，一个有效量测仅属于一个目标对象，从而确认矩阵的每一行仅有一个“1”；单一量测准则是指任一对象在雷达帧中对应的有效量测唯一，也即，在某一雷达帧内，一个目标对象只能对应一个有效量测，从而确认矩阵的每一列最多只有一个“1”。
(5)对于目标对象中的第j个对象，基于联合事件的互联概率，确定第j个对象的加权量测，j为正整数。
在拆分确认矩阵得到联合事件后，服务器可以进一步计算该联合事件的互联概率，下面示出了联合事件θ i(k)的互联概率的表达式：
Figure PCTCN2020095889-appb-000003
其中，c表示归一化常数；V表示相关波门体积；P d表示检测概率，m k表示有效量测的个数，T表示目标对象的个数；N jt[z i(k)]由
Figure PCTCN2020095889-appb-000004
给出；δ t(θ i(k))和τ j(θ i(k))均表示二元指示变量；φ(θ i(k))由
Figure PCTCN2020095889-appb-000005
给出；v j(k)由
Figure PCTCN2020095889-appb-000006
给出，z j(k)称为一个有效量测，
Figure PCTCN2020095889-appb-000007
表示目标t在当前帧的预测位置，该预测位置由服务器基于雷达帧对应的聚合量测滤波得到，S t(k)表示第t个目标对象实际的误差协方差矩阵。
通过上述公式可以得到有效量测j与目标对象t之间的互联概率为：
Figure PCTCN2020095889-appb-000008
根据该互联概率对第t个目标对象的实际状态进行更新，得到：
Figure PCTCN2020095889-appb-000009
对第t个目标对象的误差协方差进行更新，得到：
Figure PCTCN2020095889-appb-000010
由此，可以进一步得出目标对象t的加权量测：
Figure PCTCN2020095889-appb-000011
需要说明的一点是，上述是以目标对象t为例说明加权量测的计算过程，对于第j个目标对象，该第j个目标对象的加权量测的计算可以参考目标对象t的加权量测的计算，在此不多赘述。
(6)根据第j个对象的加权量测，确定第j个对象的位置信息。
服务器在确定了第j个对象在某一雷达帧内针对各个有效量测的加权量测之后，可以根据第j个对象的加权量测，进一步确定第j个对象的位置信息。
综上所述，本申请实施例提供的技术方案，通过对聚合量测进行滤波处理，然后基于滤波处理后的量测选取有效量测，针对某一雷达帧，在该雷达帧内不包含有效量测的情况下，直接将滤波处理后的量测作为目标对象的位置信息，由于滤除处理后的量测进一步缩小了聚合量测的误差，因此，将该滤波处理后的量测确定为位置信息可以提升目标跟踪的准确性；在该雷达帧内包含有效量测的情况下，基于有效量测进一步确定加权量测，并根据加权量测来确定目标对象的位置信息，通过在滤波处理后的量测的基础上进一步确定有效量测，并基于加权量测来确定位置信息，可以进一步提升位置信息的准确性，提升目标跟踪的性能。
下面以一个示例性实施例对本申请的完整技术方案进行介绍说明。
假设环境辅助生活系统中有两个目标对象，分别为对象1和对象2，假设这两个目标对象均为人体，其中，对象1以目标位置为起点，以超宽带雷达为终端，径直向靠近超宽带雷达的方向行走；对象2以超宽带雷达为起点，以目标位置为终端，径直向远离超宽带的方向行走，并且，在对象1和对象2行走过程中，不断交换起点和终点，以达到在目标位置和超宽带雷达之间往返多次的目的，其中，目标位置是指在超宽带雷达的正前方，与超宽带雷达相隔一段距离的位置。假设对象1和对象2的行走速度均为0.5m/s，采集时长设置为20秒。为了和动作识别、步态识别等需要大量训练数据的应用场景保持软、硬件一致，超宽带雷达的帧率设置为150帧，也即，相邻雷达帧之间的时间间隔为1/150秒。
如图6所示，服务器先按照上述设置获取雷达帧，并且该雷达帧存储为基带信号形式。由于目标对象的行走速度缓慢，针对滤除杂波信号这一过程而言，没有必要采用过高的帧率，因此，服务器将每15个雷达帧按照时序排列为二维数据矩阵，并且采用SVD算法滤除杂波，并且取平均作为一帧，通过这种操作，帧率相当于降低为每秒10帧。接着，针对每一个雷达帧，服务器采用CLEAN算法提取得到量测。下述表一示出了CLEAN算法输入分别为射频信号形式的雷达帧和基带信号形式的雷达帧的检测对比。
表一 不同信号形式的雷达帧的处理性能的对比
  射频信号形式的雷达帧 基带信号形式的雷达帧
平均每帧处理时间/s 0.038 0.023
真实目标的量测/个 3-10 1-3
从上述表一中，可以看出采用基带信号形式的雷达帧相对于射频信号形式的雷达帧，平均每帧处理时间缩短了39.7％，处理速度上有了明显的提升。并且，采用基带信号形式的雷达帧，提取得到的真实目标的量测也有了明显的减少，通过散射点聚合算法后，距离扩展目标进一步趋向于点目标，这样，也就降低了后续处理过程中确认矩阵的维度，在目标对象的轨迹交叉处，也减少了相交波门的数量，防止出现组合爆炸的情况。另外，由于本申请实施 例中，基带信号形式的雷达帧的距离分辨率设置为射频信号形式的雷达帧的八倍，因此，基带信号形式的雷达帧的存储空间相比于射频信号形式的雷达帧，降低了八分之一。
图7示出了通过上述设置和计算过程得到的目标对象的跟踪滤波结果。为了说明本申请实施例提供的技术方案具有良好的性能，通过下述公式计算图7所示跟踪滤波结果的均方根误差：
Figure PCTCN2020095889-appb-000012
其中，z(i)表示目标对象的位置的真实值；
Figure PCTCN2020095889-appb-000013
表示目标对象的跟踪滤波结果
如下述表二所示，其示出了对象1和对象2的均方根误差。
表二 目标跟踪结果的均方根误差
目标对象 均方根误差(米)
对象1 0.1750
对象2 0.2650
从上述表二可以看出，目标对象的跟踪误差控制在0.3米之内。尽管在上述示例中，对象1和对象2的轨迹出现了交叉，但是量测与对象之间的关联准确，每个目标对象的轨迹都得到了正确地更新。因此，本申请实施例提供的技术方案具有良好的性能以及应用前景。
需要说明的一点是，本申请实施例仅以服务器执行上述方法的各个步骤为例进行举例说明，实际应用中，上述方法的各个步骤可以全部由服务器执行，也可以一部分由服务器执行，另一部分由终端设备执行，例如，上述方法步骤中对雷达帧对应的量测进行聚合处理的步骤可以由终端设备(如图1所示系统架构中的维护者终端22)来执行，其它步骤由服务器来执行。具体如何实现上述方法的各个步骤可以结合应用需求、系统架构设计等确定，本申请实施例对此不作限定，但这些方式均应属于本申请的保护范围之内。
下述为本申请装置实施例，可以用于执行本申请方法实施例。对于本申请装置实施例中未披露的细节，请参照本申请方法实施例。
请参考图8，其示出了本申请一个实施例提供的基于超宽带雷达的目标跟踪装置的框图。该装置具有实现上述方法示例的功能，所述功能可以通过硬件实现，也可以通过硬件执行相应的软件实现。该装置可以是图1所示系统架构中的服务器24，也可以设置在图1所示系统架构中的服务器24中。如图8所示，该装置800可以包括：信息获取模块810、量测确定模块820、聚合处理模块830和位置确定模块840。
信息获取模块810，用于获取超宽带雷达对回波信号进行采样得到的雷达帧。
量测确定模块820，用于基于所述雷达帧提取散射点，得到所述雷达帧对应的量测，所述量测是指所述散射点所在的距离单元，且所述量测与所述散射点一一对应。
聚合处理模块830，用于对所述雷达帧对应的量测进行聚合处理，得到所述雷达帧对应的聚合量测。
位置确定模块840，用于根据所述雷达帧对应的聚合量测，确定至少一个目标对象的位置信息。
在一个示例中，所述雷达帧对应的量测按照大小呈顺序排列，且所述雷达帧对应的量测的数量为n，所述n为正整数；所述聚合处理模块830，用于：对于所述n个量测中的第i个量测，确定所述第i个量测对应的搜索范围内是否包含其它量测，所述其它量测大于所述第i个量测，所述i为小于所述n的正整数；在所述搜索范围内包含所述其它量测的情况下，对所述第i个量测和所述其它量测进行平均处理，得到平均量测，所述雷达帧对应的聚合量测包括所述平均量测。
在一个示例中，所述聚合处理模块830，还用于：在所述搜索范围内不包含所述其它量测的情况下，保留所述第i个量测，所述雷达帧对应的聚合量测包括所述第i个量测。
在一个示例中，所述搜索范围包括前边界和后边界，所述前边界为所述第i个量测，所述后边界大于所述前边界，且所述后边界与所述前边界之间相隔r个距离单元，所述r为正 整数。
在一个示例中，如图9所示，所述装置800还包括：雷达帧获取模块850，用于获取在时域上连续的m个雷达帧；矩阵确定模块860，用于根据所述m个雷达帧得到二维数据矩阵；矩阵处理模块870，用于将所述二维数据矩阵分解得到的奇异向量中最大的k个奇异值置零，得到处理后的二维数据矩阵，所述k为正整数；雷达帧确定模块880，用于基于所述处理后的二维数据矩阵，得到滤波后的雷达帧，所述滤波后的雷达帧用于提取散射点得到所述雷达帧对应的量测；其中，所述二维数据矩阵中第x行第y列的元素用于指示第y个雷达帧在第x个距离单元处的回波强度，所述y为小于或等于所述m的正整数，所述x为正整数，且所述x小于或等于所述每个雷达帧包含的距离单元的总个数。
在一个示例中，所述k为1或2。
在一个示例中，所述m的取值范围为10至20。
在一个示例中，所述量测确定模块820，用于：获取波形模板，所述波形模板的信号形式和所述雷达帧的信号形式相同，且所述波形模板的距离分辨率和所述雷达帧的距离分辨率相同；采用所述波形模板提取所述雷达帧中的散射点。
在一个示例中，如图9所示，所述位置确定模块840，包括：滤波处理单元841，用于对所述雷达帧对应的聚合量测进行滤波处理，得到滤波后量测；量测选取单元842，用于从所述滤波后量测中选取有效量测；矩阵构建单元843，用于基于所述有效量测构建确认矩阵；事件确定单元844，用于按照目标准则拆分所述确认矩阵，得到至少一个联合事件，所述联合事件用于指示所述有效量测与所述目标对象之间的对应关系；量测加权单元845，用于对于所述目标对象中的第j个对象，基于所述联合事件的互联概率，确定所述第j个对象的加权量测，所述j为正整数；信息确定单元846，用于根据所述第j个对象的加权量测，确定所述第j个对象的位置信息；其中，所述确认矩阵中的第p行第q列的元素的取值为0或1，在所述取值为0的情况下，所述元素用于指示所述第p个有效量测没有位于所述第q个目标对象对应的距离波门内；在所述取值为1的情况下，所述元素用于指示所述第p个有效量测位于所述第q个目标对象对应的距离波门内。
在一个示例中，所述目标准则包括单一源准则和单一量测准则，所述单一源准则是指任一有效量测对应的对象唯一，所述单一量测准则是指任一对象在所述雷达帧中对应的有效量测唯一。
在一个示例中，如图9所示，所述量测选取单元842，用于：对于所述目标对象中的第j个对象，获取所述第j个对象的最大运动速度，以及所述超宽带雷达的距离分辨率；根据所述第j个对象的最大运动速度和所述距离分辨率，设置所述第j个对象对应的距离波门；将所述滤波后量测中位于所述第j个对象对应的距离波门内的量测确定为所述有效量测。
在一个示例中，如图9所示，所述信息确定单元846，还用于：在所述滤波后量测中没有位于所述第j个对象对应的距离波门内的量测的情况下，将所述滤波后量测确定为所述第j个对象的位置信息。
在一个示例中，所述雷达帧的信号形式为基带信号形式。
在一个示例中，所述超宽带雷达的体制为脉冲体制。
在一个示例中，如图9所示，所述装置800还包括：轨迹确定模块890，用于对于所述目标对象中的第j个对象，按照x个雷达帧在时域上的先后顺序，串联所述第j个对象的位置信息，得到所述第j个对象的运动轨迹，所述x为大于1的整数，所述j为正整数。
综上所述，本申请实施例提供的技术方案，通过获取超宽带雷达对回波信号进行采样得到的雷达帧，并基于雷达帧提取散射点，将散射点所在的距离单元作为量测，以得到雷达帧对应的量测，然后对量测进行聚合处理，得到聚合量测，以降低量测的数量，提升服务器的处理速度，之后根据聚合量测进一步确定目标对象的位置信息。本申请实施例提供的技术方案，由于对量测进行了聚合处理，降低了量测的数量，从而有效缓解了服务器的存储压力和降低了服务器的计算开销，避免资源浪费。另外，由于超宽带雷达采集回波信号的过程中不 需要拍摄图像和视频，且超宽带雷达具备非侵入性的特点，本申请实施例通过将超宽带雷达应用于目标对象的检测和跟踪，充分考虑了用户的使用习惯，有助于保护用户的隐私。
需要说明的是，上述实施例提供的装置在实现其功能时，仅以上述各功能模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能模块完成，即将设备的内部结构划分成不同的功能模块，以完成以上描述的全部或者部分功能。另外，上述实施例提供的装置与方法实施例属于同一构思，其具体实现过程详见方法实施例，这里不再赘述。
请参考图10，其示出了本申请一个实施例提供的计算机设备的结构框图。
本申请实施例中的计算机设备可以包括一个或多个如下部件：处理器1010和存储器1020。
处理器1010可以包括一个或者多个处理核心。处理器1010利用各种接口和线路连接整个计算机设备内的各个部分，通过运行或执行存储在存储器1020内的指令、程序、代码集或指令集，以及调用存储在存储器1020内的数据，执行计算机设备的各种功能和处理数据。可选地，处理器1010可以采用数字信号处理(Digital Signal Processing，DSP)、现场可编程门阵列(Field-Programmable Gate Array，FPGA)、可编程逻辑阵列(Programmable Logic Array，PLA)中的至少一种硬件形式来实现。处理器1010可集成中央处理器(Central Processing Unit，CPU)和调制解调器等中的一种或几种的组合。其中，CPU主要处理操作系统和应用程序等；调制解调器用于处理无线通信。可以理解的是，上述调制解调器也可以不集成到处理器1010中，单独通过一块芯片进行实现。
可选地，处理器1010执行存储器1020中的程序指令时实现上述各个方法实施例提供的方法。
存储器1020可以包括随机存储器(Random Access Memory，RAM)，也可以包括只读存储器(Read-Only Memory，ROM)。可选地，该存储器1020包括非瞬时性计算机可读介质(non-transitory computer-readable storage medium)。存储器1020可用于存储指令、程序、代码、代码集或指令集。存储器1020可包括存储程序区和存储数据区，其中，存储程序区可存储用于实现操作系统的指令、用于至少一个功能的指令、用于实现上述各个方法实施例的指令等；存储数据区可存储根据计算机设备的使用所创建的数据等。
上述计算机设备的结构仅是示意性的，在实际实现时，计算机设备可以包括更多或更少的组件，比如：显示屏等，本实施例对此不作限定。
本领域技术人员可以理解，图10中示出的结构并不构成对计算机设备的限定，可以包括比图示更多或更少的组件，或者组合某些组件，或者采用不同的组件布置。
本申请实施例还提供了一种计算机可读存储介质，所述存储介质中存储有计算机程序，所述计算机程序用于被计算机设备的处理器执行，以实现上述基于超宽带雷达的目标跟踪方法。
本申请实施例还提供了一种芯片，所述芯片包括可编程逻辑电路和/或程序指令，当所述芯片在计算机设备上运行时，用于实现如上述基于超宽带雷达的目标跟踪方法。
本申请实施例还提供了一种计算机程序产品，当计算机程序产品在计算机设备上运行时，使得计算机设备执行上述基于超宽带雷达的目标跟踪方法。
本领域技术人员应该可以意识到，在上述一个或多个示例中，本申请实施例所描述的功能可以用硬件、软件、固件或它们的任意组合来实现。当使用软件实现时，可以将这些功能存储在计算机可读介质中或者作为计算机可读介质上的一个或多个指令或代码进行传输。计算机可读介质包括计算机存储介质和通信介质，其中通信介质包括便于从一个地方向另一个地方传送计算机程序的任何介质。存储介质可以是通用或专用计算机能够存取的任何可用介质。
以上所述仅为本申请的示例性实施例，并不用以限制本申请，凡在本申请的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本申请的保护范围之内。

Claims (32)

  1. 一种基于超宽带雷达的目标跟踪方法，其特征在于，所述方法包括：
    获取超宽带雷达对回波信号进行采样得到的雷达帧；
    基于所述雷达帧提取散射点，得到所述雷达帧对应的量测，所述量测是指所述散射点所在的距离单元，且所述量测与所述散射点一一对应；
    对所述雷达帧对应的量测进行聚合处理，得到所述雷达帧对应的聚合量测；
    根据所述雷达帧对应的聚合量测，确定至少一个目标对象的位置信息。
  2. 根据权利要求1所述的方法，其特征在于，所述雷达帧对应的量测按照大小呈顺序排列，且所述雷达帧对应的量测的数量为n，所述n为正整数；
    所述对所述雷达帧对应的量测进行聚合处理，得到所述雷达帧对应的聚合量测，包括：
    对于所述n个量测中的第i个量测，确定所述第i个量测对应的搜索范围内是否包含其它量测，所述其它量测大于所述第i个量测，所述i为小于所述n的正整数；
    在所述搜索范围内包含所述其它量测的情况下，对所述第i个量测和所述其它量测进行平均处理，得到平均量测，所述雷达帧对应的聚合量测包括所述平均量测。
  3. 根据权利要求2所述的方法，其特征在于，所述方法还包括：
    在所述搜索范围内不包含所述其它量测的情况下，保留所述第i个量测，所述雷达帧对应的聚合量测包括所述第i个量测。
  4. 根据权利要求2或3所述的方法，其特征在于，所述搜索范围包括前边界和后边界，所述前边界为所述第i个量测，所述后边界大于所述前边界，且所述后边界与所述前边界之间相隔r个距离单元，所述r为正整数。
  5. 根据权利要求1至4任一项所述的方法，其特征在于，所述基于所述雷达帧提取散射点之前，还包括：
    获取在时域上连续的m个雷达帧；
    根据所述m个雷达帧得到二维数据矩阵；
    将所述二维数据矩阵分解得到的奇异向量中最大的k个奇异值置零，得到处理后的二维数据矩阵，所述k为正整数；
    基于所述处理后的二维数据矩阵，得到滤波后的雷达帧，所述滤波后的雷达帧用于提取散射点得到所述雷达帧对应的量测；
    其中，所述二维数据矩阵中第x行第y列的元素用于指示第y个雷达帧在第x个距离单元处的回波强度，所述y为小于或等于所述m的正整数，所述x为正整数，且所述x小于或等于所述每个雷达帧包含的距离单元的总个数。
  6. 根据权利要求5所述的方法，其特征在于，所述k为1或2。
  7. 根据权利要求5或6所述的方法，其特征在于，所述m的取值范围为10至20。
  8. 根据权利要求1至7任一项所述的方法，其特征在于，所述基于所述雷达帧提取散射点，包括：
    获取波形模板，所述波形模板的信号形式和所述雷达帧的信号形式相同，且所述波形模板的距离分辨率和所述雷达帧的距离分辨率相同；
    采用所述波形模板提取所述雷达帧中的散射点。
  9. 根据权利要求1至8任一项所述的方法，其特征在于，所述根据所述雷达帧对应的聚合量测，确定至少一个目标对象的位置信息，包括：
    对所述雷达帧对应的聚合量测进行滤波处理，得到滤波后量测；
    从所述滤波后量测中选取有效量测；
    基于所述有效量测构建确认矩阵；
    按照目标准则拆分所述确认矩阵，得到至少一个联合事件，所述联合事件用于指示所述有效量测与所述目标对象之间的对应关系；
    对于所述目标对象中的第j个对象，基于所述联合事件的互联概率，确定所述第j个对象的加权量测，所述j为正整数；
    根据所述第j个对象的加权量测，确定所述第j个对象的位置信息；
    其中，所述确认矩阵中的第p行第q列的元素的取值为0或1，在所述取值为0的情况下，所述元素用于指示所述第p个有效量测没有位于所述第q个目标对象对应的距离波门内；在所述取值为1的情况下，所述元素用于指示所述第p个有效量测位于所述第q个目标对象对应的距离波门内。
  10. 根据权利要求9所述的方法，其特征在于，所述目标准则包括单一源准则和单一量测准则，所述单一源准则是指任一有效量测对应的对象唯一，所述单一量测准则是指任一对象在所述雷达帧中对应的有效量测唯一。
  11. 根据权利要求9或10所述的方法，其特征在于，所述从所述滤波后量测中选取有效量测，包括：
    对于所述目标对象中的第j个对象，获取所述第j个对象的最大运动速度，以及所述超宽带雷达的距离分辨率；
    根据所述第j个对象的最大运动速度和所述距离分辨率，设置所述第j个对象对应的距离波门；
    将所述滤波后量测中位于所述第j个对象对应的距离波门内的量测确定为所述有效量测。
  12. 根据权利要求11所述的方法，其特征在于，所述方法还包括：
    在所述滤波后量测中没有位于所述第j个对象对应的距离波门内的量测的情况下，将所述滤波后量测确定为所述第j个对象的位置信息。
  13. 根据权利要求1至12任一项所述的方法，其特征在于，所述雷达帧的信号形式为基带信号形式。
  14. 根据权利要求1至13任一项所述的方法，其特征在于，所述超宽带雷达的体制为脉冲体制。
  15. 根据权利要求1至14任一项所述的方法，其特征在于，所述方法还包括：
    对于所述目标对象中的第j个对象，按照x个雷达帧在时域上的先后顺序，串联所述第j个对象的位置信息，得到所述第j个对象的运动轨迹，所述x为大于1的整数，所述j为正整数。
  16. 一种基于超宽带雷达的目标跟踪装置，其特征在于，所述装置包括：
    信息获取模块，用于获取超宽带雷达对回波信号进行采样得到的雷达帧；
    量测确定模块，用于基于所述雷达帧提取散射点，得到所述雷达帧对应的量测，所述量测是指所述散射点所在的距离单元，且所述量测与所述散射点一一对应；
    聚合处理模块，用于对所述雷达帧对应的量测进行聚合处理，得到所述雷达帧对应的聚合量测；
    位置确定模块，用于根据所述雷达帧对应的聚合量测，确定至少一个目标对象的位置信息。
  17. 根据权利要求16所述的装置，其特征在于，所述雷达帧对应的量测按照大小呈顺序排列，且所述雷达帧对应的量测的数量为n，所述n为正整数；所述聚合处理模块，用于：
    对于所述n个量测中的第i个量测，确定所述第i个量测对应的搜索范围内是否包含其它量测，所述其它量测大于所述第i个量测，所述i为小于所述n的正整数；
    在所述搜索范围内包含所述其它量测的情况下，对所述第i个量测和所述其它量测进行平均处理，得到平均量测，所述雷达帧对应的聚合量测包括所述平均量测。
  18. 根据权利要求17所述的装置，其特征在于，所述聚合处理模块，还用于：
    在所述搜索范围内不包含所述其它量测的情况下，保留所述第i个量测，所述雷达帧对应的聚合量测包括所述第i个量测。
  19. 根据权利要求17或18所述的装置，其特征在于，所述搜索范围包括前边界和后边界，所述前边界为所述第i个量测，所述后边界大于所述前边界，且所述后边界与所述前边界之间相隔r个距离单元，所述r为正整数。
  20. 根据权利要求16至19任一项所述的装置，其特征在于，所述装置还包括：
    雷达帧获取模块，用于获取在时域上连续的m个雷达帧；
    矩阵确定模块，用于根据所述m个雷达帧得到二维数据矩阵；
    矩阵处理模块，用于将所述二维数据矩阵分解得到的奇异向量中最大的k个奇异值置零，得到处理后的二维数据矩阵，所述k为正整数；
    雷达帧确定模块，用于基于所述处理后的二维数据矩阵，得到滤波后的雷达帧，所述滤波后的雷达帧用于提取散射点得到所述雷达帧对应的量测；
    其中，所述二维数据矩阵中第x行第y列的元素用于指示第y个雷达帧在第x个距离单元处的回波强度，所述y为小于或等于所述m的正整数，所述x为正整数，且所述x小于或等于所述每个雷达帧包含的距离单元的总个数。
  21. 根据权利要求20所述的装置，其特征在于，所述k为1或2。
  22. 根据权利要求20或21所述的装置，其特征在于，所述m的取值范围为10至20。
  23. 根据权利要求16至22任一项所述的装置，其特征在于，所述量测确定模块，用于：
    获取波形模板，所述波形模板的信号形式和所述雷达帧的信号形式相同，且所述波形模板的距离分辨率和所述雷达帧的距离分辨率相同；
    采用所述波形模板提取所述雷达帧中的散射点。
  24. 根据权利要求16至23任一项所述的装置，其特征在于，所述位置确定模块，包括：
    滤波处理单元，用于对所述雷达帧对应的聚合量测进行滤波处理，得到滤波后量测；
    量测选取单元，用于从所述滤波后量测中选取有效量测；
    矩阵构建单元，用于基于所述有效量测构建确认矩阵；
    事件确定单元，用于按照目标准则拆分所述确认矩阵，得到至少一个联合事件，所述联合事件用于指示所述有效量测与所述目标对象之间的对应关系；
    量测加权单元，用于对于所述目标对象中的第j个对象，基于所述联合事件的互联概率，确定所述第j个对象的加权量测，所述j为正整数；
    信息确定单元，用于根据所述第j个对象的加权量测，确定所述第j个对象的位置信息；
    其中，所述确认矩阵中的第p行第q列的元素的取值为0或1，在所述取值为0的情况下，所述元素用于指示所述第p个有效量测没有位于所述第q个目标对象对应的距离波门内；在所述取值为1的情况下，所述元素用于指示所述第p个有效量测位于所述第q个目标对象对应的距离波门内。
  25. 根据权利要求24所述的装置，其特征在于，所述目标准则包括单一源准则和单一量测准则，所述单一源准则是指任一有效量测对应的对象唯一，所述单一量测准则是指任一对象在所述雷达帧中对应的有效量测唯一。
  26. 根据权利要求24或25所述的装置，其特征在于，所述量测选取单元，用于：
    对于所述目标对象中的第j个对象，获取所述第j个对象的最大运动速度，以及所述超宽带雷达的距离分辨率；
    根据所述第j个对象的最大运动速度和所述距离分辨率，设置所述第j个对象对应的距离波门；
    将所述滤波后量测中位于所述第j个对象对应的距离波门内的量测确定为所述有效量测。
  27. 根据权利要求26所述的装置，其特征在于，所述信息确定单元，还用于：
    在所述滤波后量测中没有位于所述第j个对象对应的距离波门内的量测的情况下，将所述滤波后量测确定为所述第j个对象的位置信息。
  28. 根据权利要求16至27任一项所述的装置，其特征在于，所述雷达帧的信号形式为基带信号形式。
  29. 根据权利要求16至28任一项所述的装置，其特征在于，所述超宽带雷达的体制为脉冲体制。
  30. 根据权利要求16至29任一项所述的装置，其特征在于，所述装置还包括：
    轨迹确定模块，用于对于所述目标对象中的第j个对象，按照x个雷达帧在时域上的先后顺序，串联所述第j个对象的位置信息，得到所述第j个对象的运动轨迹，所述x为大于1的整数，所述j为正整数。
  31. 一种计算机设备，其特征在于，所述计算机设备包括处理器和存储器，所述存储器存储有计算机程序，所述计算机程序由所述处理器加载并执行以实现如权利要求1至15任一项所述的基于超宽带雷达的目标跟踪方法。
  32. 一种计算机可读存储介质，其特征在于，所述存储介质中存储有计算机程序，所述计算机程序用于被计算机设备的处理器执行，以实现如权利要求1或15任一项所述的基于超宽带雷达的目标跟踪方法。
PCT/CN2020/095889 2020-06-12 2020-06-12 基于超宽带雷达的目标跟踪方法、装置、设备及存储介质 WO2021248472A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202080001113.7A CN114080549A (zh) 2020-06-12 2020-06-12 基于超宽带雷达的目标跟踪方法、装置、设备及存储介质
PCT/CN2020/095889 WO2021248472A1 (zh) 2020-06-12 2020-06-12 基于超宽带雷达的目标跟踪方法、装置、设备及存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/095889 WO2021248472A1 (zh) 2020-06-12 2020-06-12 基于超宽带雷达的目标跟踪方法、装置、设备及存储介质

Publications (1)

Publication Number Publication Date
WO2021248472A1 true WO2021248472A1 (zh) 2021-12-16

Family

ID=78846759

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/095889 WO2021248472A1 (zh) 2020-06-12 2020-06-12 基于超宽带雷达的目标跟踪方法、装置、设备及存储介质

Country Status (2)

Country Link
CN (1) CN114080549A (zh)
WO (1) WO2021248472A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114492505A (zh) * 2021-12-24 2022-05-13 西安电子科技大学 基于半实测数据的空中群目标和扩展目标识别方法
CN116224379A (zh) * 2023-05-06 2023-06-06 中国科学院国家空间科学中心 一种nbrcs修正方法、装置、电子设备及存储介质
CN116755057A (zh) * 2023-08-21 2023-09-15 深圳市速腾聚创科技有限公司 参数配置方法、装置及计算机可读存储介质

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115984027B (zh) * 2023-03-20 2023-06-06 长沙迪迈数码科技股份有限公司 基于uwb的井下人员聚集预警方法、装置及存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104166135A (zh) * 2014-09-05 2014-11-26 武汉中原电子集团有限公司 一种宽带雷达目标的原始点迹凝聚处理方法
US20180341019A1 (en) * 2017-05-26 2018-11-29 Toyota Motor Engineering & Manufacturing North America, Inc. Publishing lidar cluster data
CN109188423A (zh) * 2018-08-29 2019-01-11 电子科技大学 一种基于多源聚类的分布式多目标跟踪方法
CN110542897A (zh) * 2019-08-01 2019-12-06 北京理工大学 一种基于希尔排序的距离差分多扩展目标点迹凝聚方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104166135A (zh) * 2014-09-05 2014-11-26 武汉中原电子集团有限公司 一种宽带雷达目标的原始点迹凝聚处理方法
US20180341019A1 (en) * 2017-05-26 2018-11-29 Toyota Motor Engineering & Manufacturing North America, Inc. Publishing lidar cluster data
CN109188423A (zh) * 2018-08-29 2019-01-11 电子科技大学 一种基于多源聚类的分布式多目标跟踪方法
CN110542897A (zh) * 2019-08-01 2019-12-06 北京理工大学 一种基于希尔排序的距离差分多扩展目标点迹凝聚方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LIU, JINCHAO ET AL.: "Algorithm for Multiple Human Targets Tracking with Ultra-Wideband Radar", XIANDAI-LEIDA = MODERN RADAR, vol. 37, no. 2, 28 February 2015 (2015-02-28), CN , pages 37 - 40, XP009532747, ISSN: 1004-7859 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114492505A (zh) * 2021-12-24 2022-05-13 西安电子科技大学 基于半实测数据的空中群目标和扩展目标识别方法
CN116224379A (zh) * 2023-05-06 2023-06-06 中国科学院国家空间科学中心 一种nbrcs修正方法、装置、电子设备及存储介质
CN116224379B (zh) * 2023-05-06 2023-09-12 中国科学院国家空间科学中心 一种nbrcs修正方法、装置、电子设备及存储介质
CN116755057A (zh) * 2023-08-21 2023-09-15 深圳市速腾聚创科技有限公司 参数配置方法、装置及计算机可读存储介质
CN116755057B (zh) * 2023-08-21 2024-01-16 深圳市速腾聚创科技有限公司 参数配置方法、装置及计算机可读存储介质

Also Published As

Publication number Publication date
CN114080549A (zh) 2022-02-22

Similar Documents

Publication Publication Date Title
WO2021248472A1 (zh) 基于超宽带雷达的目标跟踪方法、装置、设备及存储介质
Wang et al. m-activity: Accurate and real-time human activity recognition via millimeter wave radar
US20200166610A1 (en) Detection method, detection device, terminal and detection system
CN111695420B (zh) 一种手势识别方法以及相关装置
CN106559749B (zh) 一种基于射频层析成像的多目标被动式定位方法
CN112990026B (zh) 基于对抗训练的无线信号感知模型构建、感知方法及系统
CN113447905A (zh) 双毫米波雷达人体跌倒检测装置及检测方法
CN115308734A (zh) 一种呼吸数据计算方法以及相关设备
CN111624567A (zh) 一种恒虚警检测方法及装置
CN114038012A (zh) 基于毫米波雷达和机器学习的跌倒检测方法及系统
CN112198507B (zh) 人体跌倒特征的检测方法及装置
CN114924225A (zh) 一种高精度室内定位方法、装置、设备及介质
Qian et al. Parallel LSTM-CNN network with radar multispectrogram for human activity recognition
CN115469301A (zh) 基于超宽带雷达的老年人活动能力评估方法及系统
CN112380903B (zh) 一种基于WiFi-CSI信号增强的人体活动识别方法
Chen et al. A safe-distance control scheme to avoid new infection like COVID-19 virus using millimeter-wave radar
CN115469303A (zh) 一种检测人体姿态和生命体征的认知生物雷达方法与装置
CN117741647A (zh) 一种毫米波雷达智能家居感知方法与装置
CN110636439B (zh) 位置获取方法及装置
CN115841707A (zh) 基于深度学习的雷达人体姿态识别方法及相关设备
TWI772208B (zh) 基於毫米波雷達的人數偵測方法
CN106249241B (zh) 一种自适应杂波功率统计算法
Saw et al. Neuromorphic computing with hybrid CNN–Stochastic Reservoir for time series WiFi based human activity recognition
CN115291184A (zh) 一种毫米波雷达与深度学习结合的姿态监测方法
CN114942416A (zh) 一种状态识别方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20940227

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20940227

Country of ref document: EP

Kind code of ref document: A1