Mapping of quiesce scene using radar

Abstract

To provide a method for mapping using a floor-mounted type radar unit in a quiesce scene.SOLUTION: A floor mounted type radar unit (100) includes one set of receiver antennas (102a-h) which are configured to detect a radar signal from any direction, and measures a target speed in an individual speed jar. In the method, the set of the receiver antennas are used for continuously and chronologically collecting the radar signal for detecting a quiesce scene, then constructing an occupancy map of the quiesce scene by using confirmed detection which is determined on the basis of the collected radar signals. The detections are: detection in which, radar signal strength exceeds a detection threshold value, and the speed is within a zero speed jar; and detection in which, the radar signal strength exceeds the detection threshold value, the speed is not zero, and is the detection being low enough to store overflow information in a same jar, and in a zero speed jar.SELECTED DRAWING: Figure 1A

Description

Technical Field

[0001] The present invention generally relates to the field of radar imaging, and more particularly to a method and a control unit for mapping a stationary scene using a stationary radar unit.

Background Art

[0002] The principle of radar detection is that the transmitted radar signal is reflected by an object and detected by the radar antenna. By appropriate signal processing, properties such as the object velocity and the position relative to the radar can be estimated.

[0003] Radar mapping of stationary installations may be performed using a scanning radar, where the radar antenna is spatially moved so that detection areas are detected from various angles. In some applications, the radar is necessarily in motion. For example, in the automotive industry, the radar is mounted on a vehicle to detect the surroundings of that vehicle.

[0004] A common type of radar is based on what is called range-Doppler measurement. Typically, a range-Doppler measurement radar detects a specific combination of the distance and velocity of a target. However, this type of measurement is most suitable for detecting moving objects in the detection area and is not suitable for detecting stationary, non-moving objects.

[0005] A common approach to increasing the resolution of a stationary image detected by radar is to significantly increase the number of antennas and employ complex and computationally expensive signal processing. Increasing the number of antennas is costly and generally not desirable, and computationally expensive signal processing is not desirable as it increases the requirements for the processor.

[0006] As a result, there is room for improvement with respect to detecting stationary installations using radar.

Summary of the Invention

[0007] In consideration of the above and other drawbacks of the prior art, the object of the present invention is to provide a method for mapping a stationary scene using a stationary radar unit that mitigates at least some of those drawbacks of the prior art.

[0008] According to a first aspect of the present invention, a method for mapping a stationary scene using a stationary radar unit that operates to transmit radar signals toward the scene is provided. The stationary radar unit includes a set of receiver antennas configured to detect radar signals from any direction. The stationary radar unit is configured to measure target velocity in separate velocity bins. The method includes the following steps: detecting a stationary scene by continuously collecting radar signals over time using the set of receiver antennas; and constructing an occupancy map of the stationary scene using confirmed detections determined from the collected radar signals. Confirmed detections are detections where the radar signal intensity exceeds a detection threshold and the velocity falls into the zero velocity bin, and detections where the radar signal intensity exceeds a detection threshold and the velocity is not zero, but is low enough to produce overflow information in the same bin as a detection falling into the zero velocity bin.

[0009] This invention is based on the realization of including not only the detection of stationary objects but also the detection of nearly stationary objects in order to construct a map of stationary objects in a scene. Information regarding the detection of nearly stationary objects is often discarded because it may be overflow information in the nearly stationary region from the detection of stationary objects, i.e., information that may already be considered to form the detection of stationary objects. Furthermore, radar detection of moving objects also generates overflow information into the same nearly stationary region. Therefore, the information in the nearly stationary region includes the sum of overflow information from stationary objects and from moving objects.

[0010] The inventors have realized that they can use information about detecting targets with non-zero velocity to construct a map of a static scene.

[0011] The radar measures velocity in individual steps, ensuring that the detected velocity falls within one of several bins. Velocity detection is therefore performed, for example, for velocities of 0, and for velocities of -ΔV and +ΔV, where ΔV is the bin size. Thus, the radar unit operates according to threshold settings to include detections where the signal strength exceeds the threshold and the velocity is zero, or greater than zero but within + / -ΔV. Due to overflow, each detection for the zero bin, the -ΔV bin, or the +ΔV bin may include detections from stationary objects and detections from moving objects.

[0012] A speed low enough to generate overflow information can be interpreted as a speed low enough that there is a correlation or linear dependence between the frequencies corresponding to the target at zero speed and the target at a non-zero speed.

[0013] In addition to measuring velocity in individual velocity bins, stationary radar units may also measure target range in individual range bins, and optionally, the angle of arrival of the radar signal in individual bins. This results in overflow information between target range bins and between angle of arrival bins.

[0014] Overflow information can be considered as spectral leakage, i.e., during signal processing involving a discrete Fourier transform, the lobe or "tone" in the discrete Fourier transform relating to the detected velocity spreads to other frequencies, such as velocity. Similarly, during signal processing involving a discrete Fourier transform, the lobe or "tone" in the discrete Fourier transform relating to the detected range or the angle of arrival of the detected signal spreads to other frequencies, such as other neighboring range bins or angle of arrival bins.

[0015] A stationary radar should be interpreted as a radar that does not have moving parts to sweep the scene. Instead, the radar adapts to signals detected from all directions without using beamforming or sweeping. Furthermore, stationary radar is mounted in a stationary position and not on moving entities such as vehicles. Therefore, the stationary radar unit is not moving relative to a stationary scene.

[0016] The continuous acquisition of radar signals should be interpreted as being performed at regular intervals defined by time frames or simply "frames".

[0017] Speed ​​is preferably measured by periodically transmitting a radar signal. Suitable transmission signals are periodically repeated pulsed signals, frequency-modulated sequences, in particular linear frequency sweeps used in frequency-modulated continuous-wave radar (FMCW), phase-modulated signals, or multi-carrier signals such as orthogonal frequency division multiplexing (OFDM).

[0018] The radar signal may be processed using various known methods. Some examples include range compression using matched filters. For FMCW signals using stretching, this can be done by Fourier transform or other frequency estimation algorithms (e.g., MUSIC, CAPON). Other suitable implementations may be correlation with the transmitted signal or a frequency-domain implementation of a matched filter.

[0019] Furthermore, Doppler compression may include analyzing the periodicity of the transmitted signal and detecting the Doppler shift. This is commonly done using the Fast Fourier Transform (FFT) or other frequency estimation algorithms described above.

[0020] Furthermore, the angle of arrival in an antenna array may be estimated by estimating the phase steering vector between antennas using FFT. Other techniques may be the frequency estimation techniques described above.

[0021] Using embodiments of the present invention, the method described below provides mapping of a stationary scene using a radar with a relatively small number of antennas and low computational cost signal processing.

[0022] In each embodiment, the number of receiver antennas does not exceed 16, preferably not exceeding 10, and more preferably not exceeding 8. As a result, the radar unit can be a relatively simple and low-cost type. The receiver antennas may be physical antennas or virtual antennas. The antennas may be implemented as virtual antennas for a multiple-in multiple-out (MIMO) radar.

[0023] In each embodiment, constructing an occupancy map may include, over time, including in the occupancy map all detections where the radar signal intensity exceeds the detection threshold and there is no velocity, and all detections where the radar signal intensity exceeds the detection threshold and there is a velocity that is not zero, for each of a set of distinct radii from the radar's position. As a result, the radar unit is configured to mark all detections that satisfy the detection and velocity thresholds for each of a set of distinct radii. For example, all detections where the velocity falls within the thresholds of 0, -ΔV, +ΔV, and - are marked in the occupancy map.

[0024] In a more specific example, the occupancy map is constructed by including one radar detection for each combination of a distinct radius from the radar's position and detections related to the neighboring velocity bins of the zero velocity bin.

[0025] In each embodiment, constructing an occupation map may include increasing the occupation probability value in a corresponding first region of the occupation map for each confirmed detection, and decreasing the probability value in the region between the first region and the radar position in the occupation map. This is an efficient method for constructing an occupation map that takes into account the detection of moving objects. Preferably, it allows the map to be constructed by averaging detections over time, where stationary objects appear and moving objects are averaged out. The degree to which the probability value is increased and / or decreased is a tuning parameter that may be tuned to a particular implementation. The tuning parameter effectively controls the speed at which the occupation map is constructed, i.e., how quickly the probability value approaches a set limit.

[0026] Preferably, the region between the first region in the occupied map and the radar's position intersects the straight line between the first region and the radar's position in the occupied map. This straight line represents the free line of sight between the radar unit and the first region. Since there are no further objects in the free line of sight, the probability values ​​in the region between the first region and the radar unit are reduced with high reliability, which may improve the contrast between the region with stationary objects and the region without stationary objects in the occupied map. Objects located on the straight line but beyond the first region, which are presumably from the radar unit's position, are completely or partially obscured by shadow.

[0027] Furthermore, while constructing the occupancy map, it may include increasing the occupancy probability values ​​in neighboring regions of the first region. This allows the uncertainty in the measured target range and angle of arrival to be explained in the occupancy map. This increase may be carried out according to a probability function, for example, the probability increases more rapidly as the region approaches the first region, and decreases more rapidly as the distance from the first region increases.

[0028] To construct an accurate occupancy map, the occupancy map may be constructed as a collection of probability values over time. Since the occupancy map is effectively a collection of estimated detected environments, the differences in this estimation are reduced by averaging many measurements over time.

[0029] In each embodiment, constructing the occupancy map may include determining using selecting the frequency with the most energy content in the Fourier transform of the phase difference between radar signals collected by various receiver antennas divided along a first axis, of the azimuth angle of a confirmed detection. The first axis corresponds to the horizontal axis of the scene when the radar unit is in use.

[0030] In each embodiment, constructing the occupancy map may include determining using selecting the frequency with the most energy content in the Fourier transform of the phase difference between radar signals collected by various receiver antennas divided along a second axis, of the elevation angle of a confirmed detection. The second axis corresponds to the vertical axis of the scene when the radar unit is in use.

[0031] Each of the embodiments of the present invention is applicable to various types of stationary radar units. In one preferred embodiment, the stationary radar unit is a frequency-modulated continuous-wave radar. Further, the stationary radar unit may be operative to transmit radar signals in all directions.

[0032] According to a second aspect of the present invention, there is provided a computer program including instructions which, when the program is executed by a computer, cause the computer to execute any one of the methods of the embodiments described herein.

[0033] Each of the further embodiments of this second aspect of the present invention, and the effects obtained through them, are largely similar to those described above for the first aspect of the present invention.

[0034] According to a third aspect of the present invention, a control unit is provided for mapping a stationary scene using radar signals collected by a stationary radar unit that operates to transmit radar signals toward the scene and to detect radar signals from any direction using a set of receiver antennas. The stationary radar unit is configured to measure target velocity in individual velocity bins. The control unit is configured to detect a stationary scene by acquiring data showing radar signals collected continuously over time and to construct an occupancy map of the stationary scene using confirmed detections determined from the acquired data. Confirmed detections are detections where the radar signal intensity exceeds a detection threshold and the velocity falls into the zero velocity bin, and detections where the radar signal intensity exceeds a detection threshold and the velocity is not zero, but is low enough to produce overflow information in the same bin as a detection falling into the zero velocity bin.

[0035] Each of these further embodiments of this third aspect of the present invention, and the effects obtained therefrom, are largely similar to the above-described embodiments of the first and second aspects of the present invention.

[0036] A system is further provided that includes a stationary radar unit and a control unit according to a third embodiment. This system offers similar advantages and features to any one of the embodiments described below.

[0037] Further features of the present invention and the advantages thereof will become apparent when considering the appended claims and the following description. A person skilled in the art will recognize that various features of the present invention can be combined to create other embodiments described below, each without departing from the scope of the invention.

[0038] Various aspects of the present invention, including their specific features and advantages, will be readily apparent from the following embodiments for carrying out the invention and the accompanying drawings shown below. [Brief explanation of the drawing]

[0039] [Figure 1A] This conceptually illustrates a stationary radar unit positioned to acquire radar detection from a scene containing both stationary and moving objects. [Figure 1B] Conceptual range - Doppler diagram and power vs. range diagram for stationary objects. [Figure 1C] Conceptual range - Doppler diagram and power vs. range diagram of two stationary objects. [Figure 1D] Conceptual range - Doppler diagram and diagram of radar signal power vs. range for stationary and moving objects. [Figure 1E] This is a conceptual power-velocity diagram of radar signals between a stationary and a moving object. [Figure 2] This is a flowchart of the method steps according to each embodiment of the present invention. [Figure 3] The construction of the occupancy map according to each embodiment of the present invention is conceptually shown. [Figure 4A-4B] Examples of constructed occupancy maps for two separate time periods, according to each embodiment of the present invention, are shown. [Figure 5] These are examples of individual Fourier spectra of azimuth angle versus power according to each embodiment of the present invention. [Figure 6] This is a block diagram of the system according to each embodiment of the present invention. [Modes for carrying out the invention]

[0040] The present invention will be described in further detail below with reference to the accompanying drawings. Herein, we show current preferred embodiments of the invention. However, the present invention may be embodied in many different forms and should not be understood as being limited to the embodiments shown below. Rather, these embodiments are provided for the sake of perfection and completeness, and to fully convey the scope of the invention to those skilled in the art. Similar reference numerals indicate similar components throughout these drawings.

[0041] These drawings, particularly Figure 1, show Scene 1, which includes a set of stationary objects 102a to d and a moving object 104. A stationary radar unit 100 is configured to map Scene 1, which includes the moving object 104 and the stationary objects 102a to d. Periodic signal transmission is preferably used to measure velocity. Suitable transmission signals are periodically repeated pulsed signals, frequency modulation sequences such as linear frequency sweeps used in FMCW, phase modulation signals, or multicarrier signals such as OFDM.

[0042] The radar unit 100 generally detects the strongest signal for each radius or range from the position of the radar unit 100. Therefore, if the scene does not change and remains completely stationary, the same strongest object will be observed over time, and the observed stationary object will often be blurred across azimuth angles for each range. However, when a small interference, such as a moving object 104, is introduced into scene 1, the scene changes, and therefore the strongest detection may also change. This may result in the appearance of less prominent features of the scene. For example, the moving object 104 first causes the obstruction of the area 106a of object 102b, and later causes the obstruction of the area 106b of object 102c. Without the moving object 104, the smaller stationary object 102b may not be detected as the strongest detection due to the prominent feature, stationary object 102c. However, when the moving object 104 covers part of the stationary object 102c, the less prominent object 102b may be detected and appear as the more prominent object.

[0043] During scene mapping, radar unit 100 includes the detection of objects with zero velocity, i.e., objects 102a through d. Object detection, including sampling of radar detections at individual time points, is also performed for individual velocities, which are called velocity bins. Processing the data in this manner results in overflow into neighboring bins due to spectral leakage. Similarly, the detection of moving object 104 results in overflow into neighboring bins due to spectral leakage. If an object is moving sufficiently slowly, its detection will overflow into the same bin as the detection of a stationary object, and thus this velocity bin will contain the sum of overflows from the detection of stationary objects and the detection of low-velocity objects. Therefore, by using appropriate detection thresholds, such as the method proposed here, the variation in these stationary objects in the detection of stationary objects results in the observation of various objects and features of the scene over time due to this overflow information.

[0044] In addition to the above explanation, stationary radar units such as frequency-modulated continuous-wave radar are used to detect a single object for each individual range, i.e., for each range bin and each individual velocity bin. Therefore, even if more than one object is present in the same range at the same velocity, only one object will be detected.

[0045] An object is detected when its radar signal exceeds a detection threshold, and for each detected object, its individual range, individual radial velocity, azimuth angle, and signal strength may be measured. For each individual radius, all detections whose velocity falls within the zero velocity bin and all detections whose velocity falls within the adjacent velocity bins of the zero velocity bin are marked in the occupancy map.

[0046] If object 104 is moving slowly enough, then one or more of the combinations (R0, velocity zero), (R0, velocity +ΔV), and (R0, velocity -ΔV) will include both stationary and moving object detections. Since only one detection is performed for each combination, these combinations may not be the same as if the moving object 104 were not present. In other words, new information is added to the map.

[0047] As the moving target 104 moves with a given radius R0 and azimuth angle, the radar unit 100 will observe it at various times or frames, causing it to cast a shadow over the object, meaning that 102b and 102c will be behind object 104. Therefore, if the moving object is in front of stationary objects such as 102b and 102c, the reflections from these objects will be weakened, while other reflections from the same range will remain the same. Thus, even in this case, each combination of (velocity R0, zero), (velocity R0, +ΔV), and (velocity R0, -ΔV) contains new information that can be added to the map.

[0048] More specifically, Figure 1B shows a conceptual range-Doppler plot 200. The x-axis shows the Doppler velocity, where the radar signal measurements represent the center peak Pv and side lobe Sv. Furthermore, the y-axis shows the center peak Pr and side lobe Sr. Corresponding power graphs are shown, which represent the main lobe Pr and Pv in the Fourier transform of the collected detections. A detection is considered confirmed when the detection exceeds the detection threshold, that is, when the peak has an amplitude exceeding the detection threshold 220.

[0049] Figure 1C shows another range-Doppler plot 201, where two targets 210 and 212 are present. These targets have zero velocity and are therefore considered stationary. In other words, in the corresponding power-vs-velocity plot, similar to that shown in Figure 1B, both peaks have zero velocity. In the power-vs-range plot 202 shown in Figure 1C, the difference in range, i.e., the relative displacement along the y-axis along the zero-velocity line 203 in the range-Doppler plot, is reflected in the displacement of the corresponding peaks 212a and 210a in plot 202. Since peak 212a is lower than peak 210a, and only the highest peak is selected for a given range, object 212 is considered "invisible," i.e., not detected by radar. However, the overlap between lobes 212a and 210a and their side lobes indicate spectral leakage between two non-zero detections, which can lead to confirmed detection. Each lobe may be considered to represent a bin.

[0050] Looking at Figure 1D, we see that a moving object 214 is introduced. In the range-Doppler figure 205, the moving object 214 is shifted to the right in Figure 205, away from the zero velocity axis 203. Furthermore, the power-versus-Doppler velocity plot 204 shows the main lobes 210b, 212b, and 214a, as well as their relative displacements to each other. Again, overlaps between lobes indicate an overflow region. In particular, the side lobe 212c, which belongs to the detection of the stationary object 212 (which was undetectable as explained in relation to Figure 1C), and the side lobe 214b, which belongs to the detection of the moving object 214, overlap at 215, and the sum of these signals (not shown to avoid cluttering the drawing), i.e., the sum of the overlaps of lobes 212c and 214c, may exceed the detection threshold 220. Therefore, as explained in Figure 1B, not only the detection of objects with non-zero velocity performed within a given range, but also the detection of the combined lobes 212c and 214b, also fall into the non-zero velocity bin, thereby generating a confirmed detection of the previously "invisible" stationary object 212. The velocity of object 214 is low enough to cause an overflow into the same bin, i.e., into lobe 212c, which belongs to the non-zero detection of object 212. Furthermore, the velocity of object 214 is still low enough that the combined signal of lobes 212c and 214b exceeds the detection threshold 220.

[0051] As shown in Power vs. Velocity Figure 206 of Figure 1E, the velocity of object 214 was substantially higher, which may cause spectral leakage. The sum of overlapping lobes such as lobes 212c and 214c, as described above, does not cause a confirmed detection, i.e., a combined signal exceeding the threshold 220. Therefore, the velocity of object 214 is excessively high, causing spectral leakage, which may cause a confirmed detection where the velocity is not zero.

[0052] The above is achieved by thresholding the detection, so that detections where the radar signal strength exceeds the detection threshold and the velocity falls into the zero velocity bin are included, and detections where the radar signal strength exceeds the detection threshold but the velocity is not zero and is low enough to cause overflow information to be placed in the same bin as a detection that falls into the zero velocity bin, thereby ensuring that detections where the velocity is not zero are performed, i.e., detections in bins where the velocity is not zero.

[0053] Figure 2 is a flowchart of the method steps according to each embodiment of the present invention.

[0054] In step 102, a stationary scene is detected by continuously collecting radar signals over time using a set of receiver antennas of a stationary radar unit. The stationary radar unit 100 includes a set of receiver antennas configured to transmit radar signals toward scene 1 and to detect radar signals from any direction. As a result, the stationary radar unit 100 does not utilize beamforming. The stationary radar unit 100 is configured to measure target velocity in individual velocity bins, i.e., the velocity of detected targets is measured in individual steps.

[0055] In step 104, the occupancy map of the stationary scene is constructed using confirmed detections determined from the collected radar signals. A confirmed detection is one where the radar signal intensity exceeds the detection threshold, causing the velocity to fall into the zero velocity bin. Furthermore, confirmed detections also include detections where the radar signal intensity exceeds the detection threshold, resulting in a velocity that is not zero, and overflow information in the same bin, as a detection that falls into the zero velocity bin. As described above, this type of thresholding provides additional information for mapping the stationary scene compared to detecting only completely stationary objects.

[0056] Figure 3 conceptually illustrates the construction of the occupation map for Scene 1 by considering probability values ​​in the occupation grid 300. The occupation grid contains a grid of cells 302, where only one of the cells 302 is marked in Figure 3.

[0057] When a detection performed by the radar unit 100 is confirmed by the algorithm, the probability value in the corresponding cell of the grid is increased. For example, the cells marked in groups 302a to 302d are cells in the region where confirmed detections have occurred, and correspond to stationary objects 102a to 102d, respectively. Therefore, as further detections are performed on the same stationary object over time, the probability in the corresponding cell increases further, thereby increasing the likelihood of the stationary object being present in the corresponding region. Note that since the occupancy map 300 is still under construction, regions 302a to 302d are smaller than the stationary objects 102a to 102d.

[0058] Furthermore, to increase the contrast of the occupancy map, the probability is reduced in cells that fall within the region between the confirmed detection and the location of the radar unit 100. This further reduces the influence of moving objects appearing in the occupancy map 300. For example, at the first time step, the moving object 104 covers the stationary object 102b. At this point, the probability in region 304a increases, and since the stationary object 102b is blocked by the moving object 104, the probability in the region of object 302b does not change. However, the probability in the region between region 304a and the radar unit 100 decreases. Subsequently, the moving object 104 moves to a position that partially covers object 102c, and object 102b becomes detectable by the radar unit 100. Therefore, the probability value in the corresponding region 302b of the occupancy map 300 increases, and the probability in the region between region 302b and the radar unit 100 decreases. The detection made in region 304a fades out over time. As a result, the detection of stationary objects 102a through d will appear in the occupancy map 300.

[0059] Furthermore, the reduced area of ​​the occupied map preferably intersects the straight line 306 between a first area in the occupied map 300, such as area 302b, where detection has occurred, and the position of the radar 100.

[0060] Furthermore, the occupancy probability value is increased in neighboring regions of the first region. To reiterate, if we consider region 302b as the first region in the current frame, then when detection occurs in region 302b, the occupancy probability value may also be increased in region 303b, which is adjacent to region 302b.

[0061] The occupancy map 300 is constructed by collecting detections over a certain period of time. In other words, the occupancy map 300 is filled with matching detections, where only detections made immediately on moving objects are averaged out.

[0062] Figures 4A and 4B show examples of the constructed occupancy map 400 at two separate times. Figure 4A shows the occupancy map 400 at time t1, and Figure 4B shows the occupancy map 400 at time t2, approximately one minute apart. During that time, for example, a person exemplified by object 104 in Figures 1A and 3 is moving around the scene. In Figure 4A, only a small portion of stationary objects in the scene are detected and marked in black. As time progresses, the occupancy map 400 is formed using the method described with reference to Figure 3, for the mechanism described with reference to Figure 1A, and more of the scene can be progressively detected, i.e., more black areas appear, which may correspond to building walls, and more white areas appear, which may correspond to open areas such as lawns or empty parking spaces.

[0063] The radar unit 100 operates by performing detections at individual radii and individual velocities. While operating, one radar detection for each combination of an individual radius from the radar's position and a detection related to a neighboring velocity bin of the zero velocity bin is included in the occupied map 300. Looking at Figure 3 again, radius 308 conceptually represents one individual radius 308, which in the current frame occupies regions 302a and 304a. Here, the detection in region 302a corresponds to a confirmed detection of an object with zero velocity, i.e., one that falls within the zero velocity bin, and the detection in region 304a corresponds to a detection related to a neighboring velocity bin of the zero velocity bin.

[0064] Preferably, for each of a set of separate radii from the radar position, all confirmed detections are included in the occupancy map.

[0065] As an example, once an occupancy map 300 or 400 is constructed, the azimuth angle of a confirmed detection is determined. The azimuth angle α is the angle along the individual radius 308 where the confirmed detection occurred. During the radar detection process, a Fourier transform, preferably a discrete Fourier transform, of the phase difference between the radar signals is calculated. This generally yields a spectrum of various angles versus power, as conceptually shown in Figure 5, which includes a global peak 502, i.e., the peak with the highest energy content, and a set of side lobes 504. Determining the azimuth angle α may be done by calculating the discrete Fourier transform of the phase difference between the radar signals collected by various receiver antennas divided along a first axis 310 parallel to the azimuth plane, and selecting the frequency, i.e., the azimuth angle, with the highest energy content in that discrete Fourier transform. It may also be possible to simply select a frequency higher than the one with the highest energy content, however, in a preferred embodiment, the frequency, i.e., the azimuth angle, with the highest energy content is selected. This is because it can be difficult to accurately determine whether a peak is the actual target or merely a sidelobe. Similar selections may be made for range detection and velocity detection. See also the explanations related to Figures 1A to 1E.

[0066] Similarly, even when elevation angle is considered, constructing an occupancy map to provide a radar image along the elevation angle involves determining the elevation angle of a confirmed detection by selecting the frequency with the highest energy content in the Fourier transform of the phase difference between radar signals collected by various receiver antennas divided along a second axis perpendicular to the azimuth plane.

[0067] This is done in a similar manner to the one explained in relation to azimuth.

[0068] Figure 6 is a block diagram of a radar system 600 including a stationary radar and a control unit 602. The number of receiver antennas 604a to h of the stationary radar unit is not greater than 16, preferably not greater than 10, and more preferably not greater than 8. Eight receiver antennas are shown here, representing a relatively simple and low-cost radar unit 100. The stationary radar unit 100 may be operated to transmit radar signals in all directions. Preferably, the radar unit 100 operates without utilizing beamforming.

[0069] The channels, i.e., antennas 604a to h, may be physical or virtual antennas. Antennas 604a to h may be implemented as virtual antennas for a multiple-in, multiple-out (MIMO) radar. Receiver circuit 606 is adapted to receive radar signals from antennas 604a to h and convert them into signals that may be acquired by control unit 602. Such receiver circuit may include, for example, a mixer, a filter, an analog-to-digital converter, etc. Such receiver circuit 606 may have various components and configurations depending on the specific implementation being considered.

[0070] The control unit 602 is configured to detect stationary scenes by acquiring data showing radar signals collected continuously over time. The radar signals are detected by antennas 604a to h.

[0071] Furthermore, the control unit 602 is configured to construct an occupancy map of stationary scenes using confirmed detections determined from the acquired data. As described above, confirmed detections are those where the radar signal intensity exceeds the detection threshold and the velocity falls into the zero velocity bin, and those where the radar signal intensity exceeds the detection threshold and the velocity is not zero, but is low enough to cause overflow information to be placed in the same bin as a detection that falls into the zero velocity bin.

[0072] The control unit includes a microprocessor, microcontroller, programmable digital signal processor, or other programmable device. The control unit also, or alternatively, includes an application-specific integrated circuit, programmable gate array or programmable array logic, programmable logic device, or digital signal processor. The control unit includes programmable devices such as the aforementioned microprocessor, microcontroller, or programmable digital signal processor. The processor may further include computer-executable code that controls the operation of the programmable device.

[0073] The control functions of the Disclosure may be implemented using an existing computer processor, or by a dedicated computer processor incorporated for this or another purpose for an appropriate system, or by a hardwired system. Embodiments within the scope of the Disclosure include program products including machine-readable media for executing or having machine-executable instructions or data structures stored therein. Such machine-readable media may be any available media accessible by a general-purpose or dedicated computer or other machine having a processor. For illustrative purposes, such machine-readable media may include RAM, ROM, EPROM, EEPROM, CD-ROM, or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other media accessible by a general-purpose or dedicated computer or other machine having a processor that can be used to execute or store desired program code in the form of machine-executable instructions or data structures. When information is transferred to or provided to a machine via a network or another communication connection (either hardwired, wireless, or a combination of hardwired and wireless), the machine correctly views that connection as machine-readable media. Thus, any such connection is correctly referred to as machine-readable media. The above combinations also fall within the scope of machine-readable media. Machine-executable instructions include, for example, instructions and data that cause a general-purpose computer, a dedicated computer, or a dedicated processing machine to perform a specific function or group of functions.

[0074] While drawings may show sequences, the order of steps may differ from those depicted. Furthermore, two or more steps may be performed simultaneously, or partially simultaneously. Such variations will depend on the selected software and hardware systems, as well as the designer's choices. All such variations are within the scope of this disclosure. Similarly, software implementations can be achieved using standard programming techniques involving rule-based logic and other logic to accomplish various connection, processing, comparison, and decision steps. Additionally, while the present invention has been described with reference to its specifically illustrated embodiments, many variations and modifications will be apparent to those skilled in the art.

[0075] Furthermore, variations to the disclosed embodiments can be understood and achieved by an addressee skilled in the art in the practice of the patented invention by examining the drawings, the disclosure, and the appended claims. Furthermore, in the claims, the term “comprising” does not exclude other elements or steps. The indefinite articles “a” or “an” do not exclude the plural.

Claims

1. A method for mapping stationary objects in a stationary scene (1) using a stationary radar unit (100) adapted to detect signals from all directions without the use of beamforming or sweeping, the stationary radar unit is not moving relative to the stationary scene and is operative to transmit radar signals towards the stationary scene, the stationary radar unit including a set of receiver antennas (102a to h) configured to detect radar signals from any direction, the stationary radar unit configured to measure target speeds in discrete speed bins, the method comprising: Successively collecting radar signals over time using the set of receiver antennas of the stationary radar unit to detect stationary and moving objects in the stationary scene (S102); Constructing (S104) an occupancy map (300, 400) of stationary objects in the stationary scene using confirmed detections determined from the collected radar signals; Including, the confirmed detections being detections where the radar signal strength exceeds a detection threshold and the velocity falls in a zero velocity bin, and detections where the radar signal strength is low enough to result in spillover information in the same bin as detections where the radar signal strength exceeds the detection threshold and the non-zero velocity falls in the zero velocity bin. method.

2. The method of claim 1 , wherein the number of receiver antennas does not exceed sixteen.

3. 2. The method of claim 1 , wherein constructing the occupancy map includes including in the occupancy map, over time, all detections where the radar signal strength exceeds the detection threshold and has no velocity, and all detections where the radar signal strength exceeds the detection threshold and has a non-zero velocity, for each of a set of discrete radii from the radar location (S106).

4. 2. The method of claim 1, comprising: constructing the occupancy map by, for each of the confirmed detections, increasing an occupancy probability value in a corresponding first region (302b) of the occupancy map and decreasing a probability value in a region of the occupancy map between the first region and the radar location.

5. 5. The method of claim 4, wherein the area between the first area in the occupancy map and the radar location intersects a straight line (306) between the first area in the occupancy map and the radar location.

6. The method of claim 4, further comprising increasing the occupancy probability value in neighbouring regions (303b) of the first region.

7. The method of claim 4 , wherein the occupancy map is constructed as a collection of the probability values ​​over time.

8. The method of claim 1 , comprising including in the occupancy map one radar detection for each combination of distinct radii from the radar location and detections associated with speed bins neighboring the zero speed bin.

9. 2. The method of claim 1, wherein constructing the occupancy map includes determining the azimuth angles of the confirmed detections using selecting frequencies with the highest energy content in a Fourier transform of phase differences between radar signals collected by various of the receiver antennas split along a first axis.

10. 2. The method of claim 1, wherein constructing the occupancy map includes determining the elevation angles of the confirmed detections using selecting frequencies with the highest energy content in a Fourier transform of phase differences between radar signals collected by various of the receiver antennas split along a second axis.

11. The method of claim 1 , wherein the stationary radar unit is operative to transmit radar signals in all directions.

12. A non-transitory computer readable recording medium comprising a computer program comprising instructions that, when said program is executed by a computer, cause said computer to perform the method of any one of claims 1 to 11.

13. A control unit (602) for mapping stationary objects in a stationary scene (1) using radar signals collected by a stationary radar unit (100) adapted to detect signals from all directions without using beamforming or sweeping, said stationary radar unit not moving relative to said stationary scene and operative to transmit radar signals towards said stationary scene and detect radar signals from any direction using a set of receiver antennas (102a to h), said stationary radar unit configured to measure target speeds in discrete speed bins, said control unit: obtaining data indicative of radar signals continuously collected over time by the stationary radar unit to detect stationary and moving objects in a stationary scene; configured to construct an occupancy map of stationary objects in the stationary scene using confirmed detections determined from the acquired data; the confirmed detections being detections where the radar signal strength exceeds a detection threshold and the velocity falls in a zero velocity bin, and detections where the radar signal strength is low enough to result in spillover information in the same bin as detections where the radar signal strength exceeds the detection threshold and the non-zero velocity falls in the zero velocity bin. Control unit.

14. A radar system (600) comprising a stationary radar and a control unit according to claim 13.

15. 15. The radar system of claim 14, wherein the number of receiver antennas in the stationary radar unit does not exceed 16.

Images