Method and system for performing target detection

Abstract

A method for performing target detection, the method comprising: Transmitting frequency modulated continuous wave (FMCW) pulses as chirps from a radar system (110); Receiving reflections (145) resulting from the chirps; processing the reflections (145) to obtain a range chirp map (310) for each beam associated with the transmission (115); performing curve detection on the range chirp map (310) for each ray to obtain candidate curves; Detecting one or more targets (140) based on the curve detection; and Performing a Doppler FFT on the candidate curves obtained when performing the curve detection, wherein detecting the one or more targets (140) is based on applying a threshold to a result of the Doppler FFT.

Description

INTRODUCTION

[0001] The subject matter of the disclosure relates to target detection based on curve detection in a range chirp map.

[0002] Radar systems transmit energy in the radio frequency (HF) band and process received signals resulting from targets that reflect the RF energy. Radar systems are used in a wide range of applications for target detection and tracking. For example, vehicles (e.g., automobiles, trucks, construction equipment, agricultural machinery, automated factory equipment) are increasingly equipped with sensor systems, such as radar systems, that enable enhanced or automated vehicle operation. For example, target detection and tracking by the radar system can be used for autonomous operation, adaptive cruise control, automated steering or braking, or driver warning systems. In a frequency-modulated continuous wave (FMCW) radar system, a series of pulses of increasing or decreasing frequency, called chirps, can be transmitted.A shift in the frequencies of the received reflections from the transmitted frequencies results from the relative motion of the reflecting target and is referred to as Doppler shift. Traditionally, information from the FMCW radar system is accounted for using a range-Doppler map. In a range-Doppler map, the range lies along one axis and the Doppler lies along a perpendicular axis. The traditional detection technique using a range-Doppler map can be ineffective at detecting fainter targets and requires separate steps for target detection and classification. Accordingly, it is desirable to provide target detection based on curve detection in a range-chirp map.

[0003] US 2012 / 0093359 A1 discloses a method for target tracking and target discrimination in dense detection environments using batch detection processing. DE 10 2009 016 479 A describes a radar system for use in driver assistance systems in a motor vehicle, wherein the radar system includes a method for preventing incorrect reactions caused by interference. SUMMARY

[0004] In a first aspect, a method for target detection comprises transmitting frequency-modulated continuous wave (FMCW) pulses as chirps from a radar system. The method also comprises receiving reflections resulting from the chirps and processing the reflections to obtain a range chirp map for each beam associated with the transmission. The method further comprises performing curve detection on the range chirp map for each beam and detecting one or more targets based on the curve detection. The method further comprises performing a Doppler FFT on the candidate curves obtained from performing the curve detection, wherein detecting the one or more targets is based on applying a threshold to a result of the Doppler FFT.

[0005] In addition to one or more of the features described herein, processing the reflections includes performing an analog-to-digital conversion of the reflections to obtain sample values.

[0006] In addition to one or more of the features described herein, processing the reflections also includes obtaining a range chirp map for each channel of the radar system.

[0007] In addition to one or more of the features described herein, obtaining the range chirp map includes performing a fast Fourier transform to obtain an indication of the energy distribution of the reflections at each detectable area associated with each of the chirps.

[0008] In addition to one or more of the features described herein, processing the reflections further includes performing digital beamforming on each range chirp map for each channel to obtain the range chirp map for each beam.

[0009] In addition to one or more of the features described herein, performing digital beamforming includes determining an azimuth angle to each of the one or more targets.

[0010] In addition to one or more of the features described herein, performing curve detection involves using a Hough transform on the range chirp map for each ray.

[0011] In addition to one or more of the features described herein, performing curve detection involves iteratively processing the range chirp map for each ray.

[0012] In addition to one or more of the features described herein, the detection of the one or more targets when applying a threshold is based on a result of the Doppler FFT.

[0013] In another aspect, a system for target detection includes a radar system for transmitting frequency-modulated continuous wave (FMCW) pulses as chirps from a radar system and receiving reflections resulting from the chirps. The system also includes processing circuitry for processing the reflections to obtain a range chirp map for each beam associated with the transmission, performing curve detection on the range chirp map for each beam, and detecting one or more targets based on the curve detection. The processing circuitry is further configured to perform a Doppler FFT on the candidate curves obtained from performing the curve detection and detecting the one or more targets based on applying a threshold to a result of the Doppler FFT.

[0014] In addition to one or more of the features described herein, the processing circuitry processes the reflections based on performing an analog-to-digital conversion of the reflections to obtain sample values.

[0015] In addition to one or more of the features described herein, the processing circuitry processes the reflections based on obtaining a range chirp map for each channel of the radar system.

[0016] In addition to one or more of the features described herein, the processing circuitry obtains the range chirp map based on performing a fast Fourier transform to obtain an indication of the energy distribution of the reflections in each detectable area associated with each of the chirps.

[0017] In addition to one or more of the features described herein, the processing circuitry processes the reflections based on performing digital beamforming on each range chirp map for each channel to obtain the range chirp map for each beam.

[0018] In addition to one or more of the features described herein, the processing circuit performs curve detection based on the use of a Hough transform on the range chirp map for each ray.

[0019] In addition to one or more of the features described herein, the processing circuit performs curve detection based on iteratively processing the range chirp map for each beam.

[0020] In addition to one or more of the features described herein, the processing circuit detects the one or more targets based on applying a threshold to a result of the Doppler FFT.

[0021] In addition to one or more of the features described herein, the radar system and processing circuitry are located in a vehicle.

[0022] The above features and advantages, as well as other features and functions of the present disclosure, will be readily apparent from the following detailed description when taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Other features, advantages and details appear only by way of example in the following detailed description of the embodiments, which detailed description refers to the drawings, where: Fig. 1 is a block diagram of a processing circuit for performing target detection based on curve detection in a range chirp map obtained with a radar system in a vehicle according to one or more embodiments; Fig. 2 is a process flow of a method for performing target detection based on curve detection according to one or more embodiments; Fig. Figure 3 shows components of a range chirp map generated according to one of the Fig. 2 processes shown; Fig. 4 is a process flow outlining the curve detection process according to one or more embodiments; Fig. 5 illustrates exemplary curves detected in an exemplary range chirp map according to one or more embodiments; and Fig. Figure 6 shows an exemplary curve spectrum obtained when performing a Doppler FFT according to one of the Fig. 4 processes shown are recognized curves. DETAILED DESCRIPTION

[0024] The following description is merely exemplary in nature and is not intended to limit the present disclosure in its applications or uses. It should be understood that throughout the drawings, corresponding reference characters designate like or corresponding parts and features.

[0025] As mentioned above, an FMCW radar transmits chirps and develops a range-Doppler map from the received reflections. Typical processing of received reflections involves performing an analog-to-digital conversion and a fast Fourier transform (FFT) with respect to range (referred to as the range FFT). The result of the range FFT is a display of the energy distribution across the radar-detectable distances for each transmitted chirp, with a different range FFT associated with each receive channel and each transmit channel. Thus, the total number of range FFTs is a product of the number of transmitted chirps and the number of receive channels.

[0026] A Doppler FFT is then applied to the range FFT result. The Doppler FFT is also a well-known technique in radar detection and is used to obtain a range-Doppler map for each receive channel. For each receive channel and transmit channel pair, all chirps for each range bin of the range chirp map (obtained with the range FFT) are processed together. The result of the Doppler FFT, the range-Doppler map, displays the relative velocity of each detected target along with its range. The number of Doppler FFTs is a product of the number of range bins and the number of receive channels.

[0027] Digital beamforming results in a range-Doppler (relative velocity) map for each beam. Digital beamforming is also a well-known process and involves obtaining a vector of complex scalars from the vector of received signals and the matrix of actual received signals at each receiving element for each angle of incidence of a target reflection. Digital beamforming provides an azimuth angle to each of the detected targets based on a threshold of the complex scalars of the obtained vector. The outputs ultimately obtained by processing the received signals are the range, Doppler, azimuth, altitude, and amplitude of each target.

[0028] As also mentioned, detection based on the range-Doppler map has limitations, for example, with respect to detecting fainter targets. Thus, according to one or more embodiments, the range-chirp map obtained from the range FFT is used in the detection process before performing the Doppler FFT. Furthermore, curve detection is performed on the range-chirp map before performing the Doppler FFT. As a result, the curves generated by curve detection are subjected to a Doppler FFT, and curves with Doppler values ​​that exceed a detection threshold are used to detect targets. This modification of the detection algorithm facilitates the detection of fainter targets and the detection of multiple returns from the same target as if they were associated with a single object.

[0029] According to an exemplary embodiment, Fig. 1 is a block diagram of the processing circuit 120 for performing target detection based on curve detection in a range chirp map obtained with a radar system 110 in a vehicle 100. In the Fig. In the exemplary case illustrated in Figure 1, the vehicle 100 is an automobile 101. The radar system 110 is a known FMCW radar that includes, among other known components, one or more transmitters and one or more receivers. The transmission 115 of a chirp and a reflection 145 resulting from a target 140 are shown in Fig. 1. The transmission 115 is a frequency-modulated continuous wave signal. The chirp frequency of the transmission 115 is shifted by the target 140 in the reflection 145 according to the Doppler effect. The magnitude of the shift is a function of the distance from the radar system 110 to the target 140.

[0030] Processing circuitry 120 may be part of radar system 110 or part of a separate controller. In addition to receiving information from radar system 110, processing circuitry 120 provides target information to other vehicle systems 130, such as a collision avoidance system, an adaptive cruise control system, an automatic braking system, or an autonomous driving system. Processing circuitry 120 may include an application-specific integrated circuit (ASIC), an electronic circuit, a processor (shared, dedicated, or group processor), and memory executing one or more software or firmware programs, combinational logic circuitry, and / or other suitable components that provide the described functionality.

[0031] Fig. 2 is a process flow of a method for performing target detection based on curve detection according to one or more embodiments. At block 210, receiving reflections 145 at radar system 110 is based on transmission 115. According to alternative embodiments, radar system 110 may be a multi-input-multi-output (MIMO) system with more than one transmit channel and more than one receive channel, or may include only one transmit or receive channel. Performing an analog-to-digital conversion (ADC) at block 220 of the reflections 145 results in sampling of reflections 145.

[0032] At block 230, the processes include performing the range FFT on the samples output by the ADC at block 220. Since a specific shift in the frequency of the reflection 145 relative to the transmission 115 is associated with a particular range to the target 140, the frequency of each received reflection 145 can be associated with a range. As previously mentioned, the process for performing a range FFT is known. Fig. Figure 3 illustrates components of a range chirp map 310 obtained according to the process at block 230. The output of block 230 is a range chirp map 310 per receive channel. That is, for each receive channel, the energy for each range bin resulting from each chirp is displayed. In Fig. 3, axis 320 indicates the energy level, axis 330 indicates the distance bin, and axis 340 indicates the chirp index.

[0033] According to one or more embodiments herein, digital beamforming is performed during this phase, rather than a Doppler FFT as in conventional radar systems. Performing digital beamforming at block 240 refers to the known process of estimating the angle to the target 140 from the individual transmitter or the center of a transmitter array. As previously mentioned, the digital beamforming at block 240 results in a range chirp map 310 per beam, not per receive channel.

[0034] Performing the curve detection at block 250 is done with respect to Fig. 4. The result of the curve detection performed at block 250 is provided, for example, to other vehicle systems 130 as information at block 260. The result includes range, Doppler, azimuth, altitude, and amplitude.

[0035] Fig. 4 is a process flow breaking down the curve detection process at block 250 according to one or more embodiments. Fig. The processes illustrated in Figure 4 are performed iteratively per ray on each range chirp map 310. At block 410, a ray is selected for each iteration. At block 420, performing Hough curve detection refers to fitting the curve 510 that best matches the shape in the range chirp map 310 for the selected ray. Fig. 5 illustrates exemplary curves 510 captured in an exemplary range chirp map 310. The axis 330 shows the range, the axis 340 shows the chirp index, and the energy level is indicated by shading. The process at block 420 yields the candidate curves 510 ( Fig. 5). Performing the known curve detection technique at this stage differs from using the range-Doppler map in the conventional detection technique, and the physical property of the target 140 is represented by a curve 510 rather than a point. The actual range of the target 140 over time can be determined by detecting according to the Fig. 4, while the range range is determined by additional processing beyond the target detection processing according to conventional target detection methods using the range-Doppler map. Performing the Doppler FFT at block 430 after the Hough curve detection at block 420 results in a curve spectrum 610 ( Fig. 6). As mentioned above, the Doppler FFT provides an indication of the relative velocity of each target.

[0036] Fig. 6 shows an exemplary waveform spectrum 610 resulting from performing a Doppler FFT at block 430 on the waveforms 510 (from the range chirp map 310) acquired according to the method at block 420. The range bin is indicated by axis 330 and the Doppler by axis 620. At block 440, applying a threshold 630 to the waveform spectrum 610 obtained from the Doppler FFT at block 430 facilitates target detection for the selected beam. An exemplary target frequency response 640 is shown in Fig. 6 shown.

[0037] As already mentioned, the provisions made with reference to the Fig. 2 and Fig.4, the processes described above allow the detection of fainter targets that may be obscured by targets with higher reflectivity when using conventional detection techniques based on the range-Doppler map. Target separation is also improved by detection based on the range-chirp map. If the actual target range rate determined by the detection processes does not match the target Doppler, the target Doppler frequency is ambiguous. If a curve 510 is exactly the same as a narrower curve 510, then the curve 510 is a multipath reflection. Multiple reflections 145 from the same target 140 may be detected as being associated with the same object because they share the same characteristics of the curve 510. Finally, target detection and classification may be performed jointly at block 440, according to one or more embodiments.

[0038] While the above disclosure has been described with reference to exemplary embodiments, those skilled in the art will understand that various changes may be made and corresponding parts may be substituted for the individual parts without departing from the scope of the disclosure. Furthermore, many modifications may be made to adapt a particular material situation to the teachings of the disclosure without departing from the essential scope thereof. Therefore, it is intended that the present disclosure not be limited to the specific embodiments disclosed, but will include all embodiments falling within the scope thereof.

Claims

[1] Method for performing target detection, wherein the method comprises: Transmission of frequency-modulated continuous wave (FMCW) pulses as chirps from a radar system (110); Receiving reflections (145) resulting from the chirps; Processing the reflections (145) to obtain a distance chirp map (310) for each ray assigned to the transmission (115); Performing curve detection on the distance chirp map (310) for each ray to obtain candidate curves; Detection of one or more targets (140) based on curve detection; and Performing a Doppler FFT on the candidate curves obtained during curve detection, wherein the detection of one or more targets (140) is based on applying a threshold to a result of the Doppler FFT. [2] Method according to claim 1, wherein the processing of the reflections (145) includes performing an analog-to-digital conversion of the reflections (145) to obtain sample values. [3] Method according to claim 1, wherein the processing of the reflections (145) further comprises obtaining a range chirp map (310) for each channel of the radar system (110), the obtaining of the range chirp map (310) comprises performing a fast Fourier transform to obtain a display of the energy distribution of the reflections (145) at each detectable area associated with each of the chirps, and the processing of the reflections (145) further comprises performing digital beam shaping on each range chirp map (310) for each channel to obtain the range chirp map (310) for each beam. [4] Method according to claim 1, wherein performing the curve detection involves iteratively applying a Hough transform to the distance chirp map (310) for each beam. [5] System for performing target detection, the system comprising: a radar system (110) configured to transmit frequency-modulated continuous wave (FMCW) pulses as chirps from a radar system (110) and to receive reflections (145) resulting from the chirps; Processing circuit (120) configured to process the reflections (145) to obtain a distance chirp map (310) for each beam associated with the transmission (115), to perform curve detection on the distance chirp map (310) for each beam to obtain candidate curves, and to identify one or more targets (140) based on the curve detection, wherein the processing circuit (120) is further configured to perform a Doppler FFT on the candidate curves obtained from performing the curve detection, and to identify the one or more targets (140) based on applying a threshold to a result of the Doppler FFT. [6] System according to claim 5, wherein the processing circuit (120) is further configured to process the reflections (145) based on obtaining a range chirp map (310) for each channel of the radar system (110), to obtain the range chirp map (310) based on performing a fast Fourier transform, to obtain a display of the energy distribution of the reflections (145) at each detectable area associated with each of the chirps, and to process the reflections (145) based on performing digital beam shaping at each range chirp map (310) for each channel, to obtain the range chirp map (310) for each beam. [7] System according to claim 5, wherein the processing circuit (120) is configured to iteratively perform curve detection based on the use of a Hough transform on the distance chirp map (310) for each beam. [8] System according to claim 5, wherein the radar system (110) and the processing circuit (120) are located in a vehicle (100).

Images