Techniques for automatically adjusting a detection threshold of an fmcw lidar

Abstract

A method of adjusting a detection threshold in a frequency-modulated continuous-wave (FMCW) light detection and ranging (LIDAR) system includes determining a first confidence threshold for detecting a first target from a plurality of targets within a frequency range, where the frequency range includes different frequencies corresponding to the targets. The method also includes determining a subset of frequencies within the frequency range for detecting a second target. The second target transmits a signal within the subset of frequencies that is below the first confidence threshold. The method further includes adjusting the first confidence threshold to a second confidence threshold at the subset of frequencies for detecting the second target within the subset of frequencies and restoring the second confidence threshold to the first confidence threshold outside the subset of frequencies for detecting the first target.

Description

[0001] Related applications

[0002] This application claims the benefit of U.S. Patent Application 17 / 339,737, filed June 4, 2021, pursuant to 35 U.S. SC §119(e), which claims priority to U.S. Provisional Patent Application 63 / 093,621, filed October 19, 2020, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This application generally relates to optical detection and ranging (LIDAR) systems, and more particularly to adjusting the detection threshold for target detection in frequency modulated continuous wave (FMCW) LIDAR systems. Background Technology

[0004] Frequency-modulated continuous wave (FMCW) LiDAR systems use a tunable infrared laser to chirply illuminate the target, and a coherent receiver to detect backscattered or reflected light from the target, combined with a local copy of the transmitted signal. The local copy is mixed with the echo signal, which has a round-trip time delay to the target and back, generating a signal at the receiver whose frequency is proportional to the distance to each target in the system's field of view. For human safety reasons, a low-power laser must be used, resulting in very low signal strength from reflections from objects. The range and accuracy of a LiDAR system are functions of the signal-to-noise ratio, and traditional solutions cannot reliably detect targets with weak echo signals while limiting the detection of false targets. Summary of the Invention

[0005] This disclosure describes examples of systems and methods for automatically adjusting detection thresholds for target detection. A Light Detection and Ranging (LIDAR) system includes: an optical scanner that sends a light beam toward a target and receives an echo signal from the target; an optical processing system coupled to the optical scanner to generate a baseband signal in the time domain based on the echo signal, the baseband signal including frequencies corresponding to a LIDAR target range; and a signal processing system coupled to the optical processing system. The signal processing system includes: a processor; and a memory operatively coupled to the processor, the memory storing instructions that, when executed by the processor, cause the LIDAR system to: determine a first confidence threshold for detecting a first target from a plurality of targets within a frequency range, wherein the frequency range includes a plurality of different frequencies corresponding to the plurality of targets. The processor can also determine a frequency subset within the frequency range for detecting a second target, wherein the second target transmits a signal below the first confidence threshold within the frequency subset; adjust the first confidence threshold to a second confidence threshold at the frequency subset for detecting the second target within the frequency subset; and restore the second confidence threshold to the first confidence threshold outside the frequency subset for detecting the first target.

[0006] In some embodiments, to determine the first confidence threshold, the processor further: determines a noise estimate across the frequency range; and calculates the difference between the amplitude of the frequency and the noise estimate across the frequency range. In some embodiments, the first confidence threshold is based on at least one of a likelihood ratio and a signal-to-noise ratio providing confidence values ​​for target detection. In some embodiments, to determine the subset of frequencies within the frequency range for detecting the second target, the processor: determines the subset of frequencies based on the estimated location of the second target. In some embodiments, the processor further: identifies previous detections of the second target in the time or frequency domain; and determines the estimated location of the second target based on the previous detections of the second target.

[0007] In some embodiments, to determine the estimated location of the second target, the processor: identifies point cloud information associated with the second target; and determines the estimated location of the second target based on the point cloud information. In some embodiments, the frequency subset includes a predefined frequency range corresponding to previously detected locations of the second target. In some embodiments, to adjust a first confidence threshold, the processor reduces the first confidence threshold to a second confidence threshold within the frequency subset, wherein the frequency subset corresponds to the estimated location of the second target. In some embodiments, the processor further determines the frequency subset corresponding to the estimated location of the second target based on a confidence value associated with the estimated location of the second target. In some embodiments, the processor further: determines whether one or more previous detections of the second target include a weak detection signal; determines the estimated location of the second target; and adjusts the first confidence threshold in response to determining that one or more previous detections of the target include a weak detection signal.

[0008] In some embodiments, a method includes: determining a first confidence threshold for detecting a first target from a plurality of targets within a frequency range, wherein the frequency range includes different frequencies corresponding to the target; and determining a subset of frequencies within the frequency range for detecting a second target, wherein the second target transmits a signal below the first confidence threshold within the frequency subset. The method further includes: adjusting the first confidence threshold to a second confidence threshold at the frequency subset for detecting the second target within the frequency subset; and restoring the second confidence threshold to the first confidence threshold outside the frequency subset for detecting the first target. In some embodiments, a non-transitory computer-readable medium includes instructions that, when executed by a processing means of a LIDAR system, cause the processing means of the LIDAR system to determine a first confidence threshold for detecting a first target from a plurality of targets within a frequency range, wherein the frequency range includes different frequencies corresponding to the target. The processor also determines a frequency subset within a frequency range for detecting a second target, wherein the second target sends a signal below a first confidence threshold within the frequency subset, adjusts the first confidence threshold to a second confidence threshold at the frequency subset for detecting the second target, and restores the second confidence threshold to the first confidence threshold outside the frequency subset for detecting the first target. Attached Figure Description

[0009] To gain a more complete understanding of the various examples, please now refer to the following detailed description relating to the accompanying drawings, in which similar reference numerals correspond to similar elements:

[0010] Figure 1 This is a block diagram illustrating an example LIDAR system according to this disclosure;

[0011] Figure 2 This is a time-frequency diagram illustrating an example of a LIDAR waveform according to this disclosure;

[0012] Figure 3A This is a block diagram illustrating an example LIDAR system according to this disclosure;

[0013] Figure 3B This is a block diagram illustrating an electro-optical system according to the present disclosure;

[0014] Figure 4 This is a block diagram of an example signal processing system according to the present disclosure;

[0015] Figure 5 This is a block diagram of an example signal processing system for peak detection using threshold adjustment at the estimated target location, according to this disclosure;

[0016] Figure 6A This is a signal amplitude-frequency diagram illustrating an example method for peak detection according to this disclosure;

[0017] Figure 6B This is an example of a signal amplitude-frequency diagram illustrating the comparison between the noise estimate and the difference between the signal spectrum and the noise estimate according to this disclosure;

[0018] Figure 7 This is a signal confidence-frequency plot illustrating an example method for threshold adjustment for peak detection according to this disclosure;

[0019] Figure 8 This is a flowchart illustrating a method for noise calibration and target detection according to this disclosure; and

[0020] Figure 9 This is a block diagram of an example signal processing system according to the present disclosure. Detailed Implementation

[0021] This disclosure describes various examples of LIDAR systems and methods for automatically adjusting the detection threshold of a LIDAR system to improve target detection and reduce false detections. According to some embodiments, the described LIDAR system can be implemented in any sensing market, such as, but not limited to, transportation, manufacturing, metrology, medical, virtual reality, augmented reality, and security systems. According to some embodiments, the described LIDAR system can be implemented as part of the front end of a frequency modulated continuous wave (FMCW) device that assists automated driver assistance systems or autonomous vehicles in spatial perception.

[0022] The lidar system described in this embodiment includes coherent scanning technology to detect signals returned from a target, thereby generating a coherent heterodyne signal from which target range and velocity information can be extracted. The signal can be converted into one or more frequency intervals, each interval having an amplitude at a relevant frequency within that interval. In some scenarios, target detection may correspond to a large amplitude (i.e., a peak) between one or more frequency intervals. According to some embodiments, the lidar system can use a detection threshold to determine whether a peak has a sufficiently large amplitude to correspond to a detected target. In some scenarios, the detection threshold can be increased to reduce false target detection. On the other hand, the detection threshold can be set low to increase the probability of detecting weaker signals. Using the techniques described herein, embodiments of this disclosure can address the above problems by reducing the detection threshold in small frequency bands near the estimated location of the target. Therefore, for targets returning weak signals, the detection probability increases, while maintaining a high detection threshold for the remaining frequency domains to limit false detections.

[0023] Figure 1 An example of a LIDAR system 100 implemented according to this disclosure is illustrated. The LIDAR system 100 includes one or more of a plurality of components, but may include more than one of them. Figure 1 The components shown are fewer or additional. According to some embodiments, they can be implemented on a photonic chip. Figure 1 The image depicts one or more components. Optical path 101 may include a combination of active and passive optical components. Active optical components may generate, amplify, and / or detect optical signals, etc. In some examples, active optical components include light beams of different wavelengths and include one or more optical amplifiers, one or more optical detectors, etc.

[0024] Free-space optics 115 may include one or more optical waveguides to carry optical signals and route and manipulate them to appropriate input / output ports of an active optical path. Free-space optics 115 may also include one or more optical components, such as taps, wavelength division multiplexers (WDMs), beam splitters / combiners, polarization beam splitters (PBSs), collimators, couplers, non-reciprocal elements such as Faraday rotators, etc. In some examples, free-space optics 115 may include components for changing polarization states and guiding received polarized light to an optical detector using a PBS. Free-space optics 115 may also include diffraction elements to deflect beams of different frequencies at different angles along an axis (e.g., the fast axis).

[0025] In some examples, the LIDAR system 100 includes an optical scanner 102, which includes one or more scanning mirrors rotatable along an axis orthogonal or substantially orthogonal to the fast axis of the diffraction element (e.g., the slow axis) to guide optical signals to scan the environment according to a scanning pattern. For example, the scanning mirrors may be rotated by one or more galvanometers. The optical scanner 102 also collects light incident on any object in the environment into an echo beam, which is returned to passive optical path components of the optical path 101. For example, the echo beam may be guided to an optical detector via a polarization beam splitter. In addition to mirrors and galvanometers, the optical scanner 102 may also include components such as quarter-wave plates, lenses, anti-reflective coated windows, etc. To control and support the optical path 101 and the optical scanner 102, the LIDAR system 100 includes a LIDAR control system 110. The LIDAR control system 110 may include processing means such as a signal processing unit 112. In some examples, the signal processing unit 112 may be one or more general-purpose processing devices, such as microprocessors, central processing units, etc. More specifically, the signal processing unit 112 may be a Complex Instruction Set Computing (CISC) microprocessor, a Reduced Instruction Set Computer (RISC) microprocessor, a Very Long Instruction Word (VLIW) microprocessor, or a processor implementing other instruction sets, or a processor implementing a combination of instruction sets. The signal processing unit 112 may also be one or more special-purpose processing devices, such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), network processors, etc.

[0026] In some examples, the signal processing unit 112 is a digital signal processor (DSP). The LIDAR control system 110 is configured to output digital control signals to control the optical driver 103. In some examples, the digital control signals can be converted into analog signals by the signal conversion unit 106. For example, the signal conversion unit 106 may include a digital-to-analog converter. The optical driver 103 can then provide drive signals to the active optical components of the optical path 101 to drive light sources such as lasers and amplifiers. In some examples, several optical drivers 103 and signal conversion units 106 can be provided to drive multiple light sources.

[0027] The LIDAR control system 110 is also configured to output digital control signals for the optical scanner 102. The motion control system 105 can control the galvanometers of the optical scanner 102 based on the control signals received from the LIDAR control system 110. For example, a digital-to-analog converter (DAC) can convert coordinate routing information from the LIDAR control system 110 into signals that can be interpreted by the galvanometers in the optical scanner 102. In some examples, the motion control system 105 can also return information to the LIDAR control system 110 regarding the position or operation of components of the optical scanner 102. For example, an DAC can further convert information about the galvanometer position into signals that can be interpreted by the LIDAR control system 110.

[0028] The LIDAR control system 110 is also configured to analyze incoming digital signals. In this regard, the LIDAR system 100 includes an optical receiver 104 for measuring one or more beams received by the optical path 101. For example, a reference beam receiver may measure the amplitude of a reference beam from an active optical component, and an analog-to-digital converter converts the signal from the reference receiver into a signal interpretable by the LIDAR control system 110. A target receiver measures an optical signal in the form of a beat-modulated optical signal carrying information related to the range and velocity of the target. The reflected beam may be mixed with a second signal from a local oscillator. The optical receiver 104 may include a high-speed analog-to-digital converter to convert the signal from the target receiver into a signal interpretable by the LIDAR control system 110. In some examples, the signal from the optical receiver 104 may be signal-conditioned by a signal conditioning unit 107 before being received by the LIDAR control system 110. For example, the signal from the optical receiver 104 may be provided to an operational amplifier to amplify the received signal, and the amplified signal may be provided to the LIDAR control system 110.

[0029] In some applications, the LIDAR system 100 may additionally include one or more imaging devices 108 configured to capture images of the environment, a Global Positioning System 109 configured to provide the system's geographic location, or other sensor inputs. The LIDAR system 100 may also include an image processing system 114. The image processing system 114 may be configured to receive images and geographic locations, and to transmit the images and locations, or related information, to the LIDAR control system 110 or other systems connected to the LIDAR system 100.

[0030] In some example operations, the LIDAR system 100 is configured to use a non-degenerate light source to simultaneously measure range and velocity across two dimensions. This capability allows for real-time, remote measurement of the range, velocity, azimuth, and elevation of the surrounding environment.

[0031] In some examples, scanning processing is initiated using an optical driver 103 and a LIDAR control system 110. The LIDAR control system 110 instructs the optical driver 103 to independently modulate one or more beams, and these modulated signals propagate through a passive optical path to a collimator. The collimator guides the light at the optical scanning system, which scans the environment in a pre-programmed pattern defined by the motion control system 105. The optical path 101 may also include a polarizing waveplate (PWP) to change the polarization of the light as it leaves the optical path 101. In some examples, the polarizing waveplate may be a quarter-wave plate or a half-wave plate. A portion of the polarized light may also be reflected back into the optical path 101. For example, the lens or collimation system used in the LIDAR system 100 may have natural reflective properties or a reflective coating to reflect a portion of the light back into the optical path 101. The optical signal reflected back from the environment is transmitted through the optical path 101 to a receiver. Since the polarization of the light has been changed, it can be reflected by a polarizing beam splitter along with the portion of the polarized light reflected back into the optical path 101. Therefore, the reflected light does not return to the same fiber or waveguide as the light source, but is reflected to a separate optical receiver. These signals interfere with each other and generate a combined signal. The individual beams of signal returning from the target produce time-shifted waveforms. The time phase difference between the two waveforms generates a beat frequency measured on the optical receiver (photodetector). The combined signal can then be reflected back to optical receiver 104.

[0032] The analog signal from the optical receiver 104 is converted into a digital signal using an ADC. The digital signal is then sent to the LIDAR control system 110. The signal processing unit 112 can then receive and interpret the digital signals. In some embodiments, the signal processing unit 112 also receives position data from the motion control system 105 and a galvanometer (not shown), as well as image data from the image processing system 114. Then, as the optical scanner 102 scans additional points, the signal processing unit 112 can utilize information related to the range and velocity of points in the environment to generate a 3D point cloud. The signal processing unit 112 can also overlay the 3D point cloud data with the image data to determine the velocity and distance of objects in the surrounding area. The system also processes satellite-based navigation and positioning data to provide accurate global positioning.

[0033] Figure 2 This is a time-frequency diagram of an FMCW scan signal 201 that can be used by a LIDAR system such as system 100 to scan a target environment, according to some embodiments. In one example, it is labeled f FM The scan waveform 201 of (t) has a chirped bandwidth Δf C And the chirping period T C The sawtooth waveform (sawtooth "chirping"). The slope of the sawtooth is given as k = (Δf) C / TC ). Figure 2 The target echo signal 202 according to some embodiments is also depicted. (Labeled as f) FM The target echo signal 202 (t-Δt) is a delayed version of the scanning signal 201, where Δt is the round-trip time relative to the target illuminated by the scanning signal 201. The round-trip time is given as Δt = 2R / v, where R is the target range and v is the beam velocity, i.e., the speed of light c. Therefore, the target range R can be calculated as R = c(Δt / 2). When the echo signal 202 is optically mixed with the scanning signal, a range-dependent difference frequency (“beat frequency”) Δf is generated. R (t). Beat frequency Δf R (t) is linearly related to the time delay Δt with a sawtooth slope k. That is, Δf R (t) = kΔt. Since the target range R is proportional to Δt, the target range R can be calculated as R = (c / 2)(Δf) / (c / 2)(Δt / 2)(c ... R (t) / k). That is, the range R and the beat frequency Δf R (t) Linear correlation. Beat frequency Δf R (t) can be generated as an analog signal, for example, in the optical receiver 104 of system 100. Then, the beat frequency can be digitized by an analog-to-digital converter (ADC) in a signal conditioning unit, such as signal conditioning unit 107 in LIDAR system 100. The digitized beat frequency signal can then be digitally processed, for example, in a signal processing unit, such as signal processing unit 112 in system 100. It should be noted that if the target has a velocity relative to LIDAR system 100, the target echo signal 202 will typically also include a frequency shift (Doppler shift). The Doppler shift can be determined separately and can be used to correct the frequency of the echo signal; therefore, for simplicity and ease of interpretation, in Figure 2 Doppler offset is not shown. It should also be noted that the ADC's sampling frequency will determine the highest beat frequency that the system can handle without aliasing. Generally, the highest frequency that can be handled is half the sampling frequency (i.e., the "Nyquist limit"). In one example (but not limited to), if the ADC's sampling frequency is 1 GHz, the highest beat frequency (Δf) that can be handled without aliasing is... Rmax The maximum frequency is 500MHz. This limitation further determines the maximum range of the system to be R. max =(c / 2)(Δf Rmax / k), which can be adjusted by changing the chirp slope k. In one example, although the data samples from the ADC can be continuous, the subsequent digital processing described below can be divided into “time periods” that can be associated with some periodicity in the LIDAR system 100. In one example (but not limited to), the time period can correspond to a predetermined number of chirp periods T, or the number of full rotations of the optical scanner in azimuth. It should be noted that while embodiments of this disclosure can be used in conjunction with FMCW LIDAR, this disclosure is not limited to FMCW LIDAR, and embodiments can also be used with any other form of coherent LIDAR.

[0034] Figure 3A This is a block diagram illustrating an example FMCW LIDAR system 300 according to the present disclosure. The example system 300 includes an optical scanner 301 for transmitting an FMCW infrared (IR) beam 304 and receiving an echo signal 313 reflected from a target (such as target 312, etc.) in the field of view (FOV) of the optical scanner 301. The system 300 also includes an optical processing system 302 for generating an electrical signal 314 in the time domain based on the echo signal 313, wherein the electrical signal 314 contains a frequency corresponding to the range of the LIDAR target. According to some embodiments, the optical processing system 302 may include elements of the free-space optics 115, optical path 101, optical driver 103, and optical receiver 104 of the LIDAR system 100.

[0035] System 300 also includes a signal processing system 303, which measures the energy of the electrical signal 314 in the frequency domain, compares the energy with an estimate of the LIDAR system noise, and determines the likelihood of a detected target indicated by a signal peak in the frequency domain. According to some embodiments, the signal processing system 303 may include elements of the signal conversion unit 106, signal conditioning unit 107, LIDAR control system 110, and signal processing unit 112 in the LIDAR system 100.

[0036] Figure 3B This is a block diagram illustrating an example electro-optical system 350. According to some embodiments, the electro-optical system 350 includes components designed for… Figure 1 The optical scanner 102 illustrated and described is similar to the optical scanner 301. The electro-optical system 350 also includes an optical processing system 302, which, as described above, may include elements of the free-space optics 115, optical path 101, optical driver 103, and optical receiver 104 in the LIDAR system 100.

[0037] The optical processing system 302 includes a light source 305 for generating a frequency-modulated continuous wave (FMCW) beam 304. The beam 304 can be guided to an optical coupler 306, which is configured to couple the beam 304 to a polarization beam splitter (PBS) 307 and a sample 308 of the beam 304 to a photodetector (PD) 309. The PBS 307 is configured to guide the beam 304 toward an optical scanner 301 due to the polarization of the beam 304. The optical scanner 301 is configured to scan the target environment using the range of azimuth and elevation angles of the field of view (FOV) 310 of the LIDAR window 311 covering the housing 320 of the optical system 350. Figure 3B For ease of explanation, only the azimuth scan is shown in the image.

[0038] like Figure 3B As shown, at an azimuth angle (or angular range), beam 304 passes through LIDAR window 311 and illuminates target 312. Echo signal 313 from target 312 passes through LIDAR window 311 and is guided back to PBS 307 by optical scanner 301.

[0039] The echo signal 313 (which will have a different polarization than the beam 304 due to reflection from the target 312) is guided by the PBS 307 to the photodetector (PD) 309. In the PD 309, the echo signal 313 is optically mixed with a local sample 308 of the beam 304 to generate an electrical signal 314 (e.g., a beat frequency signal) with a frequency proportional to the range of the scanned target. The electrical signal 314 can be generated by the frequency difference between the local sample 308 of the beam 304 and the echo signal 313 relative to time (i.e., Δf). R (t) is used to generate it.

[0040] Figure 4 This is a detailed block diagram illustrating an example of a signal processing system 303 for processing electrical signal 314 according to some embodiments. As described above, the signal processing system 303 may include elements of the signal conversion unit 106, signal conditioning unit 107, LIDAR control system 110, and signal processing unit 112 in the LIDAR system 100. According to some embodiments, the signal processing system 303 includes an analog-to-digital converter (ADC) 401, a time-domain signal processor 402, a block sampler 403, a discrete Fourier transform processor 404, a frequency-domain signal processor 405, and a peak search processor 406. The component blocks of the signal processing system 303 may be implemented in hardware, firmware, software, or a combination of hardware, firmware, and software.

[0041] exist Figure 4In this circuit, electrical signal 314 (which is a continuous analog signal in the time domain) is sampled by ADC 401 to generate a series of time-domain samples 315. The time-domain samples 315 are processed by time-domain module 402, which modifies them for further processing. For example, time-domain module 402 may apply weighting or filtering to remove unwanted signal artifacts or make the signal more processable for subsequent processing. The output 316 of time-domain module 402 is provided to block sampler 403. Block sampler 403 groups the time-domain samples 316 into groups of N samples 317 (where N is an integer greater than 1) that are provided to DFT module 404. DFT module 404 transforms the groups of N time-domain samples 317 into N frequency intervals or subbands in the frequency domain (e.g., subband signal spectrum 319), thereby covering the bandwidth of electrical signal 314. N sub-band signal spectra 319 are provided to frequency domain module 405, which adjusts the sub-bands for further processing. For example, frequency domain module 405 may resample and / or average the sub-band signal spectra 319 to reduce noise. Frequency domain module 405 may also calculate signal statistics and system noise statistics. The processed sub-band signal spectra 319 are then provided to peak search module 406, which searches for signal peaks representing detected targets in the FOV of the LIDAR system 300.

[0042] In some embodiments, the subband signal spectrum 319 provided to the peak search module 406 is the sum of the energy in the target echo 313 and all the noise generated by the LIDAR system 300 while the target echo signal is being processed. In some scenarios, electronic systems have noise sources that limit the performance of these systems by creating a noise floor, which is the combined level of all noise sources in the system. In order to be detected, the signal generated by the electrical signal 314 in the electronic system (such as the subband signal spectrum 319, etc.) must be above the noise floor, without the need for dedicated signal processing techniques such as signal integration and noise averaging. Noise sources in a LIDAR system (such as LIDAR system 300, etc.) can include thermal noise, 1 / f noise, shot noise, impulse noise, RIN (relative intensity noise associated with the laser), TIA (transimpedance amplifier) ​​noise, and ADC (analog-to-digital converter) noise. System noise can be characterized, for example, as the relationship between energy and frequency distribution over a frequency interval, the first moment (mean) over a frequency interval, the second moment (variance) over a frequency interval, the third moment (asymmetry) over a frequency interval, and / or the fourth moment (kurtosis or sharpness of peaks) over a frequency interval of the spectrum.

[0043] Figure 5This is a block diagram illustrating an example of a system 500 that performs peak search using automatic adjustment of a detection threshold according to some embodiments. System 500 may be the same as or similar to the signal processing system 303 of FIG3. Furthermore, system 500 may include or be included in... Figure 1 The signal processing unit 100 is located in one or more components (e.g., signal conversion unit 106, signal conditioning unit 107, LIDAR control system 110, and signal processing unit 112). System 500 includes a peak search module 406, a target predictor 502, and a point cloud module 504. Peak search module 406 can receive the sub-band signal spectrum 319 as described above. Peak search module 406 can detect one or more peaks (as shown below) from the sub-band signal spectrum 319. Figure 6A , Figure 6B and Figure 7 (As shown in the diagram). The peak search module 406 can compare a detection confidence metric (e.g., the difference between signal energy and noise floor versus noise ratio (SN / N)) with a detection threshold. In some scenarios, if the detection confidence metric of a frequency peak is above the detection threshold, the peak search module 406 can identify the frequency peak as a confirmed target detection, while filtering peaks below the detection threshold (e.g., false alarm detections). In some scenarios, peaks exceeding the detection threshold can indicate a high probability that the peak corresponds to a valid target detection.

[0044] Peak search module 406 can provide one or more identified peaks 515 from the subband signal spectrum 319 to point cloud module 504. Point cloud module 504 can generate point cloud 505 based on one or more identified peaks and other information from the frame (i.e., the detected target). Point cloud module 504 can also identify contextual information 520 near detection (e.g., target classification, estimated target velocity, scenario type, previous point cloud information). Point cloud module 504 can provide contextual information 520 to target predictor 502.

[0045] Target predictor 502 can estimate the location 525 of a previously detected target. For example, target predictor 502 can utilize context information 520 to estimate the current target location 525. Context information 520 can include one or more previous frames that can include corresponding detections of the target. Target predictor 502 can then use the detections of the target in the previous frames to estimate the current location of the target. In another example, target predictor 502 can use the median location of the detections in the previous frames to estimate the target location 525. It should be noted that the target predictor can use any number of previous frames or neighboring points in the point cloud to perform averaging, medianing, or any other calculation to estimate the target location. In another example, context information 520 can include higher-level information from point cloud 505 regarding the object associated with the detection (e.g., a vehicle, person, stationary object, etc.) and the expected behavior of the object.

[0046] Figure 6A This is a graph illustrating the amplitude of a subband signal spectrum 319, including system noise, relative to its frequency. For ease of illustration, the subband signal spectrum is shown as a continuous waveform (rather than discrete frequency intervals or subbands). Figure 6A It can be generated and / or used by the signal processing system 303 in Figure 3. Furthermore, Figure 6A It can be by Figure 1 One or more components of the LIDAR system 100 (e.g., signal conversion unit 106, signal conditioning unit 107, LIDAR control system 110, and signal processing unit 112) generate and / or use the frequency spanning from 0 to Δf. Rmax The range. In some scenarios, for example, without more information about the sub-band signal spectrum 319, the peak search module 406 will select the highest signal peak 601 as the echo signal most likely indicating the presence of a target, and will not select the lower signal peak 602. However, if the peak search module 406 has an estimate of the system noise, it can compare the sub-band signal spectrum 319 with the system noise estimate and can make a more reliable selection based on additional selection criteria (i.e., confidence metrics). Figure 6A and Figure 6B In this context, signal and noise values ​​are described using energy as a relative frequency profile. However, as previously mentioned, system noise can also be characterized by any one of the first to fourth moments, representing the mean energy, energy variance, energy asymmetry, and kurtosis versus frequency, respectively. In addition to energy alone, the signal can also be characterized by autocorrelation statistics across frequency cells in the baseband and / or cross-correlation statistics between signal and system noise estimates across frequency cells.

[0047] In one example, the estimation of system noise can be obtained by operating a LiDAR system (such as LiDAR system 300, etc.) in a no-echo calibration mode, in which there is no detectable echo signal (e.g., echo signal 313). This operating mode generates all normal system noise mechanisms and yields a sub-band signal spectrum 319 that includes energy from only the system noise sources.

[0048] Figure 6B It is the difference between noise estimate 651 and subband signal spectrum 319 and noise estimate 651 (in Figure 6B The middle figure shows a comparison of energy versus frequency for signal noise reduction (SN) 661 (Figure 650). Figure 6B In the example, peak search module 406 can be configured to select the signal peak with the highest non-negative signal-to-noise ratio (SN) / N. Under this selection criterion, since (SN) / N 653 is greater than (SN) / N 655, signal peak 652 with (SN) / N 653 will be selected instead of signal peak 654 with (SN) / N 655.

[0049] In some examples, signal processing system 303 can be configured to modify the subband signal (e.g., subband signal spectrum 319) and system noise estimation (e.g., system noise estimation 651) to generate a confidence metric (e.g., (SN) / N) that improves the likelihood of signal peaks in the frequency domain indicating detected targets and reduces the likelihood that signal peaks from false targets in the frequency domain will be interpreted as real targets. Additionally, as mentioned above regarding... Figure 5 And below Figure 7 As discussed in the paper, the peak search module 406 can use the estimated location of the target to adjust the detection threshold of the confidence metric.

[0050] Figure 7 A signal confidence-frequency plot is drawn, illustrating an example method for threshold adjustment for peak detection according to this disclosure. Figure 7 It can be made by the signal processing system 303 in Figure 3 or Figure 4 and Figure 5 The peak search module generates and / or uses peaks. Furthermore, Figure 7 It can be by Figure 1 One or more components of the LIDAR system 100 (e.g., signal conversion unit 106, signal conditioning unit 107, LIDAR control system 110, and signal processing unit 112) are generated and / or used. Figure 7 Three potential target detection methods are described, including strong signal target detection 705, false alarm detection 710, and weak signal target detection 715. Figure 7It also describes the effective threshold of 720 for the confidence measure, which is also referred to here as the confidence threshold (as mentioned above for...). Figure 6B (As described). The effective threshold 720 of the confidence metric can be the minimum confidence threshold for a signal to be identified as a detection. Strong signal target detection 705 is greater than the higher (i.e., unadjusted) effective threshold 720 and is therefore detected without adjusting the effective threshold 720. The second potential detection is a potential false alarm detection 710 (i.e., a peak that does not correspond to the actual target). Therefore, if the effective threshold 720 is lowered too much, the false alarm detection 710 will also be identified as an actual detection even though it does not correspond to the target. The third potential detection is a weak signal target 715. As depicted, weak signal target 715 will not exceed the effective threshold 720 of the confidence metric and will therefore not be recorded as a detection.

[0051] To provide a higher probability of detecting weak-signal targets (e.g., weak target 715) while keeping the probability of false detections (e.g., false alarms 710) low, peak search module 406 can reduce the effective threshold 720 of the predicted location (i.e., predicted target band 725) around weak target 715. The predicted target band 725 can be a predefined frequency range around the expected target frequency (i.e., target location), or it can be dynamically determined based on the confidence level associated with the estimated target location. For example, peak search module 406 can determine a larger predicted target band 725 for estimated locations with low confidence (e.g., if the variance associated with the estimate is high), and a smaller predicted target band 725 for estimated locations with high confidence (e.g., if the variance associated with the estimate is low). Therefore, the reduced effective threshold 730 within the predicted target band 725 allows the peak search module 406 to detect weak targets 715 while preventing false alarms 710, because the effective threshold 720 is adjusted to the reduced effective threshold 730 only for small frequency regions. For example, the predicted target band 725 can cover frequencies associated with a physical detection range of approximately 5-10 meters. It should be noted that the predicted target band 725 can span any frequency range and any corresponding physical distance range of the LIDAR system.

[0052] In one embodiment, the target predictor (e.g., Figure 5 The target predictor 502 can estimate the location of targets for which weak signals have been previously detected. For example, the target predictor 502 can estimate the location of targets for which low SNR signals have been previously detected. In another example, the target predictor 502 can calculate the estimated location of target detections with a confidence metric just above a higher (i.e., unadjusted) effective threshold 720. The peak search module 406 can then lower the effective threshold only for these weak signal targets for which estimated locations have already been calculated.

[0053] Figure 8 This is a flowchart illustrating a method 800 for automatically adjusting a detection threshold based on the estimated location of a target in a LiDAR system (such as LiDAR system 100 or LiDAR system 300). Method 800 can be derived from... Figure 1 The signal processing is performed by one or more components of the LIDAR system 100 (e.g., signal conversion unit 106, signal conditioning unit 107, LIDAR control system 110, and signal processing unit 112).

[0054] Method 800 begins at operation 802, where processing logic (e.g., peak search module 406) determines a first confidence threshold for detecting a first target among a plurality of targets within a frequency range, wherein the frequency range includes different frequencies corresponding to the targets. The first confidence threshold may correspond to a confidence metric for the target signal. The confidence metric may be a signal-to-noise ratio (SNR), a signal-to-noise ratio ((SN) / N), or a likelihood ratio that provides a confidence value for target detection. The confidence metric may be calculated for each target signal. Determining the confidence metric may include determining a noise estimate across the frequency domain of the baseband signal and calculating the difference between the amplitude of the baseband signal's frequency and the noise estimate across the frequency domain.

[0055] At operation 804, processing logic (e.g., peak search module 406 and / or target predictor module 502) determines a subset of frequencies within a frequency range for detecting a second target, wherein the second target transmits signals below a first confidence threshold within the frequency subset. In some embodiments, the frequency subset corresponds to the estimated location of the second target. In one example, to determine the estimated location of the target, the processing logic (e.g., target predictor module 502) identifies detections of the second target in the time or frequency domain and determines the estimated location of the second target based on neighborhood detections such as along azimuth, elevation, spatial (3D point cloud), or temporal (detection from a previous frame). In one example, to determine the estimated location of the second target, the processing logic (e.g., target predictor module 502) may identify point cloud information associated with the second target and determine the estimated location of the second target based on the point cloud information.

[0056] At operation 806, processing logic (e.g., peak search module 406) adjusts a first confidence threshold to a second confidence threshold at a frequency subset for detecting a second target within the frequency subset. The processing logic (e.g., peak search module 406) can use the second confidence threshold to filter false alarm peak detections that are less than the first detection threshold. To adjust the detection threshold, the processing logic (e.g., peak search module 406) can reduce the detection threshold within a frequency subset (i.e., the target frequency band range) corresponding to the estimated location of the target. In one embodiment, the target frequency band range is a predefined frequency range near the frequency corresponding to the estimated target location. In another embodiment, the processing logic (e.g., peak search module 406) determines the width of the target frequency band range based on the confidence associated with the estimated location of the second target.

[0057] At operation 808, processing logic (e.g., peak search module 406) restores the second confidence threshold to the first confidence threshold outside the frequency subset for detecting the first target. In one embodiment, the processing logic (e.g., peak search module 406) can compare the peak of the currently detected second target with the reduced second confidence threshold. In one embodiment, in response to determining that the peak exceeds the reduced second confidence threshold, the processing logic (e.g., peak search module 406) determines the actual location of the target based on the current detection. In one embodiment, the processing logic (e.g., peak search module 406) can determine whether one or more previous detections of the target include a weak detection signal. The processing logic (e.g., peak search module 406) can also determine an estimated target location and adjust the detection threshold in response to determining that one or more previous detections of the target include a weak detection signal.

[0058] Figure 9 It is the processing system 900 in a LIDAR system (such as LIDAR System 100 or LIDAR System 300, etc.) (for example, similar to the above regarding...). Figure 4A block diagram of the signal processing system 903 shown and described. Processing system 900 includes a processing device 901, which can be any type of general-purpose or special-purpose processing device designed for a LIDAR system. Processing device 901 is coupled to a memory 902, which can be any type of non-volatile computer-readable medium (e.g., RAM, ROM, PROM, EPROM, EEPROM, flash memory, disk storage, or optical disk storage) containing instructions that, when executed by processing device 901 in the LIDAR system, cause the LIDAR system to perform the methods described herein. Specifically, memory 902 includes instructions 904 for determining a first confidence threshold for detecting a first target from a plurality of targets within a frequency range, wherein the frequency range includes different frequencies corresponding to the plurality of targets. The memory 902 includes: instructions 906 for determining a frequency subset within a frequency range for detecting a second target, wherein the second target transmits a signal within the frequency subset that is below a first confidence threshold; and instructions 908 for adjusting the first confidence threshold to a second confidence threshold within the frequency subset for detecting the second target. The memory 902 also includes instructions 910 for restoring the second confidence threshold to the first confidence threshold outside the frequency subset for detecting the first target.

[0059] The foregoing description sets forth numerous specific details, such as examples of specific systems, components, methods, etc., to provide a thorough understanding of several examples in this disclosure. However, it will be apparent to those skilled in the art that at least some examples of this disclosure can be implemented without these specific details. In other instances, well-known components or methods have not been described in detail or presented in the form of simple block diagrams to avoid unnecessarily obscuring this disclosure. Therefore, the specific details set forth are merely exemplary. Specific examples may differ from these exemplary details and are still contemplated within the scope of this disclosure.

[0060] Any reference to "an embodiment" or "example" throughout this specification means that a particular feature, structure, or characteristic described in connection with the example is included in at least one example. Therefore, the phrases "in an example" or "in the example" appearing in various places throughout this specification do not necessarily refer to the same example. Although the operation of methods is shown and described herein in a specific order, the order of operation of the various methods may be changed so that some operations can be performed in reverse order, or so that some operations can be performed at least partially concurrently with other operations. Instructions or sub-operations of different operations may be performed intermittently or alternately.

[0061] The above description of the illustrated implementations of the invention (including those described in the abstract) is not intended to exhaustively or limit the invention to the precise forms disclosed. While specific implementations and examples of the invention have been described herein for illustrative purposes, various equivalent modifications can be made within the scope of the invention, as will be appreciated by those skilled in the art. The terms “example” or “exemplary” are used herein to mean used as an example, instance, or illustration. Any aspect or design described herein as “example” or “exemplary” is not necessarily to be construed as being more preferred or advantageous than other aspects or designs. Rather, the use of the terms “example” or “exemplary” is intended to present concepts in a concrete manner. As used herein, the term “or” means inclusive “or” rather than exclusive “or.” That is, unless otherwise specified or the context clearly indicates, “X includes A or B” means any natural inclusion arrangement. That is, if X includes A; X includes B; or X includes both A and B, then “X includes A or B” is satisfied under any of the above instances. Furthermore, the terms “a” and “an” as used in this application and the appended claims should generally be interpreted as meaning “one or more” unless otherwise specified or clearly indicated from the context as singular. Additionally, the terms “first,” “second,” “third,” “fourth,” etc., as used herein refer to labels used to distinguish between different elements and do not necessarily have a meaning according to the order of their numerical names.

Claims

1. A light detection and ranging (LIDAR) system, comprising: an optical scanner to transmit a light beam toward a target and to receive a return signal from the target; an optical processing system coupled to the optical scanner to generate a baseband signal in a time domain from the return signal, the baseband signal including a frequency corresponding to a LIDAR target range; and a signal processing system coupled to the optical processing system, the signal processing system including: a processor; and a memory operatively coupled to the processor, the memory to store instructions that, when executed by the processor, cause the LIDAR system to: determine a first confidence threshold for detecting a first target from a plurality of targets within a frequency range associated with the baseband signal, wherein the first confidence threshold is based on at least one of a likelihood ratio and a signal-to-noise ratio that provide a confidence value for target detection, the frequency range including a plurality of different frequencies corresponding to the plurality of targets; determine a subset of frequencies within the frequency range for detecting a second target, wherein the second target transmits a signal within the subset of frequencies that is below the first confidence threshold; adjust the first confidence threshold to a second confidence threshold at the subset of frequencies for detecting the second target within the subset of frequencies; and restore the second confidence threshold to the first confidence threshold outside the subset of frequencies to detect the first target.

2. The system of claim 1, wherein, To determine the first confidence threshold, the processor is further to: determine a noise estimate across the frequency range; and calculate a difference between an amplitude of a frequency and the noise estimate across the frequency range.

3. The system of claim 1, wherein, To determine the subset of frequencies within the frequency range for detecting the second target, the processor is to: determine the subset of frequencies based on an estimated location of the second target.

4. The system of claim 3, wherein, The processor is further to: identify a previous detection of the second target in the time domain or a frequency domain; and determine the estimated location of the second target based on the previous detection of the second target.

5. The system of claim 3, wherein, To determine the estimated location of the second target, the processor is to: identify point cloud information associated with the second target; and determine the estimated location of the second target based on the point cloud information.

6. The system of claim 1, wherein, The subset of frequencies includes a predefined frequency range corresponding to a previously detected location of the second target.

7. A method of operating a frequency modulated continuous wave (FMCW) light detection and ranging (LIDAR) system, comprising: determining a first confidence threshold for detecting a first target from a plurality of targets within a frequency range associated with a baseband signal, wherein the first confidence threshold is based on at least one of a likelihood ratio and a signal-to-noise ratio that provide a confidence value for target detection, the frequency range including a plurality of different frequencies corresponding to the plurality of targets; determining, by a processing device, a subset of frequencies within the frequency range for detecting a second target, wherein the second target transmits a signal within the subset of frequencies that is below the first confidence threshold; adjusting, by the processing device, the first confidence threshold to a second confidence threshold at the subset of frequencies for detecting the second target within the subset of frequencies; and restoring the second confidence threshold to the first confidence threshold outside of the subset of frequencies for detecting the first target.

8. The method of claim 7, wherein, determining the first confidence threshold includes: determining a noise estimate across the frequency range; and computing a difference between a magnitude of a frequency and the noise estimate across the frequency range.

9. The method of claim 7, wherein, determining the subset of frequencies within the frequency range for detecting the second target includes: determining the subset of frequencies based on an estimated location of the second target.

10. The method of claim 9, further comprising: identifying a previous detection of the second target in a time domain or a frequency domain; and determining the estimated location of the second target based on the previous detection of the second target.

11. The method of claim 9, wherein, determining the estimated location of the second target includes: identifying point cloud information associated with the second target; and determining the estimated location of the second target based on the point cloud information.

12. The method of claim 11, wherein, the subset of frequencies includes a predefined frequency range corresponding to a previously detected location of the second target.

13. The method of claim 7, wherein, adjusting the first confidence threshold includes: reducing the first confidence threshold to the second confidence threshold within the subset of frequencies, wherein the subset of frequencies corresponds to the estimated location of the second target.

14. The method of claim 13, further comprising: determining the subset of frequencies corresponding to the estimated location of the second target based on a confidence value associated with the estimated location of the second target.

15. The method of claim 7, further comprising: determining whether one or more previous detections of the second target include a weak detection signal; determining an estimated location of the second target; and adjusting the first confidence threshold in response to determining that the one or more previous detections of the second target include the weak detection signal.

16. A light detection and ranging (LIDAR) system, comprising: an optical scanner that transmits a light beam and receives a plurality of return signals from reflections of the light beam; an optical processing system coupled to the optical scanner, the optical processing system generating electrical signals from the plurality of return signals; a signal processing system coupled to the optical processing system, the signal processing system comprising: a processing device; and a memory operatively coupled to the processing device, the memory to store instructions that, when executed by the processing device, cause the LIDAR system to: determine a first confidence threshold for a frequency range of the electrical signals to detect a first target, wherein the first confidence threshold is based on at least one of a likelihood ratio and a signal-to-noise ratio that provide a confidence value for target detection; estimate a location of a second target based on contextual information associated with the second target; and determine a second confidence threshold to detect the second target, wherein the second confidence threshold is determined for a subset of frequencies of the frequency range, the subset of frequencies corresponding to the estimated location of the second target.

17. The LIDAR system of claim 16, wherein, The second confidence threshold is lower than the first confidence threshold.

18. The LIDAR system of claim 16, wherein, The frequency subset is determined based on an estimated position of the second target and a confidence value associated with the estimated position of the second target.

Images