Managing optical phased array performance based on phase shift variations

Abstract

The method includes providing light to a first set of spaced apart optical antennas in a first optical phased array (OPA), where a plurality of phase shifters impose a phase shift on the light provided to each optical antenna of the plurality of optical antennas; and managing the imposed phase shift by using a phase-shift control module. The managing includes a first mode of operation, the first mode of operation including determining an imposed phase-shift variation less than a phase-shift variation range for a first subset of the optical antennas within a fixed distance from a center of the plurality of optical antennas, and determining an imposed phase-shift variation greater than the phase-shift variation range for a second subset of the optical antennas further away from the center of the plurality of optical antennas.

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to and benefit of U.S. Provisional Application No. 63 / 340,526, entitled "MITIGATION OF SPURIOUS RETURNS IN OPTICAL PHASED ARRAYS," filed May 11, 2022, which is incorporated by reference herein.

[0002] This disclosure relates to managing optical phased array performance based on phase shift variations. [Background technology]

[0003] An optical phased array (OPA) LiDAR (Light Detection and Ranging) system may generate a return intensity map (also called an angular intensity distribution) that includes spurious signal peaks associated with lobes other than the intended lobe of the optical phased array radiation pattern. For example, light associated with the main lobe may be reflected to provide a main signal peak that triggers a detection event, but the radiation pattern typically also includes side lobes on either side of the main lobe (e.g., based on the finite spread of the OPA) and many grating lobes at some angular intervals (e.g., based on the finite spacing between the optical antennas that make up the OPA). The grating lobes may be larger than the side lobes, and light from either the grating lobes and / or the side lobes may be reflected to provide spurious (e.g., false or unwanted) signal peaks. If the strength of these spurious signals exceeds the post-digital noise floor of the LiDAR system, these spurious signals may register as false detection events, which may be problematic for typical LiDAR applications. Summary of the Invention [Means for solving the problem]

[0004] In one aspect, in general, an apparatus includes an optical phased array (OPA) including a plurality of optical antennas separated by a set of intervals and a plurality of phase shifters configured to impose a phase shift on light provided to each optical antenna of the plurality of optical antennas; and a phase shift control module configured to manage the imposed phase shift, the managing including a first operating mode, the first operating mode including determining an imposed phase shift variation smaller than a phase shift variation range for a first subset of the optical antennas within a certain distance from a center of the plurality of optical antennas, and determining an imposed phase shift variation larger than the phase shift variation range for a second subset of the optical antennas further away than the certain distance from the center of the plurality of optical antennas.

[0005] Some embodiments can include one or more of the following features.

[0006] Determining a variation in the imposed phase shift of the first subset of the optical antennas and determining a variation in the imposed phase shift of the second subset of the optical antennas includes determining the variation according to a probability distribution.

[0007] Determining the variance according to a probability distribution includes determining the variance according to a pseudorandom function.

[0008] The managing further includes a second mode of operation that includes determining an imposed phase shift corresponding to a deterministic phase profile for the first subset of optical antennas and the second subset of optical antennas.

[0009] The deterministic phase profile includes a linear phase profile that defines a common phase shift to be imposed on each of the multiple phase shifters.

[0010] The managing further includes a first scanning mode of operation, where a first region is scanned by managing the imposed phase shifts in the first mode of operation of two or more transmit beam angles.

[0011] The managing further includes a second scanning mode of operation, where a second region is scanned by managing the imposed phase shift in the second mode of operation of the two or more transmit beam angles.

[0012] The first region is a subset of the second region.

[0013] The apparatus further includes a control module configured to receive the scattered light emitted from the OPA and estimate a distance to the target based at least in part on a characteristic of the scattered light.

[0014] The OPA is a first OPA, and the apparatus further includes a second OPA configured to receive the scattered light emitted from the first OPA and provide the scattered light to the control module.

[0015] Each interval in the set of intervals is greater than one-half of any wavelength in the spectrum of light provided to the multiple optical antennas in the OPA.

[0016] All intervals in a set of intervals are identical to each other.

[0017] In another aspect, in general, a method includes providing light to a set of spaced apart optical antennas in an optical phased array (OPA), where a plurality of phase shifters impose a phase shift on the light provided to each optical antenna of the plurality of optical antennas; and managing the imposed phase shift using a phase shift control module, the managing including a first mode of operation, the first mode of operation including determining, for a first subset of the optical antennas within a certain distance from a center of the plurality of optical antennas, a variation in the imposed phase shift less than a phase shift variation range, and determining, for a second subset of the optical antennas further away than the certain distance from the center of the plurality of optical antennas, a variation in the imposed phase shift greater than the phase shift variation range.

[0018] Determining a variation in the imposed phase shift of the first subset of the optical antennas and determining a variation in the imposed phase shift of the second subset of the optical antennas includes determining the variation according to a probability distribution.

[0019] Determining the variance according to a probability distribution includes determining the variance according to a pseudorandom function.

[0020] The managing further includes a second mode of operation that includes determining an imposed phase shift corresponding to a deterministic phase profile for the first subset of optical antennas and the second subset of optical antennas.

[0021] The deterministic phase profile includes a linear phase profile that defines a common phase shift to be imposed on each of the multiple phase shifters.

[0022] The managing further includes a first scanning mode of operation, where a first region is scanned by managing the imposed phase shifts in the first mode of operation of two or more transmit beam angles.

[0023] The managing further includes a second scanning mode of operation, where a second region is scanned by managing the imposed phase shift in the second mode of operation of two or more transmit beam angles.

[0024] The first region is a subset of the second region.

[0025] The method further includes receiving scattered light emitted from the OPA and estimating a distance to the target based at least in part on a characteristic of the scattered light.

[0026] The OPA is a first OPA, and the method further includes receiving scattered light emitted from the first OPA into the second OPA and providing the scattered light for estimating the distance.

[0027] Some embodiments may have one or more of the following advantages.

[0028] In a LiDAR system, light waves from a light source can be transmitted to a target object at a given distance using an OPA, and light backscattered from the target object can be collected using another OPA. Various techniques, such as modulation and / or time of flight, can be used to determine the distance to the target object based on information related to the detected event. The light source used in such a system can be a laser or other coherent light source that provides light waves (also referred to herein simply as "light") having a narrow linewidth and a peak wavelength that falls within a specific range (e.g., between about 100 nm and about 1 mm, or some subrange thereof).

[0029] In some environments (e.g., automotive environments) where LiDAR systems or other optical systems using transmitter OPAs and receiver OPAs may be used, return intensity maps (e.g., representing the intensity of collected return light) including spurious signal peaks associated with side lobes and / or grating lobes of the optical phased array (OPA) radiation pattern may be exacerbated by the ubiquity of retroreflectors (e.g., road signs, road markers, bicycle safety reflectors, registration stickers, etc.) that may generate return signals orders of magnitude higher than those generated by diffuse scattering surfaces. Correspondingly, return signals associated with the mixing of side lobes or grating lobes in distant retroreflectors may easily exceed the noise floor of the phased array LiDAR system, generating false detection events. Mitigation of this effect is useful for commercialization of phased array LiDAR technology in automotive and other application spaces.

[0030] In some examples, the optical systems described herein may be used to explore and measure an environment or area over a wider range of angles than existing implementations. For example, a configuration in which the grating lobes of the transmitter aperture and the grating lobes of the receiver aperture are aligned while the main lobes of the transmitter aperture and the main lobes of the receiver aperture are misaligned may allow scanning over angles outside of the area accessible to the main lobe, as described in more detail below.

[0031] Other features and advantages will become apparent from the following description and from the accompanying drawings, and from the claims.

[0032] The present disclosure will be best understood from the following detailed description when read in conjunction with the accompanying drawings, in which: It is emphasized that, according to common practice, the various features of the accompanying drawings are not to scale. Conversely, the dimensions of the various features have been arbitrarily expanded or reduced for clarity. [Brief description of the drawings]

[0033] [Figure 1A] FIG. 1 is a schematic diagram of an exemplary optical system. [Figure 1B] FIG. 1 is a schematic diagram of an exemplary LiDAR system. [Figure 2A] 3A-3C are schematic diagrams of exemplary radiation intensity patterns of a transmitter OPA and a receiver OPA. [Figure 2B] 1 is a plot of predicted example radiation intensity patterns of a transmitter OPA and a receiver OPA as a function of the angle formed relative to the center of the main lobe. [Figure 2C] 1 is a plot of predicted example radiation intensity patterns of a transmitter OPA and a receiver OPA as a function of the angle formed relative to the center of a grating lobe. [Figure 2D] 13 is a plot of predicted example radiation intensity pattern products of the transmitter OPA and receiver OPA as a function of the angle the transmitter OPA grating lobe and the receiver OPA grating lobe form with respect to an aligned boresight. [Figure 2E] 13 is a plot of predicted example radiation intensity pattern products of the transmitter OPA and receiver OPA as a function of the angle at which the transmitter OPA grating lobe and the receiver OPA grating lobe form with respect to a misaligned boresight. [Figure 3A] 3A-3C are schematic diagrams of exemplary radiation intensity patterns of a transmitter OPA and a receiver OPA. [Figure 3B]1 is a plot of predicted example radiation intensity patterns of a transmitter OPA and a receiver OPA as a function of the angle formed relative to the center of the main lobe. [Figure 3C] 1 is a plot of predicted example radiation intensity patterns of a transmitter OPA and a receiver OPA as a function of the angle formed relative to the center of a grating lobe. [Figure 4] 1 is a schematic diagram of an exemplary spurious image detection and mitigation; [Figure 5A] 1 is a schematic diagram of an exemplary optical system in the near field. [Figure 5B] 1 is a plot of predicted example radiation intensity patterns of a transmitter OPA and a receiver OPA as a function of the angle formed relative to the center of a grating lobe. [Figure 5C] 1 is a schematic diagram of an exemplary optical system in the near field. [Figure 5D] 1 is a plot of predicted example radiation intensity patterns of a transmitter OPA and a receiver OPA as a function of the angle formed relative to the center of a grating lobe. [Figure 6A] 1 is a plot of predicted exemplary absolute value of effective amplitude of apodization as a function of antenna number, which corresponds to the relative position of each antenna in an optical phased array. [Figure 6B] 1 is a plot of predicted example phases of apodization as a function of antenna number, which corresponds to the relative position of each antenna in an optical phased array. [Figure 7] 13 is a plot of predicted example radiation intensity pattern products with apodization for a transmitter OPA and a receiver OPA as a function of the angle formed relative to the center of the lobe. [Figure 8] 13 is a plot of predicted example disparity as a function of the angle of the target relative to the transmitter OPA for various optical system configurations. [Figure 9] 1 is a plot of predicted example radiation intensity patterns of a transmitter OPA and two receiver OPAs as a function of the angle formed relative to the center of the lobe. [Figure 10] 1 is a flow diagram of an exemplary spurious image detection and mitigation scheme. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0034] An optical phased array (OPA) may form a steerable light beam based on the phased array principle. For example, an array of optical antennas (sometimes called optical emitters) may each emit light with a controllable (e.g., controlled by a phase shifter coupled to the optical antenna) phase, leading to an interference pattern forming one or more light beams. Typically, each optical antenna may also receive light in addition to emitting light. A transmit OPA may be characterized by an angular intensity distribution, also called a radiant intensity pattern, which may describe the light beam formed by the transmit OPA. Thus, the light beam formed by the transmit OPA may include light emitted at one or more angles characterized by the angular intensity distribution. A receiver OPA may receive or collect light that may be used to characterize a scene or area surrounding the OPA. A receiver OPA may also be characterized by an angular intensity distribution (radiant intensity pattern), also called a gain pattern. The gain pattern characterizes the angular sensitivity of the receiver OPA to the incident light. Both the transmitter angular intensity distribution and the receiver angular intensity distribution may have lobes or peaks, also called array factor peaks, corresponding to maxima where the respective OPA has higher transmit power or receive sensitivity. An optical platform including both the receiver OPA and the transmitter OPA may be characterized by a radiation intensity pattern product. In some examples, the radiation intensity pattern product may be determined by multiplying the transmitter radiation intensity pattern and the receiver radiation intensity pattern. In other examples, the radiation intensity pattern product may depend on other factors (e.g., antenna element factors).

[0035] Some of the examples described herein may include an OPA having a receive aperture (i.e., receive OPA, receiver or receive subsystem) and a transmit aperture (i.e., transmit OPA, transmitter or transmit system). Other examples may include separate structures, where the transmit aperture and receive aperture are not physically connected together or are fabricated as stand-alone devices. Additionally, the transmit aperture and receive aperture may be the same.

[0036] In some example implementations, the optical systems described herein may be designed to operate over a determined range of optical wavelengths (e.g., λ=1500-1600 nm band or λ=1270-1330 nm band), and the base spacing pitch between optical antennas may be of similar order of magnitude to the optical wavelength. For example, for operation in the 1500-1600 nm band, 1000 nm≦a≦2000 nm may be typical. All grating lobes would be eliminated if a<λ / 2, but a practical lower limit on the grating pitch may be set by the refractive indices of available waveguide materials, as well as deleterious effects of waveguide-waveguide coupling as the pitch is reduced. For example, individually single-mode silicon waveguides designed for the 1500-1600 nm band would couple with each other with a characteristic coupling length of 50 μm-200 μm if placed at a pitch of 750 nm to completely eliminate grating lobes; therefore, phase-dependent optical redispersion between the waveguides would be inevitable and thus impair the angular intensity distribution. Thus, in practical implementations, the presence of grating lobes in the transmit (TX) radiation pattern and receive (RX) gain pattern may occur. However, the grating lobes of the TX OPA and the grating lobes of the RX OPA can be misaligned by applying uniform pitch mismatch (Δa), element factor engineering, pitch apodization, random pitch, or other methods.

[0037] FIG. 1A illustrates an exemplary optical interface 100 including a transmit OPA 102 and a receive OPA 104. The transmit OPA 102 includes an array of transmit optical antennas 106 having a separation determined by a first pitch a. The receive aperture includes an array of receive optical antennas 108 having a separation determined by a second pitch a+Δa. The transmit optical antennas 106 and the receive optical antennas 108 can be, for example, optical gratings or waveguide facets (referred to as an "end-fire array" configuration). The exemplary optical interface 100 further includes an array of transmit phase shifters 110 and an array of receive phase shifters 112, which can be, for example, thermal phase shifters, electro-optical phase shifters, or micro-electromechanical phase shifters. In some examples, each of the transmit phase shifters 110 and each of the receive phase shifters 112 can be independently controlled. The transmit phase shifter 110 may modulate the direction and radiation intensity pattern of the transmit beam emitted from the transmit OPA 102, and the receive phase shifter 112 may modulate the receiver radiation intensity pattern (i.e., the receiver gain pattern). The transmitter optical power splitters 114 direct the input lightwaves 116 to each of the transmit phase shifters 110, which are optically coupled to the respective transmit optical antennas 106. In this example, the transmitter optical power splitters 114 are connected by waveguides in a binary tree arrangement. A similar or identical binary tree arrangement to house the optical power combiner 118 combines the received light into an output lightwave 120, which may then be further manipulated, transformed, or measured. Similar systems (not depicted) may also include multiple similar or identical transmit OPAs 102 or receive OPAs 104.

[0038] 1B illustrates an example of a LiDAR system 130 in which an optical interface such as the optical interface 100 shown in FIG. 1A using one or more OPAs may be used. The LiDAR system 130 includes a laser system 132, a transmitter module 134 configured to transmit light provided by the laser system 132 (e.g., by using an OPA) to a target area, and a receiver module 136 configured to receive light (e.g., by using an OPA) and coherently mix the received light with light from a local oscillator (LO) 138 (which may be derived from the laser system 132) in a coherent detector 140. A control module 142 is configured to control various aspects of the transmitter module 134 and the receiver module 136, and to estimate a distance to a target associated with a detection event based at least in part on characteristics of the scattered light received by the receiver module 136. The laser system 132 may provide a continuous wave (CW) optical signal having a narrow linewidth and low phase noise sufficient to provide a sufficiently long temporal coherence length for coherent detection over the time scales of interest, for example. In some implementations, the laser system 132 is a frequency tunable laser system in which the frequency of the light provided may be swept to perform frequency modulated continuous wave (FMCW) LiDAR measurements. A particular transmit lobe collection angle 144 from the transmitter module 134 may correspond to a particular receive lobe collection angle 146 into the receiver module 136, as described in more detail below.

[0039] Any of a wide variety of techniques may be used to steer the transmit or receive angles of the lobes of the radiation intensity pattern. Some OPAs have optical antennas with a linear distribution. Steering about a first axis perpendicular to the linear distribution may be provided, for example, by changing the relative phase shift in phase shifters coupled to each of the optical antennas. Other techniques for steering about a second axis orthogonal to the first axis (e.g., by changing the frequency of the laser system 132) may be used.

[0040] 2A-2E show exemplary hardware configurations and techniques for grating lobe signal mitigation when the phase shifters are programmed to generate linear phase fronts in the near field (e.g., there is a common phase shift used per phase shifter). Another exemplary operating configuration or mode involving phase-apodized phase fronts is described herein. With reference to FIG. 2A-2E, misalignment of the grating lobes may cause differences in the main lobe relative to the grating lobe angles of each array element. However, the direction of the main lobe can still be controlled, for example, by using phase shifters coupled to the optical antenna.

[0041] 2A shows an example radiation intensity pattern of the transmitter OPA 202 and the receiver OPA 204. The transmitter main lobe 206 and the receiver main lobe 208 overlap, while the transmitter grating lobe 210 and the receiver grating lobe 212 do not overlap or have a significantly less overlap. Such an arrangement of lobe overlap can result from tuning phase shifters associated with the transmitter optical antenna and the receive optical antenna. In this example, return signals from objects located near one of the grating lobes are attenuated due to the reduced overlap.

[0042] Figures 2B, 2C, and 2E quantitatively show an example where the TX OPA pitch and the RX OPA pitch differ by 4 nm such that the grating lobes are misaligned when the main lobes are aligned, and Figure 2D quantitatively shows an example where the TX OPA pitch and the RX OPA pitch are the same such that the grating lobes are aligned.

[0043] FIG. 2B shows a plot of a predicted example transmitter radiation intensity pattern 220 (solid line) and an example receiver radiation intensity pattern 222 (dashed line) as a function of angle relative to the center of the overlapping main lobes located at 0°.

[0044] FIG. 2C shows a plot of a predicted example transmitter radiation intensity pattern 230 (solid line) and an example receiver radiation intensity pattern 232 (dashed line) as a function of angle for a midpoint between misaligned grating lobes located at 0°.

[0045] 2B-C show the far-field TX and RX array factor patterns at the main lobe and grating lobe, respectively, depicting the relative shift of the grating lobe when the main lobe is aligned.

[0046] FIG. 2D shows a plot of a predicted exemplary radiation intensity pattern product of the transmitter OPA and the receiver OPA as a function of the angle that the transmitter OPA grating lobe and the receiver OPA grating lobe form with respect to the aligned boresight. The radiation intensity pattern product can be determined by multiplying the transmitter radiation intensity pattern by the receiver radiation intensity pattern. In some examples, other factors can contribute to the radiation intensity pattern product (e.g., antenna element factor and parallax). Since the pitch of the transmitter OPA and the pitch of the receiver OPA are the same, the grating lobes of the transmitter OPA and the grating lobes of the receiver OPA are aligned, and the main lobes of the transmitter OPA and the main lobes of the receiver OPA are aligned. The aligned grating lobes result in grating intensity peaks 240 that can be of equal amplitude of radiation intensity relative to the main intensity peak 242.

[0047] 2E shows a plot of a predicted exemplary radiation intensity pattern product of the transmitter OPA and receiver OPA as a function of the angle at which the transmitter OPA grating lobe and receiver OPA grating lobe are formed relative to the misaligned boresight. Because the pitch of the transmitter OPA and the pitch of the receiver OPA differ by 4 nm, the transmitter OPA grating lobe and the receiver OPA grating lobe are misaligned when the transmitter OPA main lobe and the receiver OPA main lobe are aligned. The misaligned grating lobe results in a grating intensity peak 250 that may be of substantially smaller amplitude of radiation intensity compared to the main intensity peak 252.

[0048] Figure 2D-E demonstrates a 24 dB reduction in the return signal from the grating lobe direction for the unmatched case (Δa = 4 nm) compared to the matched case. In general, a larger |Δa| will result in a larger relative reduction in the return signal from the grating lobe.

[0049] 3A-3C depict an example of an alternative operating configuration that may be used in some optical system implementations. In this example, the antenna array phase shifter is programmed such that one or more grating lobes of the TX OPA overlap with one or more grating lobes of the RX OPA, and the main lobe of the TX OPA is misaligned with the main lobe of the RX OPA. In this operating configuration, the LiDAR may be predominantly sensitive to return signals originating from objects within an angular range defined by the grating lobes. The direction of the grating lobes may be controlled, for example, by using the antenna phase shifter.

[0050] 3A shows an example radiation intensity pattern of the transmitter OPA 302 and the receiver OPA 304. The transmitter main lobe 306 and the receiver main lobe 308 do not overlap, while the transmitter grating lobe 310 and the receiver grating lobe 312 overlap. Thus, the return signal from any object located in the main lobe will be attenuated relative to the return signal from any object located in the grating lobe.

[0051] FIG. 3B shows a plot of a predicted example transmitter radiation intensity pattern 320 (solid line) and an example receiver radiation intensity pattern 322 (dashed line) as a function of angle for a midpoint between the misaligned main lobes located at 0°.

[0052] FIG. 3C shows a plot of a predicted example transmitter radiation intensity pattern 330 (solid line) and an example receiver radiation intensity pattern 332 (dashed line) as a function of angle relative to the center of the overlapping grating lobes located at 0°.

[0053] 3B-3C show the far-field TX and RX array factor patterns at the main lobe and grating lobe, respectively, depicting the relative shift of the main lobe when the grating lobes are aligned.

[0054] Expanding the operating range to a grating lobe aligned configuration (e.g., FIGS. 3A-3C) adds several practical capabilities to the optical system. First, this expansion effectively expands the “addressable field of view” of the system beyond the main lobe range, which has a half-width of arcsin(λ / 2a), where λ is the wavelength of light, and a is the pitch of the OPA (i.e., the distance between the optical antennas). Such configurations and techniques allow an OPA-based LiDAR system to uniquely address pointing directions at azimuth angles |θ|≧arcsin(λ / 2a). This effectively expands the azimuth scanning range of the device, where, in the course of scanning across a scene, first the TX left grating lobe and the RX left grating lobe overlap while scanning the left grating lobe range, then the TX main lobe and the RX main lobe overlap while scanning the main range, and then the TX right grating lobe and the RX right grating lobe overlap while scanning the right grating range. In general, a user may choose to scan any subset of the full 180° hemisphere in front of the LiDAR unit. Second, the ability to uniquely address objects within the main lobe and grating lobe ranges enables techniques of retroreflector grating lobe "image" (or "ghost image" or "spurious image") mitigation.

[0055] FIG. 4 shows an example of spurious image detection and mitigation. If return signals (in some cases large return signals) are measured in a direction throughout the course of scanning a scene 400 by using overlapping TX and RX lobes, a model of the system radiation intensity pattern product and antenna element factors can be used to predict the strength and direction of spurious "ghost signals" that may occur when partially non-overlapping TX and RX lobes coincide with an object that generates a reflection and a resulting return signal (which may still be relatively strong despite the degradation due to misaligned TX and RX lobes). For example, object 402A (and 402B) may generate ghost signal 404A (and 404B). The system can then annotate "that direction is affected by grating lobe ghosts" and choose to either ignore the data or locally increase the detection threshold so that only ghost signal 404A is not registered as a detected event. Such techniques can mitigate the detrimental effects of retroreflectors on application domain performance in some cases. In another example, the system may "subtract" data corresponding to ghost signals and obtain a nearly ghost-free image.

[0056] The use of selectively aligned and misaligned TX and RX grating lobes may allow objects within the grating lobe range to be uniquely targeted. For example, the left TX grating lobe and the left RX grating lobe may be overlapped to scan the left grating lobe range 406, the TX main lobe and the RX main lobe may be overlapped to scan the main lobe range 408, and the TX right grating lobe and the RX right grating lobe may be overlapped to scan the right grating lobe range 410. The direction in which the strong return signal is detected may then be used to annotate the direction in which spurious "ghost" images are expected. Data at points affected by ghosts may then be deleted or alternatively the detection threshold at those points may be increased so that the ghost signal alone does not constitute a detection. In this example, there are two grating lobes (on either side of the main lobe) corresponding to three respective ranges corresponding to different parts of the area being scanned. In other examples, there may be any number of grating lobes on either or both sides of the main lobe, leading to four or more respective ranges corresponding to different parts of the area being scanned.

[0057] 1, 2A, 3A, and 4 together illustrate various aspects of a first exemplary method for imaging an area. The first exemplary method for imaging an area includes providing light (e.g., input lightwave 116) to a plurality of optical antennas (e.g., transmit optical antenna 106) separated by a first set of spacings (e.g., pitch α, if the spacings are all equal) in a first optical phased array (e.g., transmit OPA 102), where the phase of the light provided to each optical antenna is controlled (e.g., by an array of transmit phase shifters 110) to form a transmit beam from the first OPA. The transmit beam is characterized by a first angular intensity distribution including a first lobe (e.g., transmitter main lobe 206, transmitter main lobe 306) and a second lobe (e.g., transmitter grating lobe 210, transmitter grating lobe 310). The first exemplary method of imaging the region further includes steering the angle of the beam so that a first lobe of the first angular intensity distribution scans across a first portion of the region (e.g., main lobe range 408) and so that a second lobe of the first angular intensity distribution scans across a second portion of the region (e.g., left grating lobe range 406, right grating lobe range 410). The first exemplary method of imaging an area further includes receiving light from a plurality of optical antennas (e.g., receive optical antenna 108) separated by a second set of intervals (e.g., pitch α+Δα) in a second OPA (e.g., receive OPA 104), where at least one interval in the second set of intervals is different from at least one interval in the first set of intervals, and the phase of the light provided to each optical antenna is controlled (e.g., by an array of receive phase shifters 112) to receive light into the OPA from various directions associated with a second angular intensity distribution including a first lobe (e.g., receiver main lobe 208, receiver main lobe 308) and a second lobe (e.g., receiver grating lobe 212, receiver grating lobe 312).The phase of the first OPA and the phase of the second OPA are configured such that the first lobe of the first angular intensity distribution and the first lobe of the second angular intensity distribution approximately overlap in angle during scanning of the first lobe across a first portion of the region (e.g., as shown in FIG. 2A ), and such that the second lobe of the first angular intensity distribution and the second lobe of the second angular intensity distribution approximately overlap in angle during scanning of the second lobe across a second portion of the region (e.g., as shown in FIG. 3A ).

[0058] 1, 2A, 3A, and 4 together illustrate various aspects of a second exemplary method for imaging an area. The second exemplary method for imaging an area includes providing light (e.g., input lightwave 116) to a plurality of optical antennas (e.g., transmit optical antenna 106) separated by a first set of spacings (e.g., pitch α) in a first optical phased array (e.g., transmit OPA 102), where the phase of the light provided to each optical antenna is controlled (e.g., by an array of transmit phase shifters 110) to form a transmit beam from the first OPA. The transmit beam is characterized by a first angular intensity distribution including a first lobe (e.g., transmitter main lobe 206, transmitter main lobe 306) and a second lobe (e.g., transmitter grating lobe 210, transmitter grating lobe 310). The second exemplary method of imaging an area further includes steering the angle of the beam such that a first lobe of the first angular intensity distribution scans across a first portion of the area (e.g., main lobe range 408) and a second lobe of the first angular intensity distribution scans across a second portion of the area (e.g., left grating lobe range 406, right grating lobe range 410). The second exemplary method of imaging an area further includes receiving light from a plurality of optical antennas (e.g., receive optical antenna 108) separated by a second set of intervals (e.g., pitch α+Δα) in a second OPA (e.g., receive OPA 104), where at least one interval in the second set of intervals is different from at least one interval in the first set of intervals, and the phase of light provided to each optical antenna is controlled (e.g., by an array of receive phase shifters 112) to receive light into the OPA from various directions associated with the second angular intensity distribution. A second exemplary method of imaging an area further includes characterizing a potential detected event (e.g., a ghost signal 404A) during a scan of a first lobe of the first angular intensity distribution across a first portion of the area based at least in part on whether a detected event (e.g., an object 402A) is detected during a scan of a second lobe of the first angular intensity distribution across a second portion of the area at a beam angle associated with the potential detected event.

[0059] In the examples depicted in Figures 2A-E and 3A-C, the phase shifters associated with each TX and RX antenna are set to generate approximately flat phase fronts in the near field of the TX and RX apertures. In other examples, the phase shifters may be configured to generate other non-flat phase fronts in the near field.

[0060] 5A shows a schematic diagram of an example transmitter optical phased array (TX OPA) 500 configured to generate a nearly flat transmitter phase front corresponding to the distribution of phase 502A in the near field of the beam transmitted from the transmitter aperture 506, and an example receiver optical phased array (RX OPA) 501 configured to receive a nearly flat receiver phase front corresponding to the distribution of phase 504A in the near field of the light received at the receiver aperture 508. The arrows shown in relation to the near field phase front of the transmit beam represent the relative phase shifts applied by the phase shifters to the optical field emitted from the optical antenna of the TX OPA 500. The arrows shown in relation to the near field phase front of the receive beam represent the relative phase shifts applied by the phase shifters to the optical field received at the optical antenna of the RX OPA 501. In this example, three lobes (corresponding to a main lobe and grating lobes on either side of the main lobe) begin to emerge as the optical field begins to transition from the near field to the far field of each OPA.

[0061] 5B shows plots of a predicted exemplary transmitter radiation intensity pattern 510A (dashed line) of a transmitter optical phased array and a predicted exemplary receiver radiation intensity pattern 512A (solid line) of a receiver optical phased array as a function of the angle relative to the midpoint between the peaks of the grating lobes. Referring to FIG. 5A, if the optical power is approximately uniformly distributed across the antennas in the transmitter aperture 506 and uniformly collected from the antennas in the receiver aperture 508 (e.g., by using a binary divider tree), the transmitter optical phased array 500 and the receiver optical phased array 501 will have the sinc 2 A radiant intensity pattern can be generated. Referring to FIG. 2The possibly gradual drop-off in the profile may limit the relative attenuation of ghost images located at misaligned TX and RX grating lobes relative to the main lobe. This drawback may apply to splitter networks other than binary trees, for example, that produce gradual drops in the far field due to their near-field amplitude profile.

[0062] 5C-D show additional modes of operation that may be used in some implementations to further reduce the effects of ghost images.

[0063] 5C shows a schematic diagram of an example, where phase shifters are configured to generate a spatially varying transmitter phase 502C and a spatially varying receiver phase 504C in the near field of a transmitter aperture 506 and a receiver aperture 508. The spatially varying phase shift applied by the phase shifters results in a near field profile that has a smooth and gradual falloff towards the edges of the aperture.

[0064] 5D shows plots of a predicted exemplary transmitter radiation intensity pattern 510C (dashed line) of a transmitter optical phased array and a predicted exemplary receiver radiation intensity pattern 512C (solid line) of a receiver optical phased array as a function of angle relative to the midpoint between the grating lobe peaks. The far-field beam profile corresponds to the configuration shown in FIG. 5C and shows a sharper roll-off away from the array factor peak. In some examples, the near-field phase, for sufficient grating lobe misalignment, reduces to the sinc shown in FIG. 5B. 2 The near-field phase misalignment can be designed to be further reduced compared to the degraded case. In some instances, the near-field phase misalignment can reduce the peak intensity of the main lobe, imposing a small link penalty on the overall return signal of the system. Typical values ​​for this penalty are 0-10 dB. However, for some phase plane configurations, the relative attenuation of ghost signals can greatly exceed the induced link penalty (in some cases by more than 10 dB).

[0065] In general, the various phase apodization techniques described herein may be used to improve peak roll-off and reduce ghosting (e.g., the contribution of ghost signals to the received signal that are the result of reflections from objects outside the area being scanned by the intended lobe). Furthermore, the various phase apodization techniques may be realized in a variety of implementations. Two possible manifestations (example algorithms for determining the preferred phase to apply to each antenna) are described below.

[0066] FIG. 6A is a plot of predicted example absolute values ​​of the effective amplitude of apodization as a function of antenna number, which corresponds to the relative position of each antenna in an optical phased array.

[0067] FIG. 6B is a plot of predicted example phases of apodization as a function of antenna number, which corresponds to the relative position of each antenna in the optical phased array.

[0068] 6A-B show an example of apodization by applying random phases to each antenna in the transmitter OPA (TX) and receiver OPA (RX), where each phase is selected from a suitable probability distribution (e.g., Gaussian probability distribution) with a standard deviation of variation over the width of the respective aperture by using a suitable function to generate random values ​​(e.g., by using a spurious random number generating function). The procedure of this example is as follows: First, a desired "effective amplitude apodization" profile is selected for the TX and RX apertures, as shown in FIG. 6A. The design of this apodization profile can be similar to the design of a finite impulse response low-pass filter, inasmuch as the result (i.e., the far-field profile) can be the Fourier transform of a finite list of values. In some implementations, a finite impulse response design procedure can be used for this step. Second, for each antenna, the phase is randomly selected from a Gaussian distribution with zero mean and standard deviation σ=sqrt[−2ln(A)], where A is the value of the effective amplitude apodization profile at that antenna and ln is the natural logarithm.

[0069] Figure 6B shows a plot of a prophetic example of such an antenna phase profile. The antenna phase configuration shown in Figure 6B produces a far field profile whose side lobe intensity is approximately equal to that which would result if the antenna's launch amplitude were apodized according to the design amplitude profile shown in Figure 6A (as opposed to the launch phase shown in Figure 6B). The far field profile shown in Figure 5D was produced using this methodology (e.g., the antenna phase configuration shown in Figure 6B).

[0070] Another exemplary methodology for determining the antenna phase configuration involves solving an optimization problem by incorporating a calibration model of each individual OPA LiDAR system, where the optimization figure of merit may be minimization of ghost strength (i.e., minimization of the TX / RX radiation intensity pattern product in the non-coincident region) constrained by an acceptable amount of attenuation of the radiation intensity pattern product peak or main lobe. A particularly desirable solution to this optimization problem may result in a deep (>10 dB) dip in the RX pattern noise floor coinciding with the TX peak in the non-coincident region and a deep (>10 dB) dip in the TX pattern noise floor coinciding with the RX peak in the non-coincident region, thus resulting in a TX / RX radiation intensity pattern product with highly attenuated ghosts.

[0071] 6A-B may yield additional benefits, for example, such phase apodization may lead to a sharpening of the main peak of the TX / RX radiation intensity pattern product as shown in FIG.

[0072] FIG. 7 shows a plot of predicted exemplary radiated intensity pattern products with apodization for the transmitter OPA and receiver OPA as a function of angle relative to the peak of the intensity pattern product. Compared to the unapodized linear phase front case, where the central lobe at 0° may have about 10 side lobes on either side that rise above the noise floor, phase apodization reduces this number to 2 side lobes on either side for the 2048 element TX and RX arrays in this example. The reduced number of side lobes in turn reduces the azimuth width of spurious signals associated with the coincidence of these side lobes onto strong reflectors. The practical manifestation of this effect is that the "apparent width" of retroreflectors (which may be many times wider than their physical width in the unmitigated case) can be significantly reduced. In FIG. 7, the distance between the side lobes is expressed in terms of the angular offset relative to the center of the angle on the horizontal axis, and the relative widths are compared in terms of arbitrary units (labeled as basketball diameters at a distance of 200 m).

[0073] Due to the link penalty associated with phase apodization, practical applications of the mitigation techniques may involve switching between a non-phase-apodized near-field phase profile (e.g., a linear phase profile based on a common phase shift without using a probability distribution or pseudorandom function or other deterministic phase profile (such as a lensed linear / parabolic phase profile) and a phase-apodized near-field phase profile depending on the situation. For example, the presently described OPA The LiDAR may be operated in linear / lensed linear modes during the initial scan of the scene. In areas of the scene where retroreflectors (e.g., recognized by strong return signals compared to a predefined threshold) or ghosts (annotated at known angles to the left and right of the strong return signals) are suspected, the system may reconfigure the phase shifters to generate a phase-apodized near-field phase front and rescan the affected areas, taking advantage of reduced sidelobe overlap in the former case (near retroreflectors) and reduced offset peak overlap in the latter case (at ghosts), increasing sensitivity to weaker return signals that may otherwise be masked by spurious return signals. Since the percentage of the scene affected by retroreflectors is expected to be very low (e.g., <5%), a similarly low amount of time overhead is expected in the rescanning process.

[0074] 1 and 6B together illustrate various aspects of an exemplary device. The exemplary device includes an optical phased array (e.g., transmit OPA 102) including a plurality of optical antennas (e.g., transmit optical antenna 106) separated by a set of intervals (e.g., pitch α) and a plurality of phase shifters (e.g., transmit phase shifter 110) configured to impose a phase shift on light provided to each optical antenna of the plurality of optical antennas. A phase shift control module is configured to manage the imposed phase shift, the managing including a first mode of operation. The phase shift control module may provide an analog electrical signal to the plurality of phase shifters. For example, a computer processing unit (CPU) is disposed proximate to and electrically connected to the plurality of phase shifters. A digital-to-analog converter (DAC) may be used to convert a digital signal from the exemplary phase shift control module to an analog signal that is then transmitted to the plurality of phase shifters. In another example, an application specific integrated circuit (ASIC) may be used as the phase control module. The first operating mode includes determining an imposed phase shift variation smaller than the phase shift variation range for a first subset of optical antennas within a certain distance from the center of the plurality of optical antennas (e.g., antennas having antenna numbers between 800 and 1200 in FIG. 6B ), and determining an imposed phase shift variation larger than the phase shift variation range for a second subset of optical antennas further away from the center of the plurality of optical antennas (e.g., antennas having antenna numbers less than 800 and greater than 1200 in FIG. 6B ).

[0075] In some implementations, an OPA LiDAR may use multiple receiver OPAs to capture light from one or more transmitter OPAs (e.g., multiple transmitters in a multistatic configuration). In this case, additional measures may be taken to identify or suppress back reflections. For example, a subset of the receiver apertures may be configured to overlap with the main lobe of the transmitter beam, while the remainder of the receiver apertures are configured to overlap with the grating lobes. In this case, the receiver apertures configured to overlap within the grating lobes may be monitored for strong return signals and used to reject ghost return signals that appear within the receiver apertures configured to overlap with the main lobe of the TX beam.

[0076] Multistatic LiDAR may also include information that can be used to identify and reject retroreflections, with some or all of the receiver apertures configured to overlap with the transmitter's main lobe. Prior to reaching the far field, the offset of the transmitter and receiver may cause losses at the receiver due to parallax between the transmitter and receiver. This loss may be based on geometry and may be calculated assuming that the target object is located at the transmitter's main lobe or even at the transmitter's grating lobe, as shown in FIG. 8.

[0077] Figure 8 shows predicted example disparity as a function of the angle of the target (assumed to be at 10 m distance) relative to the transmitter OPA for various optical system configurations. The difference between the expected disparity and the measurement in the main lobe can be used to mark and reject high-power retroreflectors that couple into the grating lobe.

[0078] FIG. 9 shows a plot of predicted exemplary radiation intensity patterns of the transmitter OPA and two receiver OPAs as a function of angle relative to the lobe reference angle. For a multistatic LiDAR configuration (such as shown in FIG. 9), each RX aperture may use a different pitch than the TX aperture and also the other RX apertures. In this case, different grating lobe coupling is expected for each receiver aperture. Return signals with intensity reduction due to expected mismatch in grating lobe coupling can be marked as retroreflectors and ignored. In this example, the Rx2 receiver beam grating lobe has a predictable lower overlap with the transmit beam grating lobe than the Rx1 receiver beam grating lobe.

[0079] Another spurious signal that can affect LiDAR systems, such as FMCW (frequency modulated continuous wave) LiDAR, is the frequency domain spreading of a strong reflector into neighboring frequency bins (as shown in FIG. 10). The data generated by a single FMCW LiDAR acquisition at a single polar azimuth angle consists of a finite-time complex signal. The frequency content of this signal directly reflects the light intensity received from various distances from the system, and the frequency bins are approximately proportional to the reflector distance. A strong reflector generates a strong sine wave in the frequency content. A strong sine wave generates a peak with a wide frequency spread after a finite-time Fourier transform, potentially masking weak signals below the frequency domain shoulder of the peak. One way to remove the peak and its frequency spread includes, for example, inferring the phase and amplitude of the original sine wave from the Fourier domain information, subtracting the sine wave from the time domain data, and re-evaluating the Fourier transform to reveal the previously obscured weak signals. However, other algorithms may work equally well.

[0080] 10 shows a flow diagram of an exemplary spurious image detection and mitigation scheme 1000 in an exemplary FMCW LiDAR system. A retroreflector signal 1002, a target signal 1004, and a noise signal 1006 are summed in a summer 1008. A Fourier transform 1010 is performed on the output of the summer 1008. The Fourier transform signal 1012 indicates that the detection of the target signal 1004 is masked by the retroreflector signal 1002. A mitigation algorithm 1014 attempts to convert the contribution of the retroreflector signal into the Fourier transform signal 1012. One exemplary mitigation algorithm 1014 finds peaks in the zero-padded Fourier transform signal 1012, combines a sine wave with a matching complex amplitude, and subtracts the combined sine wave from the original output of the summer 1008. Finally, the Fourier transform 1010 is re-evaluated, resulting in a retroreflector relaxation signal 1016 that contains the previously ambiguous target signal peak 1018.

[0081] In some aspects, the mitigation techniques described herein generally involve optical phased array systems in which the TX and RX optical phased arrays have unmatched grating lobes. This allows for linear or lensed linear modes of operation featuring maximum link budget as well as reduced ghosts (i.e., spurious return signals) associated with multiple grating orders launched by the TX and RX antenna arrays. This also allows for a phase apodized mode featuring further reduced grating lobe ghosts as well as reduced side lobe ghosts (reduced spurious lateral spread of strong reflectors) in exchange for a small link penalty. These techniques may involve the use of these modes separately or in tandem, where a scene is first scanned using the former mode of operation, and then areas of the scene suspected of containing spurious signals are rescanned using the latter mode of operation.

[0082] Although the present disclosure has been described in connection with several embodiments, it should be understood that the disclosure is not intended to be limited to the disclosed embodiments, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the scope of the appended claims, which claims are to be accorded the broadest interpretation so as to encompass all such modifications and equivalent arrangements as permitted under law.

Claims

1. An apparatus comprising: an optical phased array (OPA); Phase shift control module and Including, The OPA is a plurality of optical antennas separated by a set of intervals; a plurality of phase shifters configured to impose a phase shift on light provided to each optical antenna of the plurality of optical antennas; Including, the phase shift control module is configured to manage the imposed phase shift; the managing includes a first mode of operation; The first mode of operation includes: determining an imposed phase shift variation less than a phase shift variation range for a first subset of optical antennas within a distance from a center of the plurality of optical antennas; determining, for a second subset of optical antennas farther than the distance from a center of the plurality of optical antennas, a variation in the imposed phase shift that is greater than the phase shift variation range; 13. An apparatus comprising:

2. 2. The apparatus of claim 1 , wherein determining a variation in an imposed phase shift of the first subset of optical antennas and determining a variation in an imposed phase shift of the second subset of optical antennas comprises determining a variation according to a probability distribution.

3. The apparatus of claim 2 , wherein determining the variance according to a probability distribution comprises determining the variance according to a pseudorandom function.

4. The managing further includes a second mode of operation; 2. The apparatus of claim 1, wherein the second mode of operation includes determining an imposed phase shift for the first subset of optical antennas and the second subset of optical antennas that corresponds to a deterministic phase profile.

5. The apparatus of claim 4 , wherein the deterministic phase profile comprises a linear phase profile that defines a common phase shift imposed on each of the plurality of phase shifters.

6. 5. The apparatus of claim 4, wherein the managing further comprises a first scanning mode of operation, a first region being scanned by managing the imposed phase shift in the first mode of operation of two or more transmit beam angles.

7. 7. The apparatus of claim 6, wherein the managing further comprises a second scanning mode of operation, a second region being scanned by managing the imposed phase shift in the second mode of operation of two or more transmit beam angles.

8. The apparatus of claim 7 , wherein the first region is a subset of the second region.

9. The apparatus of claim 1 , further comprising a control module configured to receive scattered light emitted from the OPA and estimate a distance to a target based at least in part on a characteristic of the scattered light.

10. 10. The apparatus of claim 9, wherein the OPA is a first OPA, the apparatus further comprising a second OPA configured to receive the scattered light emitted from the first OPA and provide the scattered light to the control module.

11. The apparatus of claim 1 , wherein the intervals in the set of intervals are each greater than ½ of any wavelength in the spectrum of light provided to the multiple optical antennas in the OPA.

12. The apparatus of claim 1 , wherein all of the intervals in the set of intervals are identical to one another.

13. 1. A method comprising: providing light to a set of spaced apart optical antennas in an optical phased array (OPA), wherein a plurality of phase shifters impose a phase shift on the light provided to each optical antenna of the plurality of optical antennas; managing the imposed phase shift by using a phase shift control module; Including, the managing includes a first mode of operation; The first mode of operation includes: determining an imposed phase shift variation less than a phase shift variation range for a first subset of optical antennas within a distance from a center of the plurality of optical antennas; determining an imposed phase shift variation greater than the phase shift variation range for a second subset of optical antennas farther than the distance from a center of the plurality of optical antennas; A method comprising:

14. 14. The method of claim 13, wherein determining a variation in an imposed phase shift of the first subset of optical antennas and determining a variation in an imposed phase shift of the second subset of optical antennas comprises determining a variation according to a probability distribution.

15. The method of claim 14 , wherein determining the variance according to a probability distribution includes determining the variance according to a pseudorandom function.

16. The managing further includes a second mode of operation; 14. The method of claim 13, wherein the second mode of operation includes determining an imposed phase shift for the first subset of optical antennas and the second subset of optical antennas that corresponds to a deterministic phase profile.

17. 17. The method of claim 16, wherein the deterministic phase profile comprises a linear phase profile that defines a common phase shift imposed on each of the plurality of phase shifters.

18. 17. The method of claim 16, wherein the managing further comprises a first scanning mode of operation, a first region being scanned by managing the imposed phase shift in the first mode of operation of two or more transmit beam angles.

19. 20. The method of claim 18, wherein the managing further includes a second scanning mode of operation, and a second region is scanned by managing the imposed phase shift in the second mode of operation of two or more transmit beam angles.

20. 20. The method of claim 19, wherein the first region is a subset of the second region.

21. The method of claim 13 , further comprising receiving scattered light emitted from the OPA and estimating a distance to a target based at least in part on a characteristic of the scattered light.

22. 22. The method of claim 21, wherein the OPA is a first OPA, the method further comprising receiving scattered light emitted from the first OPA into a second OPA, and providing the scattered light for estimating the distance.

Images