On-chip monitoring and calibration circuit for frequency modulated continuous wave lidar

Abstract

A LiDAR chip of a solid-state frequency-modulated continuous-wave (FMCW) light detection and ranging (LiDAR) system. The LiDAR chip includes a light switch network and a switchable coherent pixel array (SCPA). The light switch network is configured to selectively provide coherent light to one or more of a plurality of output waveguides. The SCPA includes coherent pixels (CPs), and each CP is configured to emit the coherent light provided by a corresponding output waveguide of the plurality of output waveguides. The LiDAR chip further includes a monitoring component for calibrating the light switch network and / or an interferometer for calibrating a shape of a waveform used to generate the coherent light.

Description

[0001] Cross-references to related applications

[0002] This application claims priority to U.S. Provisional Patent Application Serial No. 62 / 965,094, filed January 23, 2020, and U.S. Provisional Patent Application Serial No. 62 / 966,983, filed January 28, 2020, both of which are incorporated herein by reference in their entirety. Technical Field

[0003] This disclosure generally relates to frequency modulated continuous wave (FMCW) optical detection and ranging (LiDAR), and more specifically, to solid-state FMCW LiDAR systems. Background Technology

[0004] Conventional frequency FMCW LiDAR uses a reference interferometer to help characterize and correct for nonlinearities in laser chirp. Traditional LiDAR or optical coherence tomography (OCT) systems rely on a reference interferometer with a balanced photodiode to indirectly estimate the laser frequency and calibrate any measurement data. To ensure accuracy of phase extraction from the reference interferometer and balanced photodiode, the reference interferometer typically needs to have a long or short delay line and careful bias control—i.e., it is located off-chip. Furthermore, long delay lines are generally a challenging problem in integrated photonics due to high losses in small waveguides. Additionally, the performance of an FMCW laser can drift across different timescales during multiple measurements under varying temperature or other environmental conditions, or throughout the lifespan of the FMCW sensor.

[0005] Furthermore, traditional FMCW LiDAR systems use mechanical moving parts and bulk optical lens elements (i.e., refractive lens systems) to steer the laser beam in two directions. And for many applications (such as automotive), they are too bulky, too expensive, and unreliable. Summary of the Invention

[0006] A LiDAR chip for a solid-state FMCW LiDAR system. The LiDAR chip includes an optical switch network, a switchable coherent pixel array (SCPA), and a monitoring component. The optical switch network is on the LiDAR chip. The optical switch network is configured to selectively provide coherent light to one or more of a plurality of output waveguides. The SCPA is on the LiDAR chip. The SCPA includes coherent pixels (CPs), and each CP is configured to emit coherent light provided by a corresponding output waveguide of the plurality of output waveguides. The monitoring component is on the LiDAR chip. The monitoring component includes a plurality of photodetectors, and each of the plurality of photodetectors is configured to generate an output signal in response to a hierarchy of light detected from a corresponding output waveguide of the plurality of output waveguides. The optical switch network is calibrated (e.g., by a controller) by adjusting the drive strength of a switch driver for the optical switch network based on the output signal from the monitoring component.

[0007] In some embodiments, the LiDAR chip includes a beam splitter, an interferometer, an optical switching network, and a SCPA. The beam splitter is on the LiDAR chip. The beam splitter is configured to split coherent light into a first portion and a second portion. The coherent light is chirped according to a waveform. The interferometer is on the LiDAR chip. The interferometer is configured to use the first portion of the coherent light to generate in-phase (I) and quadrature (Q) signals. The optical switching network is on the LiDAR chip. The optical switching network is configured to selectively provide the second portion of the coherent light to one or more of a plurality of output waveguides. The SCPA is on the LiDAR chip. The SCPA includes coherent pixels (CPs), and each CP is configured to emit coherent light provided by a corresponding output waveguide of the plurality of output waveguides. A controller is configured to identify deviations in the frequency of the coherent light, partly based on the I and Q signals, and partly based on the shape of a control waveform to compensate for the identified deviations. Note that in some embodiments, the LiDAR chip may also include monitoring components as described in the preceding paragraph.

[0008] In some embodiments, the LiDAR chip is part of the focal plane array (FPA) system of a solid-state FMCW LiDAR system. The FPA system includes a beam splitter, an interferometer, an optical switching network, a SCPA, a monitoring component, and a lens system. The beam splitter is on the LiDAR chip. The beam splitter is configured to split coherent light into a first portion and a second portion, and to chirp the coherent light according to a waveform. The interferometer is on the LiDAR chip. The interferometer is configured to use the first portion of the coherent light to generate in-phase (I) and quadrature (Q) signals. The optical switching network is on the LiDAR chip. The optical switching network is configured to selectively provide the second portion of the coherent light to one or more of a plurality of output waveguides. The SCPA is on the LiDAR chip. The SCPA includes coherent pixels (CPs), and each CP is configured to emit coherent light provided by a corresponding output waveguide of the plurality of output waveguides. The monitoring component is on the LiDAR chip. The monitoring component includes multiple photodetectors, each configured to generate an output signal in response to a hierarchy of light detected from a corresponding output waveguide of multiple output waveguides. A lens system is positioned to guide coherent light emitted from the SCPA as one or more beams into the environment, each beam emitted at a specific angle, which is based in part on the position of the CP (Concurrent Phase) on the LiDAR chip that generates the coherent light forming the one or more beams. A controller is configured to identify deviations in the frequency of the coherent light, based in part on I and Q signals, and in part on the shape of a control waveform to compensate for the identified deviations. The controller is also configured to calibrate the optical switching network based on the output signal from the monitoring component. Attached Figure Description

[0009] The embodiments disclosed herein have other advantages and features, which will become more apparent from the following detailed description and the appended claims when taken in conjunction with the examples in the accompanying drawings, in which:

[0010] Figure 1 Solid-state scanning using a switchable coherent pixel array chip on a LiDAR chip is illustrated according to one or more embodiments.

[0011] Figure 2 The basic structure and signal flow of an on-chip monitoring component according to one or more embodiments are shown.

[0012] Figure 3 This is a diagram of a cross-switching readout scheme for a linear array of monitoring photodetectors in a multi-channel switchable coherent pixel array having a reduced number of input / output (I / O) units, according to one or more embodiments.

[0013] Figure 4AThis is a diagram of a hierarchical readout scheme for a monitoring photodetector in a single-channel switchable coherent pixel array with a reduced number of I / Os, according to one or more embodiments.

[0014] Figure 4B This is a diagram of a hierarchical readout scheme for a monitoring photodetector in a multi-channel switchable coherent pixel array with a reduced number of I / Os, according to one or more embodiments.

[0015] Figure 5A This is a diagram of a hybrid coupled interferometer having post-processing feedback to a direct laser driver, according to one or more embodiments.

[0016] Figure 5B This is a diagram of a hybrid coupled interferometer having post-processing feedback to a modulator driver, according to one or more embodiments.

[0017] Figure 6 A process for laser waveform generation and FMCW calibration according to one or more embodiments is illustrated.

[0018] Figure 7 A solid-state LiDAR system incorporating an FPA system is described according to one or more embodiments. Detailed Implementation

[0019] Solid-state FMCW LiDAR systems determine depth information (e.g., distance, velocity, and acceleration of one or more objects) within the system's field of view. Solid-state FMCW LiDAR directly measures the distance and velocity of an object by directing a frequency-modulated collimated beam to a localized region. The light reflected from the object within this localized region is mixed with a tapped version of the beam (called a local oscillator (LO)). The resulting radio frequency (RF) beat signal, once corrected for the Doppler shift required for additional measurements, is proportional to the distance from the object to the solid-state FMCW LiDAR system. These two measurements, which may or may not occur simultaneously, provide information about the target's distance and velocity.

[0020] Solid-state FMCW LiDAR systems utilize on-chip monitoring and calibration circuitry to achieve high-performance solid-state beam steering and laser chirping. A solid-state FMCW LiDAR system comprises a focal plane array (FPA) system. The FPA includes one or more switchable coherent pixel arrays (SCPAs). One or more SCPAs can be positioned at the focal plane of the lens system, enabling the FPA system to perform solid-state beam steering for one-dimensional and / or two-dimensional applications. The direction of the incident beam is mapped to the discrete location of the focal point, and vice versa. A challenge for the SCPA is maintaining optimal calibration settings for the switching network to achieve low insertion loss, high extinction ratio, and low crosstalk at all times. Solid-state FMCW LiDAR systems utilize on-chip feedback mechanisms to achieve in-situ calibration or real-time closed-loop control of high-performance solid-state beam steering.

[0021] On-chip feedback mechanisms help maintain high-quality laser chirps using solid-state FMCW LiDAR, which senses distance by measuring the interference between optical signals from the local path and the target path. By scanning the laser frequency, the interference signal becomes an oscillation with a frequency proportional to the target distance. The FMCW laser is modulated to perform a linear frequency scan from low to high frequency and then from high to low frequency in a triangular pattern. Typically, lasers tuned in this way must be tuned using specific drive signals, or the frequency scan can deviate significantly from linearity. This linearity deviation leads to significant inaccuracies in distance and velocity measurements derived using FMCW LiDAR.

[0022] In some embodiments, the solid-state FMCW LiDAR system utilizes on-chip monitoring and calibration circuitry for solid-state beam steering implemented by one or more integrated SCPAs. The solid-state FMCW LiDAR system uses on-chip optical power monitoring circuitry to achieve in-situ calibration or real-time closed-loop control of the optical switching network. For example, the LiDAR chip includes (i.e., on-chip) an optical switching network, a switchable coherent pixel array (SCPA), and a monitoring component. The optical switching network is configured to selectively provide coherent light to one or more of a plurality of output waveguides. The SCPA includes coherent pixels (CPs), and each CP is configured to emit coherent light provided by a corresponding output waveguide of the plurality of output waveguides. The monitoring component includes a plurality of photodetectors, and each of the plurality of photodetectors is configured to generate an output signal in response to a hierarchy of light detected from a corresponding output waveguide of the plurality of output waveguides. The optical switching network is calibrated (e.g., by a controller) by adjusting the drive strength of a switch driver for the optical switching network based on the output signal from the monitoring component.

[0023] One or more SCPAs are placed at the focal plane of the lens system for rapid solid-state beam steering and coaxial FMCW LiDAR operation. On-chip optical monitoring circuitry with optical couplers and monitoring photodetectors (PDs) monitors the optical power at the output port of the optical switching network. This design enables in-situ calibration and real-time closed-loop control without affecting coherent pixels or interrupting normal LiDAR operation. For large-scale or multi-channel switchable coherent pixel arrays, signals can be read from any monitoring PD using cross-switch types or hierarchical (e.g., binary tree) connection schemes, while significantly reducing the number of I / Os and receivers used for monitoring circuitry.

[0024] In some embodiments, the solid-state FMCW LiDAR system utilizes on-chip monitoring and calibration circuitry to generate high-quality laser chirped signals. For example, the LiDAR chip may include (i.e., on-chip) a beam splitter, an interferometer, an optical switching network, and a SCPA. The beam splitter is configured to split coherent light into a first portion and a second portion. The coherent light is chirped according to a waveform. The interferometer is configured to use the first portion of the coherent light to generate in-phase (I) and quadrature (Q) signals. The optical switching network is configured to selectively provide the second portion of the coherent light to one or more of a plurality of output waveguides. The SCPA includes coherent pixels (CPs), and each CP is configured to emit coherent light provided by a corresponding output waveguide of the plurality of output waveguides. The controller is configured to identify deviations in the frequency of the coherent light, partly based on the I and Q signals, and to control the shape of the waveform to compensate for the identified deviations. Note that in some embodiments, the LiDAR chip may also include monitoring components as described in the preceding paragraph.

[0025] The solid-state FMCW LiDAR system uses a swept-frequency laser source and a frequency discriminator interferometer with optical mixing for laser driver calibration. In some embodiments, a programmable laser driver (current or voltage source) directly drives the tuned laser source to create alternating positive and negative frequency sweeps. In other embodiments, a programmable modulator driver directly drives the modulator to induce positive and negative frequency sweeps on a seed laser beam. Following this is the interferometer, which includes: a beam splitter that sends light along two paths, a "local" path and a "reference" path; an optical combiner called a "90-degree optical mixing"; a photodetector with multiple photodetectors; and a controller for signal processing. The controller uses the output from the optical mixer to calculate the instantaneous signal phase and laser frequency. The resulting instantaneous laser frequency is fed back to the drive signal generator to compensate for deviations in laser frequency from linearity. Furthermore, the interferometer and optical mixer can be used to calibrate any nonlinearities caused by residual errors in the pre-distortion process or laser / environment drift. Therefore, the solid-state FMCW LiDAR system performs in-situ generation of laser driver signals, as well as in-situ calibration of residual nonlinearity and laser performance drift.

[0026] Note that in some embodiments, the monitoring components for calibrating the optical switch network and the interferometer with optical hybridity for calibrating the laser driver can both be implemented on the same LiDAR chip. Therefore, the LiDAR chip can calibrate not only the optical switch network but also the laser driver.

[0027] As mentioned above, conventional LiDAR systems that use interferometers to help characterize and correct nonlinearities in laser chirp involve long or short delay lines and careful bias control. Long delay lines are problematic for integrated photonics due to high losses in small waveguides and may be located off-chip. Similarly, careful bias control is typically associated with increased complexity of the control circuitry. In contrast, solid-state FMCW LiDAR systems use short on-chip delay lines to perform laser frequency measurements without complex bias control. Furthermore, solid-state FMCW LiDAR systems are configured to measure laser frequencies and dynamically adjust the laser drive waveform to accommodate changes in laser characteristics over time or with environmental conditions.

[0028] Please note that the LiDAR chip is capable of steering light emitted from a solid-state LiDAR system in at least a first angular dimension (e.g., elevation). Furthermore, the solid-state FMCW LiDAR system may include, for example, scanning mirrors (e.g., moving mirrors, polygon mirrors, etc.) to steering the light in different angular dimensions (e.g., azimuth). And in some embodiments, one or more optical antennas within the SCPA are arranged in a two-dimensional configuration, enabling the LiDAR chip to steering the beam in two dimensions (e.g., azimuth and elevation). The ability to steering the beam without moving parts reduces many of the size, cost, and reliability issues present in conventional mechanically driven LiDAR systems.

[0029] Figure 1 Solid-state scanning using a switchable coherent pixel array chip on a LiDAR chip 106 is illustrated according to one or more embodiments. The LiDAR chip 106 is part of an FPA system configured to scan a local area. The LiDAR chip 106 is based on photonic integrated circuits (e.g., silicon photonics). The LiDAR chip 106 includes one or more FMCW LiDAR transceiver channels 101. Each FMCW LiDAR transceiver channel 101 includes an FMCW light source 102, an optical switch network 103, a monitoring component 110, and an SCPA 115. As shown, the FMCW light source 102 is directly integrated onto the LiDAR chip 106. In other embodiments, the FMCW light source 102 is not part of the LiDAR chip 106, but rather light from the FMCW light source 102 is coupled into the LiDAR chip 106 from an external source. The FMCW light source 102 may be split light on the LiDAR chip 106 or even between different FMCW LiDAR transceiver channels on different LiDAR chips. Light can also be amplified by an optical fiber amplifier or a semiconductor amplifier chip. The optical switch network 103 switches the guided light between output ports and activates the coherent pixel associated with the selected port. The optical switch network 103 is configured to selectively provide coherent light to one or more of a plurality of output waveguides coupled to various coherent pixels of the SCPA 115. The monitoring component 110 can be placed after the optical switch network 103 or anywhere on the chip that is part of the optical switch network 103. As an example, as shown, the monitoring component 110 is placed between the optical switch network 103 and the SCPA 104. The monitoring component 110 includes a plurality of photodetectors. Each of the plurality of photodetectors is configured to generate an output signal in response to the hierarchy of light detected from a corresponding output waveguide of the plurality of output waveguides. The optical switch network 103 is calibrated by adjusting the drive strength of the switch driver for the optical switch network 103 based on the output signal from the monitoring component 110. References below... Figure 2 , 34A and 4B are used to describe embodiments of the monitoring component 110.

[0030] SCPA 115 includes coherent pixels, each configured to emit coherent light provided by a corresponding output waveguide of a plurality of output waveguides. Each coherent pixel includes an optical antenna 105 for transmitting and receiving optical signals, and other passive and active optical components for generating RF signals, such as waveguides, couplers, mixers, gratings, and photodetectors. SCPA 115 is positioned at the focal plane of lens system 107. Lens system 107 includes one or more optical elements (e.g., positive lenses, freeform lenses, Fresnel lenses, etc.) that map the physical location of each coherent pixel to a unique orientation. In some embodiments, lens system 107 is positioned to collimate transmitted signals emitted via the plurality of optical antennas 105. Lens system 107 is configured to project transmitted signals emitted from the optical antennas of the plurality of antennas into a corresponding portion of the field of view of the FPA system and to reflect the transmitted signals back to the optical antennas. Each optical antenna transmits and receives light from a different angle. Thus, discrete beam scanning is achieved by switching between different antennas. The FPA system scans the target laser beam 108 across the FPA system's field of view, and coherent pixels in the FPA system generate electrical signals, which are then digitally processed to create a LiDAR point cloud. A lens system 107 generates collimated transmission signals that scan the transceiver's field of view along one or more angular dimensions (e.g., performing a two-dimensional scan of a local region). Note that by switching light to different coherent pixels, the LiDAR chip 106 emits or receives the collimated laser beam 108 at different angles, enabling discrete solid-state scanning and coaxial optical sensing of a single channel or multiple parallel channels.

[0031] Figure 2The basic structure and signal flow of an on-chip monitoring component 110 according to one or more embodiments are illustrated. The on-chip monitoring component 110 monitors coherent light emitted from the output of an optical switching network 103. The on-chip monitoring component 110 includes one or more monitoring circuits 200. The monitoring circuits 200 may include, for example, an optical coupler 202, a monitoring photodetector (PD) 203, and output waveguides 204A and 204B. The optical switching network 103 has multiple output ports. The optical switching network 103 is configured to switch light from an FMCW source between the output ports. Each output port is coupled to a corresponding output waveguide (e.g., output waveguide 201). Furthermore, some output waveguides 201 may be internally routed waveguides within the optical switch. The monitoring component 110 includes multiple optical couplers configured to tap portions of coherent light supplied to the multiple output waveguides and to supply such portions of light to multiple photodetectors. In some embodiments, each optical coupler has a different corresponding photodetector among the multiple photodetectors to which the optical coupler supplies the tapped portion of coherent light. For example, optical coupler 202 taps the optical power (i.e., coherent light) from output waveguide 201 at the output port of optical switch network 103, and provides the tapped optical power (i.e., a portion of the coherent light output from the output port) to monitoring PD 203 via output waveguide 204A. The optical power is converted into an electrical signal via monitoring PD 203. Optical coupler 202 outputs the remaining optical power via output waveguide 204B. Output waveguide 204B provides light to the next stage (e.g., coherent pixels) in FMCW LiDAR transceiver channel 101.

[0032] The electrical signals from one or more monitoring circuits 200 are then processed via receiver 206. The receiver may include, for example, an amplifier, integrator, switch, etc. The data output from receiver 206 is quantized by analog-to-digital converter (ADC) 207. The output of ADC 207 is then processed in controller 208. Controller 207 may include, for example, control circuitry, a computer processor, field-programmable gate array (FPGA), digital signal processor, microcontroller, application-specific integrated circuit (ASIC), or some combination thereof. As shown, receiver 206, ADC 207, controller 208, and switch driver 209 are separate from LiDAR chip 106. In other embodiments, some or all of receiver 206, ADC 207, controller 208, and switch driver 209 may also be integrated into LiDAR chip 106. Furthermore, although a single receiver 206 and a single ADC 207 are shown, in some embodiments, multiple receivers 206 and corresponding multiple ADCs may exist. For example, for each monitoring circuit, a separate receiver 206 and a separate corresponding ADC 207 may exist.

[0033] Closed-loop calibration and / or control are performed by adjusting the drive strength of the switch driver 209 of the optical switch network 103 based on the output of the monitoring circuit. Calibration of the optical switch network 103 helps ensure that light is delivered to one or more target CPs using minimal power and reduces light delivered to non-target CPs. Note that the calibration of the optical switch network 103 occurs within the solid-state FMCWLiDAR system and does not require external equipment. For large-scale or multi-channel switchable coherent pixel arrays, there may be hundreds of coherent pixels, hundreds of switch ports, and therefore hundreds of monitoring PDs. Therefore, in some cases, assigning individual electrical I / O pads / traces to each monitoring PD may become impractical due to electrical I / O limitations.

[0034] Figure 3 This is a diagram of a cross-switch readout scheme for a linear array of monitoring photodetectors in a multi-channel switchable coherent pixel array with a reduced number of input / output (I / O) pins, according to one or more embodiments. The cross-switch connection scheme enables signal readout from any monitoring PD 203 while significantly reducing the number of electrical I / O pins required by the monitoring circuitry. Figure 3 In this FPA system, a LiDAR chip 106 is included, and the LiDAR chip 106 includes multiple LiDAR transceiver channels 101. The LiDAR transceiver channels 101 can also be sub-blocks in a larger switching network. The LiDAR chip includes "n" channels and "N" rows of monitoring circuitry (and corresponding coherent pixels), where n and N are integers. Therefore, each monitored PD can be identified using channels and rows. Thus, each photodetector of each monitoring component has a corresponding row value ranging from 1 to N and a corresponding channel value ranging from 1 to n. For example, a monitored PD for channel "k" and row "j" is labeled "PD_k_j".

[0035] The cross-switch connection scheme is independent of the polarity of the monitoring PD. In this example, the cathodes of monitoring PDs with the same row number are connected to form corresponding signal groups (also called nodes), and the anodes of monitoring PDs with the same channel value are connected to form corresponding bias groups. For example, as shown, there are n channels, and the cathodes associated with the Nth row are connected to form corresponding signal group 303. Therefore, there are N signal groups. Similarly, the anodes of monitoring PDs with the same channel value are connected to form corresponding signal groups (also called nodes), so there are n signal groups. For example, as shown, the anodes of the monitoring PD for channel 1 are connected to form corresponding signal group 302.

[0036] Any monitored PD can be selected for readout by selecting the corresponding pixel and row number on one or two analog multiplexers (MUX) (e.g., multiplexers (MUX) 304 and MUX 306). MUX 304 and / or MUX 306 can be controlled by controller 208. For example, controller 208 can be configured to read out one or more monitored PDs. MUX 304 and 306 (e.g., switches) can be implemented on the same LiDAR chip 106 or outside of LiDAR chip 106. For example, the output of MUX 306 can be connected to a constant bias voltage 307 to provide reverse bias to the monitored PD, and a signal group (e.g., signal group 302) can be used to output a current signal. To read current from PD_k_j, the FPA system turns on all switches except switch "j" and processes the signal output from the k-th monitored PD. In this example, when switch 3 of the MUX 306 is on and the other switches are off, only the third monitoring PD in each channel is activated. The monitoring PD output can be further multiplexed to reduce the number of receiver channels using the MUX 304. This scheme enables independent optical power monitoring at all ports across the entire switch network without requiring a separate electrical I / O trace / pad for each monitoring PD. To avoid any leakage current from unselected monitoring PDs in other active channels, it is best to keep one channel active during monitoring and calibration. This can be achieved by shutting down the laser source or laser amplifier of the other channels where no monitoring PD is selected. The monitoring and calibration process can occur during power-on or frame transition.

[0037] Figure 4A and Figure 4B A layered readout scheme for monitoring photodetectors is shown, which overcomes the limitations of the cross-switch readout scheme. Figure 4A This is a diagram illustrating a hierarchical readout scheme for a monitoring photodetector in a single-channel switchable coherent pixel array with a reduced number of I / Os, according to one or more embodiments. Figure 4A In this context, the optical switch network adopts a binary tree structure. Figure 4A In this configuration, multiple optical switching units, multiple optical couplers, and multiple photodetectors are positioned to form a binary tree with multiple levels. In this example, the 1 to 8 switch tree takes a single input and routes it to one of eight coherent pixels (P0 to P7) via a three-level process using a 1x2 optical switching unit 402. Each optical switching unit is capable of diverting optical power from the input waveguide to one of the two output waveguides. In this scheme, there are optical couplers 403 and monitoring PDs 404 at both output ports of each switching unit to monitor the power flow and calibrate the switch tree in a hierarchical manner. For 1 to 8 switches, there are a total of 14 monitoring PDs.

[0038] To reduce the number of electrical I / Os, the PD bias 406 and PD output signals can be connected together. For each level in the binary tree, the outputs of the monitoring PDs with odd-numbered indices are connected together to form a first signal, and the outputs of the monitoring PDs with even-numbered indices are connected together to form a second signal. For switches 1 to 8 with 14 hierarchical monitoring PDs, the total number of electrical I / Os for optical power monitoring and the corresponding receivers 405 is reduced from 14 to 6. More generally, for switches 1 to 2... N The number of switches, monitoring I / O, and receivers ranges from 2. N+1 This decreases to 2N, and becomes more pronounced as N increases.

[0039] For example, a calibration switch is set to direct all light into the coherent pixel P2. Controller 208 begins by reading monitoring signals from receivers L0 and H0 and optimizing a control signal SW0 to maximize the L0 reading and minimize the H0 reading. The controller (e.g., controller 208) then moves to the next stage and optimizes a control signal SW1_0 to maximize H1 and minimize L1. For the final stage SW2_1, the controller attempts to maximize L2 and minimize H2. This hierarchical calibration process minimizes leakage and crosstalk from unselected photodetectors. It also electrically decouples from the electrically sensitive coherent pixel units, which requires low-noise and high-speed operation of the FMCW LiDAR. This enables in-situ calibration and real-time closed-loop control without affecting the coherent pixels or interrupting normal LiDAR operation.

[0040] Figure 4B This is a diagram of a hierarchical readout scheme for a monitoring photodetector in a multi-channel switchable coherent pixel array with a reduced number of I / Os, according to one or more embodiments. Figure 4B It shows how to Figure 4A The solution expands from a single channel to multiple channels. Figure 4B In this LiDAR chip, n channels are included (in this case n = 4, but in other embodiments n may have some other value), and each channel includes a corresponding optical switch network, a corresponding SCPA, and a corresponding monitoring component. Each channel includes multiple optical switch units, multiple optical couplers, and multiple photodetectors positioned to form a binary tree with multiple levels. Figure 4B In this configuration, the output signals from even and odd monitoring PDs at each level are bound together on the LiDAR chip 106 via different channels. As long as the same pixel is active at any given time, the controller (e.g., controller 208) is able to perform the same calibration and closed-loop control on all channels simultaneously.

[0041] Figure 5AThis is a diagram of a hybrid coupled interferometer with post-processing feedback to a direct laser driver, according to one or more embodiments. The waveform of a periodic wave (voltage or current) on the laser driver 512 is used to drive the laser 506. Typically, the output of the laser 506 is an up-scan and down-scan sequence of laser frequencies, referred to as the up-chirp and down-chirp, respectively. This laser chirp sequence is used for both FMCW detection and the frequency discrimination process described below. A beam splitter 500 converts some of the laser power to a LiDAR transceiver 505 so that the chirped laser is used as a detection field in the FMCW sensor. The LiDAR transceiver 505 may include one or more FMCW LiDAR transceiver channels 101, each including a corresponding monitoring component 110. Therefore, the hybrid coupled interferometer with post-processing feedback to a direct laser driver shown herein can also be coupled with… Figure 1 , 2 The features shown in 3, 4A, 4B, or some combination thereof. In some embodiments, the LiDAR transceiver 505 may not include the monitoring component 110. The beam splitter 500 is configured to split coherent light into a first portion and a second portion, wherein the coherent light is chirped according to a waveform. The interferometer 550 is configured to use the first portion of the coherent light to generate in-phase (I) and quadrature (Q) signals. The beam splitter 500 also converts some power to another beam splitter 501 of the interferometer 550. The beam splitter 501 divides the power between two arms, one of which has a delay arm 502. The delay arm 502 is delayed relative to the other arm, such that when the two arms are combined at the optical mixer 503, a beat signal is generated. The optical mixer 503 has four outputs configured to be optically phase-shifted relative to each other to create optical signals with 0-degree, 90-degree, 180-degree, and 270-degree phase shifts. A 0-degree and 180-degree phase-shifted signal is measured at a balanced photodetector 504 to create an "I-channel" signal, and a 90-degree and 270-degree phase-shifted signal is measured at another balanced photodetector to create a "Q-channel" signal. The I-channel and Q-channel signals are phase-shifted by 90 degrees to each other. Each channel is buffered and amplified by receiver circuitry 509 and sampled by ADC 510. The sampled I and Q signals are used by controller 511. Controller 511 may be an embodiment of controller 208. The resulting measured laser frequency is used by controller 511 to generate a new waveform to compensate for linear deviations in the laser frequency. This waveform is generated by controller 511 and used to drive laser driver 512. Deviations in the interferometer temperature can cause deviations in the effective exponent of delay arm 502, therefore temperature sensor 507 can be used to compensate for this deviation in calculations performed by controller 511.

[0042] Figure 5B This is a diagram of a hybrid coupled interferometer having post-processing feedback to a modulator driver, according to one or more embodiments. Figure 5B Basically similar Figure 5B In addition to: using a seed laser 506 and a laser modulator 514 to generate the laser chirp. The laser modulator 514 can be a phase modulator (such as a dual Mach-Zehnder modulator for I / Q modulation) or an intensity modulator. The laser modulator 514 is driven by a voltage or current signal from the modulator driver 513. The chirp generated at the output of the laser modulator 514 is transmitted through a beam splitter 500. The components within the dashed line 515 can be... Figure 5A Those identical blocks within 515. (To be consistent with...) Figure 5A In the same manner, controller 511 generates a control signal that serves as an input to modulator driver 513, which compensates for deviations in the linear laser frequency chirp at the output of laser modulator 514.

[0043] Figure 6 A process for laser waveform generation and FMCW calibration according to one or more embodiments is illustrated. The calibration process can reduce and eliminate nonlinearities from laser up-chirp and down-chirp. It can be performed by the controller of a solid-state FMCW LiDAR system. Figure 6 The process is illustrated in the diagram. In other embodiments, other entities may perform this process. Figure 6 Some or all of the steps in the process. Implementations may include different and / or additional steps, or perform these steps in a different order.

[0044] The solid-state FMCW LiDAR system loads a 605 drive waveform. For example, the microcomputer of the LiDAR processing engine in the solid-state FMCW LiDAR system can load the drive waveform. The drive waveform can be generic or a previously stored drive waveform.

[0045] The solid-state FMCW LiDAR system uses a loaded drive waveform frequency to modulate a 610 laser source. The modulated light forms one or more laser chirps. Frequency modulation can be performed by a laser controller that modulates the K-channel laser array according to instructions from the LiDAR processing engine.

[0046] Solid-state FMCW LiDAR systems measure one or more laser chirps to generate I and Q signals. For example, a solid-state FMCW LiDAR system can use an optical hybrid photodetector to measure one or more laser chirps to generate I / Q signals, as described above regarding, for example... Figure 5A and 5B As shown and described.

[0047] The solid-state FMCW LiDAR system processes 620 I and Q signals. For example, the solid-state FMCW LiDAR system can filter and / or sample the I and Q signals. The solid-state FMCW LiDAR system can use a LiDAR processing engine to process the I and Q signals.

[0048] The solid-state FMCW LiDAR system determines the phase of the processed I and Q signals. The solid-state FMCW LiDAR system can use a LiDAR processing engine to determine the phase of the processed I and Q signals. The phase can be determined, for example, by calculating the arctangent of the quotient of the processed I and Q signals. This is equivalent to measuring the phase angle of a signal created by adding the I channel to a Q channel modified by multiplying the Q channel by an imaginary number i.

[0049] Solid-state FMCW LiDAR systems use phase to determine the instantaneous frequency of a 630 laser. The instantaneous frequency of the laser can be determined, for example, by dividing the phase calculated in a previous step by the optical path time delay of the delay arm (e.g., delay arm 502).

[0050] The solid-state FMCW LiDAR system controls the drive waveform of a 635 laser source in part based on instantaneous frequency to generate a modified output beam. The system monitors the intensity of deviations in the instantaneous frequency of the laser at different time instances. Based on the intensity of these deviations, the system adjusts the drive waveform (e.g., adjusts its shape) to compensate for slower or faster chirps. This adjustment can be done either at once by updating a preloaded laser model and adjusting the drive waveform analytically, or iteratively by tuning the parameterized drive waveform using a gradient descent optimization algorithm. The controlled drive waveform is then reapplied to the laser source to generate the modified output beam. Note that steps 615 through 635 can be iterative and looped once or multiple times during calibration.

[0051] The solid-state FMCW LiDAR system uses a modified output beam to collect 640FMCW measurements. The system scans the modified output beam across a local region (e.g., via an FPA system) and measures the reflection of the modified output beam from one or more objects within that local region to generate the FMCW measurements.

[0052] Solid-state FMCW LiDAR systems use FMCW measurements to determine range and / or velocity data. The solid-state FMCW LiDAR system uses FMCW measurements to estimate range and velocity data based on the laser's expected chirp rate. If residual deviations from linearity still exist, they are measured by the same processes 605 to 630 and used to adjust the calculations of the range and / or velocity data. This process results in the generation of a more accurate point cloud.

[0053] Figure 7 A solid-state LiDAR system including an FPA system 705 is described according to one or more embodiments. The FPA system 705 may be a reciprocal system. The FPA system 705 includes a lens system 702 and a LiDAR chip 106. The LiDAR chip 106 and the FPA system 705 include as described above. Figures 1 to 6 Some or all of the components and / or some or all of the functionality are described. The CP in LiDAR chip 106 is part of one or more SPCAs (e.g., FMCW LiDAR transceiver channels 101) controlled by FPA driver 710. One or more individual CPs in LiDAR chip 106 can be activated to emit and receive light. Light emitted by LiDAR chip 106 is generated by K-channel laser array 715. K-channel laser array 715 is a laser array having K parallel channels, where K is an integer. K-channel laser array 715 can be directly integrated with LiDAR chip 106, or it can be a separate module packaged with LiDAR chip 106. K-channel laser array 715 is controlled by laser controller 720. In some embodiments, K-channel laser array 715 is tunable within a wavelength range.

[0054] The laser controller 720 receives control signals from the LiDAR processing engine 725 via a digital-to-analog converter 730. This processing also controls the FPA driver 710 and sends and receives data from the LiDAR chip 106.

[0055] The LiDAR processing engine 725 includes a microcomputer 735. The microcomputer 735 processes data from the FPA system and sends control signals to the FPA system via the FPA driver 710 and the laser controller 720. Note that the microcomputer 735 may include controller 208 and / or controller 511. The LiDAR processing engine 725 also includes an N-channel receiver 740. Signals are received by the N-channel receiver 740 and digitized using a set of M-channel analog-to-digital converters (ADCs) 745.

[0056] Note that the LiDAR chip 106 is capable of directing light emitted from the solid-state LiDAR system in one or more angular dimensions. In some embodiments, the LiDAR chip 106 is configured to direct the beam only in a first angular dimension (e.g., elevation). The FPA system 705 may include one or more scanning mirrors (not shown) capable of directing the beam in a second dimension (e.g., orthogonal to the first angular dimension—e.g., azimuth). The scanning mirrors receive light from the lens system 702 and guide the light into a target area along a specific angular field of view determined by the first angular dimension (controlled by the LiDAR chip 106) and the second angular dimension (controlled by one or more scanning mirrors). Note that the above example using one or more scanning mirrors is in a scenario where the LiDAR chip 106 is configured to scan only in the first angular dimension. However, in some embodiments, one or more scanning mirrors may be used with the LiDAR chip 106, which is configured to scan in multiple angular dimensions (e.g., azimuth and elevation). For example, using a two-dimensional arrangement of optical antennas (e.g., a rectangular grid), signals from multiple optical antennas can be scanned in two dimensions within the field of view of one or more scanning mirrors.

[0057] Additional configuration information

[0058] The accompanying drawings and the foregoing description relate to preferred embodiments only by way of illustration. It should be noted that, based on the foregoing discussion, alternative embodiments of the structures and methods disclosed herein will readily be considered as feasible alternatives that can be employed without departing from the claimed principles.

[0059] While the detailed description contains numerous details, these should not be construed as limiting the scope of the invention, but are merely illustrative of different examples. It should be understood that the scope of this disclosure includes other embodiments not discussed in detail above. Various other modifications, alterations, and variations, which will be apparent to those skilled in the art, may be made in the arrangement, operation, and details of the methods and apparatus disclosed herein without departing from the spirit and scope defined by the appended claims. Therefore, the scope of the invention should be determined by the appended claims and their legal equivalents.

[0060] Alternative embodiments can be implemented using computer hardware, firmware, software, and / or combinations thereof. Implementations can be carried out in a computer program product tangibly contained in a machine-readable storage device for execution by a programmable processor; and method steps can be executed by a programmable processor that executes an instruction program to perform a function by manipulating input data and generating output. Embodiments can be advantageously implemented in one or more computer programs that can be executed on a programmable system including at least one programmable processor coupled to receive and send data and instructions to a data storage system, at least one input device, and at least one output device. Each computer program can be implemented in a high-level procedural or object-oriented programming language, if desired, or also in assembly or machine language; and in any case, the language can be a compiled or interpreted language. For example, suitable processors include general-purpose and special-purpose microprocessors. Typically, the processor receives instructions and data from read-only memory and / or random access memory. Generally, a computer will include one or more mass storage devices for storing data files; such devices include disks, such as internal hard disks and removable disks; magneto-optical disks; and optical disks. Storage devices suitable for tangibly representing computer program instructions and data include all forms of non-volatile memory, such as semiconductor storage devices like EPROM, EEPROM, and flash memory devices; disks like internal hard disks and removable disks; magneto-optical disks; and CD-ROM disks. Any of the above can be supplemented or incorporated into them by ASICs (Application-Specific Integrated Circuits) and other forms of hardware.

Claims

1. A light detection and ranging (LiDAR) chip for a solid-state frequency modulated continuous wave (FMCW) light detection and ranging (LiDAR) system, the light detection and ranging (LiDAR) chip comprising: The optical switching network on the optical detection and ranging (LiDAR) chip is configured to selectively provide coherent light to one or more of a plurality of output waveguides; The switchable coherent pixel array (SCPA) on the optical detection and ranging (LiDAR) chip includes coherent pixels (CPs), and each of the coherent pixels (CPs) is configured to emit coherent light provided by a corresponding output waveguide among the plurality of output waveguides; The monitoring component on the LiDAR chip includes a plurality of photodetectors, and each of the plurality of photodetectors is configured to generate an output signal in response to a hierarchy of light detected from a corresponding output waveguide of the plurality of output waveguides. A first beam splitter on the LiDAR chip is configured to split coherent light into a first portion and a second portion, wherein the coherent light is chirped according to a waveform, and the second portion of the coherent light is coherent light selectively provided by the optical switching network to one or more of the plurality of output waveguides; and The interferometer on the optical detection and ranging (LiDAR) chip is configured to generate a signal using the first portion of the coherent light; The waveform shape is controlled in part based on in-phase and quadrature signals to compensate for laser frequency deviations. The optical switching network is calibrated by adjusting the drive strength of the switch driver used in the optical switching network based on the output signal from the monitoring component.

2. The light detection and ranging (LiDAR) chip of claim 1, wherein, The monitoring component includes: Multiple optical couplers are configured to tap portions of the coherent light supplied to the multiple output waveguides and to supply the portions of the light to the multiple photodetectors.

3. The light detection and ranging (LiDAR) chip of claim 2, wherein, Each of the optical couplers has a different corresponding photodetector among the plurality of photodetectors, and the optical coupler provides a split portion of the coherent light to the different corresponding photodetectors.

4. The LiDAR chip according to claim 1, wherein, The LiDAR chip includes n channels, and each channel includes a corresponding optical switch network, a corresponding switchable coherent pixel array (SCPA) including N coherent pixels, and a corresponding monitoring component. The optical switch network, the switchable coherent pixel array (SCPA), and the monitoring component are part of the first channel, and n and N are integers. Each photodetector in each of the monitoring components has a corresponding row value ranging from 1 to N, and a corresponding channel value ranging from 1 to n.

5. The light detection and ranging (LiDAR) chip of claim 4, wherein, The first electrodes of photodetectors with the same channel value are coupled together to form corresponding first nodes, such that there are n first nodes.

6. The LiDAR chip according to claim 5, wherein, The second electrodes of photodetectors with the same row value and different channel values ​​are coupled together to form corresponding second nodes, such that there are N second nodes.

7. The LiDAR chip according to claim 6, wherein, The first electrode is the anode, and the second electrode is the cathode.

8. The LiDAR chip according to claim 6, wherein, The n first nodes are electrically coupled to a first switch, and the N second nodes are electrically coupled to a second switch, and the first and second switches are configured to selectively read out any photodetector of any of the monitoring components.

9. The LiDAR chip according to claim 3 further includes a plurality of optical switching units, and the plurality of optical switching units, the plurality of optical couplers and the plurality of photodetectors are positioned to form a binary tree with multiple levels.

10. The LiDAR chip according to claim 9, wherein, For the first level of the plurality of levels The outputs of the photodetectors with odd-numbered indices are connected together to form a first node coupled to the first receiver, and The outputs of the photodetectors with even-numbered indices are connected together to form a second node coupled to the second receiver; For the second level of the plurality of levels The outputs of the photodetectors with odd-numbered indices are connected together to form a third node coupled to a third receiver, and The outputs of the photodetectors with even-numbered indices are connected together to form a fourth node coupled to a fourth receiver.

11. The LiDAR chip according to claim 10, wherein, The LiDAR chip includes n channels, and each channel includes a corresponding optical switch network, a corresponding switchable coherent pixel array (SCPA), and a corresponding monitoring component. The optical switch network, the switchable coherent pixel array (SCPA), and the monitoring component are part of the first channel, and n is an integer. Each channel includes multiple optical switch units, multiple optical couplers, and multiple photodetectors, which are positioned to form a binary tree with multiple levels. as well as For each of the n channels The outputs of the photodetectors with odd indices at the same level in the plurality of levels are connected to the first receiver, and the outputs of the photodetectors with even indices at the same level in the plurality of levels are connected to the second receiver.

12. A light detection and ranging (LiDAR) chip for a solid-state frequency modulated continuous wave (FMCW) light detection and ranging (LiDAR) system, the light detection and ranging (LiDAR) chip comprising: The first beam splitter on the LiDAR chip is configured to split coherent light into a first part and a second part, wherein the coherent light is chirped according to a waveform. The interferometer on the optical detection and ranging (LiDAR) chip is configured to use the first portion of the coherent light to generate in-phase (I) and quadrature (Q) signals; The optical switching network on the optical detection and ranging (LiDAR) chip is configured to selectively provide the second portion of the coherent light to one or more of a plurality of output waveguides; The switchable coherent pixel array (SCPA) on the optical detection and ranging (LiDAR) chip includes coherent pixels (CP), and each of the coherent pixels (CP) is configured to emit coherent light provided by a corresponding output waveguide among the plurality of output waveguides; The controller is configured as follows: The frequency deviation of the coherent light is identified in part based on the in-phase signal and the quadrature signal, and Part of it is based on controlling the shape of the waveform to compensate for the identified deviation.

13. The LiDAR chip according to claim 12 further comprises: A temperature sensor is configured to monitor the temperature of the delay arm of the interferometer, wherein the controller uses the monitored temperature to compensate for temperature-induced deviations in the refractive index of the delay arm.

14. The LiDAR chip according to claim 12, wherein, The interferometer includes: A second beam splitter is configured to split the first portion of the coherent light into a first arm and a second arm, wherein the second arm introduces a delay; An optical mixing combiner configured to receive light output from a first arm and a second arm, and to output light optically phase-shifted relative to each other across a first output, a second output, a third output, and a fourth output; A first balanced photodetector, configured to generate the in-phase signal using the first output and the third output; and A second balanced photodetector is configured to use the second output and the fourth output to generate the quadrature signal.

15. The LiDAR chip according to claim 12, wherein, The coherent light is generated using a seed laser and a laser modulator, and the laser module is driven by a modulator driver. The controller controls the shape of the waveform by controlling the modulator driver.

16. The LiDAR chip according to claim 12, wherein, The in-phase signal and the quadrature signal are processed, and the controller is configured to: Determine the phase of the in-phase and quadrature signals being processed; The instantaneous frequency of the coherent light is determined using the phase of the in-phase signal and the quadrature signal; and The determined instantaneous frequency is used to identify deviations in the frequency of the coherent light.

17. The LiDAR chip according to claim 16, wherein, The in-phase signal is buffered and amplified using a first receiver, and then sampled using a first analog-to-digital converter (ADC) to generate the processed in-phase signal. The quadrature signal is buffered and amplified using a second receiver, and then sampled using a second analog-to-digital converter (ADC) to generate the processed quadrature signal.

18. The LiDAR chip according to claim 12, further comprising: The monitoring component on the LiDAR chip includes a plurality of photodetectors, each of which is configured to generate an output signal in response to a hierarchy of light detected from a corresponding output waveguide among the plurality of output waveguides. The optical switching network is calibrated by adjusting the drive strength of the switch driver used in the optical switching network based on the output signal from the monitoring component.

19. A focal plane array (FPA) system for a solid-state frequency modulated continuous wave (FMCW) optical detection and ranging (LiDAR) system, the focal plane array (FPA) system comprising: A first beam splitter on a LiDAR (Light Detection and Ranging) chip, the first beam splitter being configured to split coherent light into a first part and a second part, wherein the coherent light is chirped according to a waveform; The interferometer on the optical detection and ranging (LiDAR) chip is configured to use the first portion of the coherent light to generate in-phase (I) and quadrature (Q) signals; The optical switching network on the optical detection and ranging (LiDAR) chip is configured to selectively provide the second portion of the coherent light to one or more of a plurality of output waveguides; The switchable coherent pixel array (SCPA) on the optical detection and ranging (LiDAR) chip includes coherent pixels (CP), and each of the coherent pixels (CP) is configured to emit coherent light provided by a corresponding output waveguide among the plurality of output waveguides; The monitoring component on the LiDAR chip includes a plurality of photodetectors, each of which is configured to generate an output signal in response to a hierarchy of light detected from a corresponding output waveguide among the plurality of output waveguides. A lens system is positioned to guide coherent light emitted from the switchable coherent pixel array (SCPA) as one or more beams into the environment, and each of the one or more beams is emitted at a specific angle, and the specific angle is based in part on the position of the coherent pixels (CP) on the light detection and ranging (LiDAR) chip that generate the coherent light forming the one or more beams. The controller is configured as follows: The frequency deviation of the coherent light is identified partly based on the in-phase signal and the quadrature signal. Partially based on controlling the shape of the waveform to compensate for the identified deviation; and The optical switch network is calibrated based on the output signal from the monitoring component.

Images