WO2024131914A1

WO2024131914A1 - Method for dynamically adjusting single-photon detector and idar

Info

Publication number: WO2024131914A1
Application number: PCT/CN2023/140781
Authority: WO
Inventors: Feng Yuan; Shuaiqi XU; Shaoqing Xiang
Original assignee: Hesai Technology Co., Ltd.
Priority date: 2022-12-21
Filing date: 2023-12-21
Publication date: 2024-06-27
Also published as: CN118225258A

Abstract

Methods, devices, apparatuses, and systems for dynamically adjusting a single-photon detector include: determining a current environment noise level; and when the environment noise level saturates the single-photon detector within a predetermined duration, adjusting a driving voltage of the single-photon detector to reduce the photon detection efficiency of the single-photon detector.

Description

METHOD FOR DYNAMICALLY ADJUSTING SINGLE-PHOTON DETECTOR AND LIDAR

CROSS-REFERENCE TO RELATED APPLICATION (S)

This application claims priority to Chinese Patent Application No. 202211647680.7, filed on December 21, 2022, the content of which is incorporated herein by reference in its entirety.

TECHNICAL FIELD

This disclosure typically relates to a technical field of light detection, in particular to a method for dynamically adjusting a single-photon detector and a LiDAR.

BACKGROUND

Single-photon detectors (e.g. single-photon avalanche diodes or “SPADs” ) are commonly used for performing photoelectric detection. For example, the single-photon detector can be used in a LiDAR as a detector for performing time-of-flight ( “ToF” ) ranging. A SPAD receives a light signal and generates a photocurrent. For SPAD devices, avalanche phenomenon can occur when the SPAD devices are excited by photons, and the SPAD devices can be saturated and a light response pulse can be generated. The SPAD cannot detect photons for a certain period of time. The time period during which the SPAD cannot perform detection is called recovery time or dead time. The length of this recovery time affects total count of photons per unit time that the detector can output (also referred to as a maximum count rate or “MCR” ) , which refers to a maximum number of light response pulses that a detector can output per second. A LiDAR based on a planar array of SPADs can have weaker detection capabilities for objects with high reflectivity compared with objects with low reflectivity in strong environmental lighting conditions. This is not ideal for a LiDAR, which is a problem that can be addressed.

SUMMARY

To take into account the detection capabilities for objects with high reflectivity in strong noise conditions and improve the overall detection performance of the LiDAR, this disclosure provides a method for feedback-adjusting the SPAD (s) by measuring environment noise, which can balance the detection capability for objects with low reflectivity and objects with high reflectivity under this noise.

In view of at least one disadvantage in the existing technology, this disclosure provides a method for dynamically adjusting a single-photon detector. The method includes: S1, determining a current environment noise level; and S2, when the environment noise level saturates the single-photon detector within a predetermined duration, adjusting a driving voltage of the single-photon detector to reduce a photon detection efficiency of the single-photon detector.

Optionally, the step S1 includes: S11, determining a number of noise pulses in an output signal of the single-photon detector based on the output signal; and S12, determining the environment noise level based on the number of noise pulses.

Optionally, the step S12 includes: S121, determining a first number of noise pulses of the single-photon detector under a current photon detection efficiency; S127: reducing the photon detection efficiency of the single-photon detector by a predetermined amplitude, and determining a second number of noise pulses; and S128, determining whether the single-photon detector is saturated based on a change trend, and the change trend is determined by the first number of noise pulses and the second number of noise pulse.

Optionally, the step S12 includes: S120, based on a mapping relationship between the number of input noise photons and the number of noise pulses, determining the number of noise photons incident on the single-photon detector based on the number of noise pulses on the basis of a state where the single-photon detector is not saturated; S122, calculating a first value of the photon detection efficiency of the single-photon detector based on the number of noise photons; S123, adjusting the driving voltage of the single-photon detector to adjust the photon detection efficiency of the single-photon detector to the first value; S124, determining the number of noise pulses in the output signal of the single-photon detector based on the output signal after adjusting the photon detection efficiency; S125, determining whether the single-photon detector is saturated based on the numbers of noise pulses determined in step S11 and the numbers of noise pulses determined in step S124; and S126, when the single-photon detector is saturated, reducing the photon detection efficiency of the single-photon detector by a predetermined amplitude; and when the single-photon detector is not saturated, maintaining the photon detection efficiency of the single-photon detector as the first value.

Optionally, the step S122 includes: calculating the first value of the photon detection efficiency of the single-photon detector based on both N_noise1 and Deadtime, the N_noise1 indicates the number of noise photons incident on the single-photon detector, and the Deadtime indicates dead time of the single-photon detector.

Optionally, the step S12 includes: when the environment noise level does not saturate the single-photon detector, adjusting the driving voltage of the single-photon detector to increase the photon detection efficiency of the single-photon detector to the first value.

Optionally, the step S12 includes: when the environment noise level saturates the single-photon detector, adjusting the photon detection efficiency of the single-photon detector can change a reflectivity of an object corresponding to a peak value of the output signal of the single-photon detector.

Optionally, the step S11 includes: determining the number of noise pulses based on an output of the single-photon detector within a time window; or determining in advance a ratio or position of a valid signal within the time window and determining the number of noise pulses based on the output of the single-photon detector except the valid signal within the time window; or determining the number of noise pulses based on the output of the single-photon detector beyond the time window; or determining the number of noise pulses based on the output of the single-photon detector not used for ranging currently; or determining the number of noise pulses based on the output of the single-photon detector during non-ranging periods within the time window.

Optionally, saturation of the single-photon detector corresponds to a situation where the single-photon detector reaches the maximum count rate.

Optionally, the method further includes: S3, when the environment noise level does not saturate the single-photon detector within the predetermined duration, adjusting the driving voltage of the single-photon detector to increase the photon detection efficiency of the single-photon detector.

This disclosure also provides a LiDAR. The LiDAR includes a transmitter unit, a receiver unit and a controller. The transmitter unit is configured to transmit a detection light to detect an object. The receiver unit includes a planar array of single-photon detectors configured to receive an echo of the detection light reflected on the object. The controller is coupled to the transmitter unit and the receiver unit and is configured to perform operations of: S1, determining a current environment noise level; and S2, when the environment noise level saturates the single-photon detector within a predetermined duration, adjusting a driving voltage of the single-photon detector to reduce a photon detection efficiency of the single-photon detector.

Optionally, the operation S1 includes: S11, determining the number of noise pulses in an output signal of the single-photon detector based on the output signal; and S12, determining the environment noise level based on the number of noise pulses.

Optionally, the operation S12 includes: S121, determining a first number of noise pulses of the single-photon detector under a current photon detection efficiency; S127: reducing the photon detection efficiency of the single-photon detector by a predetermined amplitude, and determining a second number of noise pulses; and S128, determining whether the single-photon detector is saturated based on a change trend, and the change trend is determined by the first number of noise pulses and the second number of noise pulse.

Optionally, the operation S12 includes: S120, based on a mapping relationship between the number of input noise photons and the number of noise pulses, determining the number of noise photons incident on the single-photon detector based on the number of noise pulses on the basis of a state where the single-photon detector is not saturated; S122, calculating a first value of the photon detection efficiency of the single-photon detector based on the number of noise photons; S123, adjusting the driving voltage of the single-photon detector to adjust the photon detection efficiency of the single-photon detector to the first value; S124, determining the number of noise pulses in the output signal of the single-photon detector based on the output signal after adjusting the photon detection efficiency; S125, determining whether the single-photon detector is saturated based on the numbers of noise pulses determined in the operation S11 and the numbers of noise pulses determined in the operation S124 ; and S126, when the single-photon detector is saturated, reducing the photon detection efficiency of the single-photon detector by a predetermined amplitude; and when the single-photon detector is not saturated, maintaining the photon detection efficiency of the single-photon detector as the first value.

Optionally, the operation S122 includes: calculating the first value of the photon detection efficiency of the single-photon detector based on both N_noise1 and Deadtime, the N_noise1 indicates the number of noise photons incident on the single-photon detector, and the Deadtime indicates recovery time of the single-photon detector.

Optionally, the operation S3 includes: when the environment noise level does not saturate the single-photon detector, adjusting the driving voltage of the single-photon detector to increase the photon detection efficiency of the single-photon detector to the first value.

Optionally, the operation S3 includes: when the environment noise level saturates the single-photon detector, adjusting the photon detection efficiency of the single-photon detector can change a reflectivity of an object corresponding to a peak value of the output signal of the single-photon detector.

Optionally, the operation S11 includes: determining the number of noise pulses based on an output of the single-photon detector within a time window; or determining in advance a ratio of a valid signal within the time window and determining the number of noise pulses based on the output except the valid signal within the time window; or determining the number of noise pulses based on the output of the single-photon detector beyond the time window; or determining the number of noise pulses based on the output of the single-photon detector not used for ranging currently; or determining the number of noise pulses based on the output of the single-photon detector during non-ranging periods within the time window.

Optionally, the controller is further configured to perform operations of: S3, when the environment noise level does not saturate the single-photon detector within a predetermined duration, adjusting the driving voltage of the single-photon detector to increase the photon detection efficiency of the single-photon detector.

This disclosure provides a method for feedback-adjusting the PDE of SPAD (s) by measuring environment noise, so that the LiDAR can balance the detection capability for objects with high reflectivity and objects with low reflectivity under different noise levels. When the environment noise exceeds a predetermined threshold, to take in to account the detection capability for objects with high reflectivity under the high noise, the PDE can be reduced by adjusting a voltage over break voltage ( “VOV” ) appropriately, so as to make the detection capability of the LiDAR for objects with high reflectivity not lower than that for objects with low reflectivity under the high noise as far as possible. In addition, when the environment noise is lower than the predetermined threshold, a maximum PDE that can be set in a LiDAR can improve the detection capability for objects with high reflectivity and objects with low reflectivity. Consistent with embodiments of this disclosure, the overall detection performance of LiDAR can be improved by adjusting PDE based on magnitude of different environment noises.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings are provided to help further understand this disclosure and form part of the specification. The drawings are used in conjunction with some embodiments of this disclosure to explain this disclosure, which do not form limitation to this disclosure.

FIG. 1 shows a relationship between the reflectivity of objects and the signal-to-noise ratio of the output of a detector under different environment light.

FIG. 2a shows a variation of a detection probability of a LiDAR for objects with high reflectivity when using different PDEs under strong noise.

FIG. 2b shows a variation of the detection probability of a LiDAR for objects with low reflectivity when using different PDEs under strong noise.

FIG. 3 shows an example method for dynamically adjusting SPAD (s) , consistent with some embodiments of this disclosure.

FIG. 4 shows an example scenario of determining an environment noise level based on the output of the LiDAR, consistent with some embodiments of this disclosure.

FIG. 5 shows an example planar array LiDAR, consistent with some embodiments of this disclosure.

FIG. 6a shows a curve of an example mapping relationship between the total count of output noise photons and incident noise photons, consistent with some embodiments of this disclosure.

FIGS. 6b and 6c show curves of the example variation of the output of the SPAD (s) with the PDE of the SPAD (s) when the noise level is fixed, consistent with some embodiments of this disclosure.

FIG. 7 shows an example method for determining the environment noise level based on the output of the LiDAR, consistent with some embodiments of this disclosure.

FIG. 8 shows example detection results of SPAD (s) for objects with different reflectivity before and after adjusting PDE using the method, consistent with some embodiments of this disclosure.

FIG. 9 shows an example method for performing step S12, consistent with some embodiments of this disclosure.

FIG. 10 shows an example circuit diagram using SPAD (s) , consistent with some embodiments of this disclosure.

FIG. 11 shows an example process of multiple detections, consistent with some embodiments of this disclosure.

FIG. 12 shows an example accumulation result of a histogram, consistent with some embodiments of this disclosure.

FIG. 13 shows an example structure of a planar array LiDAR, consistent with some embodiments of this disclosure.

FIG. 14 shows an example method of this disclosure perform by a LiDAR to perform this disclosure, consistent with some embodiments of this disclosure.

DETAILED DESCRIPTION

In the following, only some example embodiments are described. The described embodiments may be modified in multiple different ways without departing from the spirit or scope of this disclosure, as can be obvious to those skilled in the art. Accordingly, the drawings and descriptions are to be regarded as illustrative and not restrictive in nature.

In the description of this disclosure, it needs to be understood that the orientation or position relations denoted by such terms as “central, ” “longitudinal, ” “latitudinal, ” “length, ” “width, ” “thickness, ” “above, ” “below, ” “front, ” “rear, ” “left, ” “right, ” “vertical, ” “horizontal, ” “top, ” “bottom, ” “inside, ” “outside, ” “clockwise, ” “counterclockwise, ” or the like are based on the orientation or position relations for example, as shown in the accompanying drawings, and are used only for the purpose of facilitating description of this disclosure and simplification of the description, instead of indicating or suggesting that the denoted devices or elements must be oriented specifically, or configured or operated in a specific orientation. Thus, such terms should not be construed to limit this disclosure. In addition, such terms as “first” and “second” are only used for the purpose of description, rather than indicating or suggesting relative importance or implicitly indicating the number of the denoted technical features. Accordingly, features defined with “first” and “second” may, expressly or implicitly, include one or more of the features. In the description of this disclosure, “plurality” means two or more, unless otherwise defined explicitly and specifically.

In the description of this disclosure, it needs to be noted that, unless otherwise specified and defined explicitly, such terms as “installation” “coupling” and “connection” should be broadly understood as, for example, fixed connection, detachable connection, or integral connection; or mechanical connection, electrical connection or intercommunication; or direct connection, or indirect connection via an intermediary medium; or internal communication between two elements or interaction between two elements. For those skilled in the art, the specific meanings of such terms herein can be construed in light of the specific circumstances.

Herein, unless otherwise specified and defined explicitly, if a first feature is “on” or “beneath” a second feature, this may cover direct contact between the first and second features, or contact via another feature therebetween, other than the direct contact. Furthermore, if a first feature is “on, ” “above, ” or “over” a second feature, this may cover the case that the first feature is right above or obliquely above the second feature, or just indicate that the level of the first feature is higher than that of the second feature. If a first feature is “beneath, ” “below, ” or “under” a second feature, this may cover the case that the first feature is right below or obliquely below the second feature, or just indicate that the level of the first feature is lower than that of the second feature.

The disclosure below provides many different embodiments or examples so as to realize different structures described herein. To simplify the disclosure herein, the following can give the description of the parts and arrangements embodied in specific examples. They are only for the example purpose, not intended to limit this disclosure. Besides, this disclosure may repeat a reference number and/or reference letter in different examples, and such repeat is for the purpose of simplification and clarity, which does not represent any relation among multiples embodiments and/or arrangements as discussed. In addition, this disclosure provides examples of multiple specific processes and materials, but those skilled in the art can also be aware of application of other processes and/or use of other materials. The terms ″or″ and ″and/or″ of this disclosure describe an association relationship between associated objects, and represent a non-exclusive inclusion. For example, each of ″A and/or B″ and ″A or B″ can include: only ″A″ exists, only ″B″ exists, and ″A″ and ″B″ both exist, where ″A″ and ″B″ can be singular or plural. For another example, each of ″A, B, and/or C″ and ″A, B, or C ″ can include: only ″A″ exists, only ″B″ exists, only ″C″ exists, ″A″ and ″B″ both exist, ″A″ and ″C″ both exist, ″B″ and ″C″ both exist, and ″A″ , ″B″ , and ″C″ all exist, where ″A, ″ ″B, ″ and ″C″ can be singular or plural.

Some embodiments of this disclosure are described below with reference to the drawings. It should be appreciated that the embodiments described here are only for the purpose of illustrating and explaining, instead of limiting, this disclosure.

Single-photon detectors, such as single-photon avalanche diodes ( “SPADs” ) , refer to avalanche photodiodes ( “APDs” ) that operate in Geiger mode in this disclosure. Single-photon detectors can perform single-photon detection, in which a single photon can be detected. The process of photon detection can include the following steps. A certain reverse bias voltage Vbias to the avalanche photodiode can be applied. Photons carrying energy can be incident on a P-N junction of the avalanche photodiode and transfer energy to electrons on covalent bond. The electrons can be caused to break away from the covalent bond and form electron-hole pairs, which can be referred to as photogenerated carriers. If the reverse bias voltage Vbias is high enough, the photogenerated carriers in a depletion layer can obtain sufficient kinetic energy, and cause the covalent bond to break when colliding with the lattice and generate more electron-hole pairs. This process can also be referred to as collision ionization. The new carriers can produce new collision ionization continuously, which can form a chain effect, resulting in an avalanche multiplication effect of carriers, and eventually obtaining a pulse current of a magnitude of mA that is large enough to be detected and can help finish the single-photon detection. For a single SPAD device, photon detection efficiency ( “PDE” ) is an important parameter of the SPAD (s) . PDE represents an average probability that photons can excite avalanches after the photons entering the SPAD (s) device and being detected. The PDE can be expressed by:
PDE=εgeo*QE*εtrigger (Equation 1)

where, εgeo represents a geometric fill factor. QE represents a quantum efficiency, which is a probability of generating electron-hole pairs. εtrigger represents the probability of further exciting avalanches by electron-hole pairs.

PDE can also represent the capability of the SPAD (s) device to detect single-photon signals and can be expressed as the number of detected photons divided by the total number of incident photons.

To take into account the detection capability for objects with high reflectivity under strong noise and improve the overall detection performance of LiDAR, this disclosure provides a method for feedback-adjusting PDE through measuring the environment noise. This method allows the LiDAR to balance the detection capabilities for objects with low reflectivity and objects with high reflectivity under different noise levels. When the environment noise exceeds a predetermined threshold, to take into account the detection capability for objects with high reflectivity in this high noise environment, the PDE can be reduced appropriately by adjusting the voltage over break voltage ( “VOV” ) . By doing so, the detection capability for objects with high reflectivity can be greater than or equal to the detection capability for objects with low reflectivity in this high noise environment as far as possible. In addition, when the environment noise is below the preset threshold, a maximum PDE that can be set in the LiDAR system can improve the detection capabilities for both objects with high reflectivity and objects with low reflectivity. By using the technical solution of this disclosure, the detection capability of the LiDAR can be improved by adjusting the PDE based on the magnitude of different environment noises.

Before describing the technical solutions of some embodiments of this disclosure in detail, the technical terms to be used in this disclosure are explained first.

The number of noise photon can also be referred to as the number of incident noises. The number of incident noises can represent the number of photons incident on the surface of the SPAD (s) array within a certain duration, which can reflect the intensity of external environment light.

A total count of photons or the number of responded photons can represent a sum of the number of pulses output after being responded by the SPAD (s) , which is the output of the SPAD (s) .

A maximum count rate ( “MCR” ) can represent a characteristic of the SPAD (s) device, such as the maximum value that the SPAD (s) device can count per unit time. The status that the output of the SPAD (s) reaches the maximum count rate can represent that the SPAD (s) is saturated.

Referring to FIG. 1, FIG. 1 illustrates a curve of variation of the signal-to-noise ratio ( “SNR” ) with the reflectivity of the object during object detection of the LiDAR under different environment noise. The environment noise corresponding to curves 1-10 becomes higher gradually. For example, curve 1 corresponds to environment noise of 100 W/ (m*m*um) , curve 6 corresponds to environment noise of 600 W/ (m*m*um) , and curve 10 corresponds to environment noise of 1000 W/ (m*m*um) . For example, still referring to FIG. 1, FIG. 1 shows that under high noise, for example, when the solar noise exceeds 600 W/ (m*m*um) , the SNR of the LiDAR for objects with 10%reflectivity is better than the SNR for objects with 100%reflectivity.

However, a detection capability of a LiDAR for an object with 10%reflectivity under strongest noise is typically considered as inferior detection capability of the LiDAR. Users of the LiDAR generally do not accept a LiDAR having a detection performance for objects with high reflectivity lower than the detection performance for objects with low reflectivity, which is also a problem.

In addition, a long distance ranging capability of the LiDAR for objects with 100%reflectivity when using different PDEs under a condition where an environment light is 100 K lx is simulated. A long distance ranging capability of the LiDAR for objects with 10%reflectivity when using different PDEs under a condition where an environment light is 100 K lx is also simulated. 1Klx = 10 W/(m*m*um) , and such environment light can be understood as strong environment light, such as the situation where the noise is strong enough to saturate the SPAD (s) . For example, referring to FIGS. 2a and 2b, FIG. 2a illustrates variation of the long distance ranging capability of the LiDAR for objects with high reflectivity, such as objects with 100%reflectivity, under different PDEs; and FIG. 2b illustrates the long distance ranging capability of the LiDAR for objects with low reflectivity, such as objects with 10%reflectivity, under different PDEs. In FIGS. 2a and 2b, PDE varies with VOV, each curve corresponding to a different VOV. VOVs corresponding to curves 1-8 increases gradually. For example, the VOV corresponding to curve 1 can be 4.3V, the VOV corresponding to curve 5 can be 6.3V, and the VOV corresponding to curve 8 can be 7.8V.

For example, still referring to FIG. 2a, as the VOV decreases, the long distance ranging capability of the LiDAR for objects with 100%reflectivity improves gradually. This is because at the same expected detection distance, the probability for the LiDAR detecting objects at that distance is higher when the VOV is lower. For example, at the same detection probability, when the VOV is lower, the LiDAR can detect objects at a greater distance.

For example, still referring to FIG. 2b, FIG. 2b shows that as the VOV varies, the long distance ranging capability of the LiDAR for objects with 10%reflectivity is basically unchanged. This is because at the same expected detection distance, the probability for the LiDAR detecting objects at that distance is basically the same for at least some VOVs.

In some embodiments, based on the simulation results in FIGS. 2a and 2b, FIGS. 2a and 2b show that, under strong noise, reducing the PDE appropriately has a significant benefit for improving long distance ranging capability for objects with high reflectivity, but does not significantly worsen the detection capability for objects with low reflectivity.

Consistent with embodiments of this disclosure, methods for dynamically adjusting SPAD (s) are provided in this disclosure. In some embodiments, a method can include: monitoring the environment noise condition of the SPAD (s) , and when the environment noise is low, increasing the PDE; when the environment noise is high, reducing the PDE. The detection capability of the LiDAR for objects with high reflectivity at this time can be equal to the detection capability for objects with low reflectivity. By doing so, the overall detection performance of the LiDAR can be improved.

For example, referring to FIG. 3, FIG. 3 shows a method 100 for dynamically adjusting the single-photon detector SPAD (s) in some embodiments of this disclosure. The method is described as follows in detail with reference to FIG. 3.

At step S1, a current environment noise level can be determined. In some embodiments of this disclosure, the current environment noise level can be determined based on the output of the SPAD (s) .

At step S2, when the environment noise level saturates the SPAD (s) within a predetermined duration, a driving voltage of the SPAD (s) can be adjusted to reduce the photon detection efficiency of the SPAD (s) .

For a SPAD (s) device, there is recovery time in an order of nanoseconds after the SPAD device is excited by photons and a light response pulse is generated. During this recovery time, the SPAD (s) device cannot respond to incident light signals, and thus this period is referred to as dead time. For a SPAD (s) device, the length of the dead time can affect the maximum number of light response pulses that a detector can output per second. For example, the Maximum Count Rate (MCR) . In some embodiments, saturation of the SPAD (s) refers to a situation that a single-photon detector reaches its maximum count rate, under an external environment noise level and a predetermined reflectivity of object, at a predetermined distance, within a predetermined duration. For example, if the external noise, such as, the solar noise is 700W/ (m*m*um) and the reflectivity of the object is 90%, at a distance of 50m, the single-photon detector reaches its maximum count rate within 1s, which indicates that the external noise level saturates the SPAD (s) . Additionally, this predetermined duration can also be set as a detection time window. For example, a duration reserved for receiving light after the LiDAR transmits light, which can be related to an expected maximum detection distance of the LiDAR.

In addition, in some embodiments of this disclosure, the method 100 further includes step S3, adjusting the driving voltage of the SPAD (s) to improve the photon detection efficiency of the SPAD (s) when the environment noise level does not saturate the SPAD (s) .

As mentioned above, because PDE of the SPAD (s) is typically less than 1, the SPAD (s) can respond to a portion of photons incident thereon based on a certain probability, and a current or voltage signal is generated after the SPAD (s) is triggered. The photon detection efficiency of the SPAD (s) can also be understood as a probability of the SPAD (s) responding to photons incident thereon. The photon detection efficiency of the SPAD (s) is typically related to a voltage difference between the cathode and anode of the SPAD (s) . In this disclosure, the saturation of the SPAD (s) refers to that within a duration, the output of one or more SPAD (s) included in one pixel reaches the maximum count rate. For example, reaching the maximum value of responses that can be counted.

In some embodiments of this disclosure, the photon detection efficiency PDE of the SPAD (s) to be dynamically adjusted can be increased or decreased by a predetermined amplitude. By comparing the output of the SPAD (s) before and after the increment or decrement, whether the SPAD (s) has reached saturation is determined based on the comparison. During this dynamic adjustment of PDE, the reflectivity and distance of external objects do not change. In addition, when the environment noise level saturates the single-photon detector within a predetermined duration, the detection probability of the LiDAR for objects with high reflectivity (e.g., objects with 100%reflectivity) is lower than that for objects with low reflectivity (e.g., objects with 10%reflectivity) .

In some embodiments of this disclosure, when the SPAD (s) is currently in a high noise environment, such as in harsh sunlight at noon, where the noise level is sufficient to saturate the SPAD (s) to be dynamically adjusted. For example, referring to FIG. 1 to FIG. 2b, this disclosure proposes to reduce the photon detection efficiency of the SPAD (s) by reducing the voltage difference across two ends of the SPAD (s) , which can improve the detection capability of the SPAD (s) for objects with high reflectivity while maintaining the detection capability for objects with low reflectivity. Thus, the overall detection performance of the LiDAR under this environment noise is improved.

In some embodiments of this disclosure, when the SPAD (s) is currently in a relatively low noise environment, such as on a cloudy day or at dusk, where the noise level is not sufficient to saturate the SPAD (s) to be dynamically adjusted, with reference to the analysis in FIG. 1 to FIG. 2b, this disclosure proposes to increase the photon detection efficiency of the SPAD (s) by increasing the voltage difference across the two ends of the SPAD (s) , which can improve the detection performance of the SPAD (s) for objects with high reflectivity and objects with low reflectivity. Thus, the overall detection capability of the LiDAR is improved.

A detection capability of a LiDAR for objects with 10%reflectivity under strongest noise is typically defined as the worst detection capability of the LiDAR. However, as mentioned above, for devices such as SPAD (s) , the detection capability of the SPAD (s) for objects with 100%reflectivity can be severely weaker than that for objects with 10%reflectivity under strong noise. To balance the detection capabilities for objects with different reflectivity under the same noise, in some embodiments of this disclosure, the detection capability of the LiDAR for objects with the highest reflectivity (100%) is not lower than that for objects with the lowest reflectivity (10%) .

In some embodiments of this disclosure, the number Nn of photons of environment noise in an environment where the LiDAR is located currently can be deduced based on the total count of noise photons output by the LiDAR within a certain duration, which can determine the environment noise level. In embodiments of this disclosure, step S1 includes: S11, determining the number of noise pulses in an output signal of the SPAD (s) based on the output signal; S12, determining the environment noise level of the SPAD (s) based on the number of noise pulses. There can be multiple different ways for implementing calculation. First, different ways for implementing step S11 are described.

In a first method, although the output of the SPAD (s) within a duration includes both response signals to the noise photons and response signals to valid signal photons, in some embodiments, the number of all output pulses of the SPAD (s) can be taken as the number of noise pulses. The deviation is negligible in the first method.

For example, in one sweep window of a LiDAR, the time window for receiving valid signals is very short. For example, in one sweep window, what is received most of the time is noise. For example, the time window of one sweep of a LiDAR is about 2us, and after waiting for 70ns, the next sweep starts. The pulse width of the valid signals is about 10.4 ns, so when the noise is strong, the output of the SPAD (s) within a sweep time window of 2us can be understood as noises. For example, the total count of photons = the number of responded photons = the number of noises being responded to+ the number of valid signals being responded to ≈ the number of noises being responded to. In such as case, the total count of photons can be approximately taken as the number of noises being responded to, which is the number of noise photons.

In a second method, the ratio or positions of valid signals within a time window can be determined in advance, and the number of noise pulses can be determined based on the output of SPAD (s) except the valid signal within the time window. For example, because multiple sweeps (e.g., 400 times, 1000 times) can be performed by a LiDAR with an planar array of SPADs to complete the detection of the distance and reflectivity of a single point in a point cloud, the positions of valid signals (corresponding to objects) can be roughly determined after approximately 200 sweeps, for example. Accordingly, those responses other than the positions of valid signals can be understood as noise. For example, referring to FIG. 4 as follows, in a time window of 10us, based on the first 200 sweeps that the valid signals are within a range of 0.3-0.7 us. Accordingly, the output of SPADs during the time periods of 0-0.3 us and 0.7-1 us can be considered as noises.

In a third method, the number of noise pulses can be determined based on the output of SPAD (s) beyond the time window. For example, the LiDAR can detect the distance based on a certain frequency, and there can be a certain time slot between every two sweeps. After completing a batch of sweeps for a point in the point cloud but before starting the next batch of sweeps for a next point, there is also a certain time slot. In this disclosure, these time slots can monitor and confirm the current external environment noise level. For example, before the detection of the ranging, a time slot can measure background light by using the SPAD (s) .

In a fourth method, the number of noise pulses can be determined based on the output of SPAD (s) that are not used for ranging currently. For example, referring to FIG. 5, when an planar array LiDAR performs detection, the detectors may not always be in an operating state. Instead, the SPAD (s) can be activated that the corresponding laser (e.g., a series of lasers) transmits lights, or slightly in ahead of or slightly behind the time when the corresponding laser (e.g., a series of lasers) transmits lights. For example, at a certain time, there are some SPAD (s) performing ranging, and there are some SPAD (s) not performing ranging. For example, still referring to FIG. 5, when the laser 2 is activated, a field of view of the laser 2 for external detection can partially overlap with the third, fourth, and fifth columns of SPAD (s) . Accordingly, the third, fourth, and fifth columns of SPAD (s) starts to perform ranging, while other columns such as the seventh column of SPAD (s) are non-ranging SPAD (s) at that time, which can measure background light.

In a fifth method, the number of noise pulses can be determined based on the output of SPAD (s) during non-ranging periods within the time window. For example, the LiDAR with an planar array of SPADs performs multiple sweeps (e.g., 400 times, 1000 times) to complete the detection of the distance and reflectivity of a single point in the point cloud. In some embodiments of this disclosure, several sweeps of these sweeps can be selected to measure the noise. For example, when the number of sweeps configured by the LiDAR system for detecting one point in the point cloud is 500, the data of the previous 450 sweeps can be accumulated and the distance information is obtained by analysis and processing. For other 50 sweeps of these sweeps, the corresponding lasers do not emit light, and the output of SPAD (s) during this sweep period can be read as the number of noise pulses.

It should be noted that when methods 1-5 are used, there is no need to provide additional SPAD (s) or other sensors for monitoring the condition of the environment light, which can be more cost-effective. In some embodiments of this disclosure, the following method can also be used.

In a sixth method, additionally providing SPAD (s) or other sensors for monitoring the condition of the environment light to determine the current external environment condition by reading output of the provided sensor (s) within a certain duration. By using this method, ranging and environment noise measurement can be performed in parallel without interfering with each other, which improves real time performance and thus provides more accurate monitoring results for environment light.

By using any one of the described methods or a combination thereof, the number of noise pulses output by SPAD (s) (which are the noises responded by the SPAD (s) ) can be obtained. In step S12, the environment noise level of SPAD (s) can be determined based on the number of noise pulses output by SPAD (s) .

For step 12, it should be noted that there is a certain mapping relationship between the number of noise pulses output by SPAD (s) and the environment noise level of SPAD (s) , and the mapping relationship can be pre-stored. Accordingly, the environment noise level in an external environment where the SPAD (s) is located can be determined based on the number of noise pulses output by SPAD (s) .

For example, referring to FIG. 6a, FIG. 6a illustrates a curve of mapping relationship between the output total count of noise photons and the number Nn of incident noise photons, at this time, the photon detection efficiency and dead time Deadtime of the SPAD (s) are unchanged. For example, FIG. 6 illustrates the output of the SPAD (s) within a certain duration with change of the external noise Nn when the PDE and Deadtime of the SPAD (s) device are unchanged.

It can be seen that the relationship between the two is non-monotonic. Accordingly, after determining the output number of noise photons indicated on the vertical axis, for example, still referring to FIG. 6a, obtaining a unique value for the number Nn of incident noise photons on the horizontal axis is not possible. For example, still referring to FIG. 6a, when the output number of noise photons is referred to as x, the corresponding horizontal coordinate on the horizontal axis can be Nn1 or Nn2 (e.g., the horizontal coordinates corresponding to the two black dots in FIG. 6a) . FIG. 6b and 6c illustrates the curves showing the variation of the output of the SPAD (s) with the PDE, at this time the external noise level is fixed. It can be seen that when Nn is relatively small. For example, the external environment noise is weak, increasing the PDE can strengthen the output signal, satisfying FIG. 6b. However, when Nn is relatively large, for example, the external environment noise is strong, increasing the PDE can weaken the output signal, satisfying FIG. 6c. This mapping relationship or curve can be obtained and stored in a memory of the LiDAR in advance.

For example, still referring to FIG. 6a, one value of the output total count of photons can correspond to two values of the number of input noise photons. Accordingly, in some embodiments of FIG. 6a, it can be impossible to accurately determine the environment noise level of the SPAD (s) based on the number of noise pulses output by the SPAD (s) .

Accordingly for example, still referring to FIGS. 6a-6c, using a change trend of the total count of photons output by the SPAD (s) before and after PDE adjustment can assist in determining whether the current noise level is Nn1 on the left side or Nn2 on the relatively right side of FIG. 6a.

For example, referring to FIG. 7, FIG. 7 show a manner for performing the step S12 bin some embodiments of this disclosure. For example, the step S12 includes the following steps.

At S121, a first number of noise pulses of the SPAD (s) can be determined under current photon detection efficiency. Regardless of the number of noise photons received, the total count of noise photons at this time is recorded firstly.

At S127, the photon detection efficiency of the SPAD (s) can be reduced by a predetermined amplitude and determining a second number of noise pulses. The PDE is reduced by a fixed step and the new total count of noise photons is collected.

At step S128, whether the SPAD (s) is saturated can be determined based on a change trend of the numbers of noise pulses, and the change trend can be determined by the first number of noise pulses and the second number of noise pulse.

For example, the change trend of the total count of noise photons before and after PDE adjustment is determined.

When the total count of noise photons decreases with the decrease of the PDE, it indicates that the current noise level is weak, which is not enough to saturate the SPAD (s) . When the total count of noise photons increases with the decrease of the PDE, this increase indicates that the current noise level is strong enough to saturate the SPAD (s) . After determining whether the SPAD (s) is saturated or not, in FIG. 6a, the current noise level can be determined to be Nn1 on the left side or Nn2 on the relatively right side of FIG. 6a.

In addition, in some embodiments of this disclosure, after determining whether the current noise level is sufficient to saturate the SPAD (s) or not, the following steps can be performed.

At S129, when the noise level is weak, which is not enough to saturate the SPAD (s) , increasing the PDE by a fixed step to improve the detection capability of the LiDAR under low noise.

When the current noise level is strong enough to saturate the SPAD (s) , decreasing the PDE by the fixed step, so as to make the detection capability of the LiDAR for objects with high reflectivity not lower than that for objects with low reflectivity under the current environment noise level as far as possible.

In addition, in practical applications, the process of S127 and S128 can be repeated until the total count of noise photons (the number of responded photons) before and after PDE adjustment no longer changes. At this time, the maximum count rate MCR is basically reached, or the minimum or maximum value of a SPAD's adjustable range is reached.

It should be noted that the above method can determine whether the current noise level is sufficient to saturate the SPAD (s) by monitoring the output of the SPAD (s) within a duration, the output of the SPAD (s) within the duration can be referred to as a noise characteristic value, representing a change in noise. For example, a certain period of time t1 can be set, such as a unit duration is set, to collect and calculate the detection response signals output by the SPAD (s) devices in one pixel, and calculate the noise characteristic value under this state based on the detection response signals. In addition, this characteristic value can be freely set or defined, as long as the calculation method used this time is in consistent with that used next time. For example, it can be an average value or a maximum value of the noise in an area or range within the Pixel.

In addition, when determining the saturation by the noise, the change in the total count of noise photons has to be greater than a fixed threshold, the noise can be considered as having changed. This is because when the SPAD (s) is just in a saturation state, the value of the output total count of photons after PDE adjustment can change as compared with that before PDE adjustment, but the amplitude of the change is small. Accordingly, a threshold can be set, and the output total count of photons is considered as being changed only when the change of the output total count of photons before and after PDE adjustment is greater than the threshold value.

In the above embodiments, whether the current external environment noise level is sufficient to saturate the SPAD (s) is determined based on the change trend of output of the SPAD (s) before and after PDE adjustment, and when the SPAD (s) is saturated, the PDE is reduced by a certain step; when the SPAD (s) is not saturated, the PDE is increased by the certain step until the difference in the output of the SPAD (s) before and after PDE adjustment is within a predetermined threshold range. For example, the output of the SPAD (s) can remain basically unchanged. This method is simple, convenient and easy to implement, which can improve or ensure the efficiency in dynamically adjusting the SPAD (s) .

Furthermore, to demonstrate the technical effect of this disclosure, for example, referring to FIG. 8, FIG. 8 is a diagram illustrating a detection effect of the SPAD (s) before and after PDE adjustment performed using some embodiments. The SPAD (s) is in a strong noise environment, and the number Nn of noise photons of an echo with 100%reflectivity is 3.89. Before adjustment, PDE is 2%and Deadtime is 40ns. After adjustment, the new PDE is 1.64%. The amplitudes of valid signals output at different reflectivity before and after PDE adjustment are shown in FIG. 8. FIG. 8shows that in this high reflectivity environment, reducing PDE improves the detection capability for objects with high reflectivity while slightly reducing the detection capability for objects with low reflectivity, by doing so the detection capability at 100%reflectivity is roughly the same as that at 10%reflectivity. In addition, it can also be seen that when the environment noise level saturates the SPAD (s) , adjusting the photon detection efficiency of the SPAD (s) can change the reflectivity of the object corresponding to a peak of output signal of the SPAD (s) . For example, still referring to FIG. 8, before adjustment, the reflectivity of the object corresponding to the peak of the output signal of the SPAD (s) is approximately 35%. After adjustment, the reflectivity of the object corresponding to the peak of the output signal of the SPAD (s) is approximately 40%.

In some embodiments of this disclosure, a more accurate adjustment method can be used. The following is a detailed explanation.

Firstly, a mathematical model can be established as follows:
y=N_signal*PDE*Refl*exp (-Deadtime*PDE*Refl*N_noise) Equation 1

where y represents an amplitude of valid signal output by a detector in one or more sweep (e.g., corresponding to the number of valid signal photons being responded to, which is proportional to signal-to-noise ratio SNR) . N_signal (which can be replaced by Ns hereafter for simplicity) represents the number of signal photons incident on the detector. N_noise (e.g., Nn as mentioned above) represents the number of noise photons incident on the detector.

Based on Equation 1, when R=10%, the theoretical amplitude of output signal can be:
y_10%=N_signal*10%*PDE*Refl*exp (-Deadtime*10%*PDE*Refl*N_noise) Equation 2

When R=100%, the theoretical amplitude of output signal can be:
y_100%=N_signal*100%*PDE*Refl*exp (-Deadtime*100%*PDE*Refl*N_noise) Equation 3

where Nn1 represents the number of incident noises in the echoes at 10%and Nn2 indicates the number of incident noises in the echoes at 100%reflectivity. Ns1 represents the number of signal photons in the echoes at 10%and Ns2 indicates the number of signal photons in the echoes at 100%, reflectivity. Nn and Ns are positively correlated with the reflectivity, for example both Nn and Ns are proportional to reflectivity.

Accordingly, when using y_10%＜=y_100%to calculate, the following equation can be obtained:

For example, Equation 4 provides a relationship among the number Nn1 of incident noises in the echo, dead time, and PDE when the reflectivity is 10%. For example, once the number Nn1 of incident noises in the echo at 10%reflectivity can be obtained, a PDE can be calculated based on Equation 4. Below the calculated PDE, the detection capability of LiDAR for objects with high reflectivity is not lower than that for objects with low reflectivity. Combining the simulation results in FIGS. 2a and 2b, it can be seen that in practice, when the environment noise Nn is not sufficient to saturate the SPAD, the detection capability of the LiDAR for objects with high reflectivity can be higher (e.g., always higher) than that for objects with low reflectivity. Accordingly, when the SPAD is not saturated, the PDE calculated using Equation 4 is the maximum PDE (less than or equal to a certain value) corresponding to the current noise level. In addition, under this low noise, as the detection capability of the LiDAR for objects with high and low reflectivity can also be improved with the increase of PDE, this PDE is optimal.

However, for example, still referring to FIG. 6a, the relationship between the responded noise and incident noise Nn is non-monotonic, so the number Nn1 of incident noises cannot be uniquely determined from the responded noises. For example, referring to FIG. 9, the following steps can be performed.

At S121, based on a mapping relationship between the number of input noise photons and the number of noise pulses, the number of noise photons incident on the SPAD (s) can be determined based on the number of noise pulses on the basis of a state where the SPAD (s) is not saturated.

For example, still referring to FIG. 6a, regardless of the total count of noise photons received, the number of input noises is first calculated using FIG. 6a based on the unsaturated process (e.g., referring to the left side of the dashed line in FIG. 6a) , Nn=Nn1.

At S122, a first value of the photon detection efficiency of the SPAD (s) can be determined based on the number of noise photons.

For example, the first value of the photon detection efficiency of the SPAD (s) can be calculated based on the calculated Nn1 and Deadtime of the device based on Equation 4. This first value is the maximum PDE that can satisfy y_10%＜=y_100%.

Next, at step S123, the driving voltage of the SPAD (s) can be adjusted to adjust the photon detection efficiency of the SPAD (s) to the first value (e.g., the new PDE obtained from the calculation) .

At step S124, after adjusting the photon detection efficiency, the number of noise pulses in an output signal of the SPAD (s) can be determined based on the output signal. After adjusting the photon detection efficiency to the first value, the number of noise pulses included in the output signal of the SPAD (s) is determined again using the above method.

At step S125, whether the SPAD (s) is saturated can be determined based on the numbers of noise pulses determined in step S121 and step S124.

In some embodiments, whether the newly calculated PDE is appropriate is determined by determining whether the number of noise photons before and after the PDE adjustment is consistent with an expected change trend.

For example, taking a PDE on horizontal axis corresponding to MCR as a boundary (about 7%in FIGS. 6b and 6c) , when a trend where SPAD (s) has not yet reached saturation as on the left side is satisfied, the response of the adjusted SPAD (s) increases when the newly calculated PDE is greater than the existing PDE; and when a trend where SPAD (s) reaches saturation as on the right side is satisfied, the response of the adjusted SPAD (s) decreases when the newly calculated PDE is greater than the existing PDE. Similarly, when a condition that noise is not enough to saturate SPAD (s) is satisfied, the response of the adjusted SPAD (s) decreases when the newly calculated PDE is less than the existing PDE; and when a condition that noise has saturated SPAD (s) on the right side is satisfied, the response of the adjusted SPAD (s) increases when the newly calculated PDE is less than the existing PDE.

At S126, when the SPAD (s) is already saturated, reducing the photon detection efficiency of the SPAD (s) by a predetermined amplitude; and when the SPAD (s) is not saturated, maintaining the photon detection efficiency of the SPAD (s) as the first value.

In some embodiments, when it is determined that the noise level Nn is high to the extent that the SPAD is already saturated, the PDE is reduced by a certain step. At this time, an effect for example, referring to FIG. 8 can be achieved, such that the LiDAR has a balanced detection capability for objects with high reflectivity and objects with low reflectivity in high noise environment.

When it is determined that the current noise Nn is not large enough to saturate the SPAD, the newly calculated PDE is maintained, which is the optimal PDE under the current environment noise level. This can improve the detection capability of LiDAR for both objects with high reflectivity and objects with low reflectivity.

Furthermore, the steps S121-S126 can be cyclically performed at a predetermined interval or in real-time, so as to correspondingly adjust the PDE of LiDAR based on external environment noises, which can improve the detection performance of LiDAR.

In the above embodiments shown in FIG. 9, assuming that the SPAD (s) is not saturated, the first value of the PDE is calculated and the PDE of the SPAD is adjusted to this first value. Whether the SPAD (s) is saturated is determined based on the change trend of the number of noise pulses obtained after the adjustment. This disclosure is not limited to this, and it is possible to obtain the numbers Nn1 and Nn2 of noise photons incident on the SPAD (s) when the SPAD (s) is in saturated, it is also possible to obtain the numbers Nn1 and Nn2 of noise photons incident on the SPAD (s) when the SPAD (s) is in unsaturated states, for example, referring to FIG. 6a. For example, one first value of the PDE can be calculated based on Nn1 and Nn2, and another first value of the PDE can be calculated based on Nn1 and Nn2, and the PDE is adjusted to the one first value of the PDE and the another first value of the PDE. Whether the SPAD (s) is saturated is determined based on the change trend of the number of noise pulses obtained after adjustment.

For example, referring to FIG. 10, FIG. 10 illustrates a circuit diagram using SPAD (s) , which uses a passively quenched structure, but this disclosure is not limited to the structure of SPAD (s) . An anode of the SPAD (s) is grounded through a resistor R1, and a cathode is connected to a high voltage end HV through a resistor R2. When receiving an incident light signal, the SPAD (s) generates a current which can generate a voltage drop through the resistor R1, and the signal is output from the anode of the SPAD (s) and passes through an amplifier to form a voltage signal. In the detection process of a LiDAR, when taking a detector array which includes single-photon avalanche diodes SPAD (s) as an example, as the SPAD (s) operates in Geiger mode, an avalanche effect of the SPAD (s) can be triggered by a single photon, thus the SPAD (s) can be easily affected by environment light noises. On the other hand, the photon detection efficiency PDE of SPAD (s) for a detection light band commonly used by LiDAR is relatively low, thus the signal intensity obtained from a single detection is relatively weak.

For example, referring to FIG. 11, FIG. 11 illustrates a process of multiple detections, and FIG. 12 illustrates an accumulated result in a histogram. FIG. 13 illustrates a structure of a LiDAR. For example, still referring to FIG. 13, the LiDAR 300 in this disclosure includes: a transmitter unit 301, a receiver unit 302, and a controller (not shown) .

The transmitter unit 301 is configured as an array of lasers, which can be an planar array for example, referring to FIG. 13 or a line array. Multiple lasers are attached to a planar plate, and the type thereof can be a vertical cavity surface emitting laser (VCSEL) . In some embodiments, light transmitted by the array of lasers is emitted vertically to a transmitting end, after being shaped by a transmitting lens group (not shown) , the light is emitted to different directions from the LiDAR 300 to cover a vertical field of view (FOV) of the LiDAR.

The receiver unit 302 includes an planar array of SPADs, and it can also be used as an planar array for example, referring to FIG. 13 or a line array. One or more photodetector is attached to the planar plate. The arrangement of the array 301 of lasers corresponds to the arrangement of the array 302 of detectors to form multiple detection channels. The light signals transmitted by the transmitter unit in the same detection channel, after being reflected by objects and returning to the LiDAR, can be received by the detection unit in that detection channel. The sub-fields of view of all detection channels together form the entire FOV of the LiDAR, so as to form a point cloud with a certain horizontal FOV and vertical FOV. In addition, the LiDAR can also be provided with a controller coupled to the transmitter unit 301 and the receiver unit 302.

To improve the signal-to-noise ratio, for a ranging LiDAR device with an array of SPADs for example, referring to FIG. 13, a time-correlated single-photon counting (TCSPC) method can be typically used for ranging. The basic idea of measuring time information of the photon is to take photons as random events and statistically count the photons after measuring the photons for multiple periods.

For example, still referring to FIG. 11, for any point, during a single detection (for obtaining distance and reflectivity information of a point in the point cloud) , the corresponding detector SPAD (s) can be in the operating mode within a predetermined time window, i.e., in a state where avalanche effect of the detector can be triggered by photons. This time window is called the ″detection time window″ , during which the SPAD (s) can be triggered several times, and it is impossible to distinguish whether the SPAD (s) is triggered by the echo signal reflected by the target or the SPAD (s) is triggered by environment light noise. To improve the long distance ranging performance and signal-to-noise ratio of the LiDAR, for example, still referring to FIG. 11, during a single detection (or measurement) process for any target point within the same FOV, the LiDAR can perform multiple repeated detection sweeps (the number of repeated sweeps can be 400-500 times, or more than or less than 400-500 times) . The controller reads the results of multiple detection sweeps and accumulates the results to obtain a histogram. Further calculations and processing can be performed based on this histogram to measure the distance, so as to obtain the distance and reflectivity information of a point in the LiDAR's point cloud.

For one or more detection sweep, the controller of the LiDAR triggers and controls the array 301 of lasers at the transmitting end, at the transmission time t1, to transmit a light pulse for detection, and records the transmission time t1. The light pulse encounters external objects, the light pulse is reflected by the objects and returns to the LiDAR 300, and is received at time t2 by the photodetectors 302 at the receiving end. When the photodetectors 302 are an array of SPADs, environment light can also trigger the avalanche of the SPAD (s) . Once the SPAD (s) receives photons and generates an avalanche electric signal, the signal is transmitted to a time-to-digital converter TDC, and the TDC outputs a time signal when the SPAD (s) is triggered and outputs a signal indicating the number of the SPAD (s) triggered at the time t2 (this is the case when one pixel includes multiple SPAD (s) ; when one pixel includes one SPAD, there is no such signal indicating the number of triggered SPAD (s) , but there are two states: triggered and not triggered) . The subsequent memory stores a timestamp obtained by subtracting the transmission time t1 from the trigger time t2 of the SPAD (s) (related to the time information represented by the horizontal axes of FIGS. 11 and 12) and the signal of the number of triggered SPAD (s) under this timestamp (hereinafter referred to as cnt) .

In addition, the controller 303 can store the number cnt of triggered SPAD (s) obtained from one or more detection sweep in the corresponding memory location based on the timestamp. When a new number cnt of triggered SPAD (s) arrives at a location corresponding to a certain timestamp, the previously stored value is accumulated to the new number cnt of triggered SPAD (s) and accumulated value is updated to that position. After multiple detection sweeps and accumulations, the data saved in the memory forms a histogram. Combined with FIGS. 11-13, the histogram reflects the sum of the number cnt of triggered SPAD (s) corresponding to different timestamps on a time axis reflected by the statistics of multiple detection sweeps. In this way, the time of flight corresponding to the echo pulses can be obtained by operations such as calculating the centroid using histogram, and the ranging result is obtained.

In some embodiments of this disclosure, the controller can also perform the above-mentioned method, for example, perform the following steps.

At S1, a current environment noise level can be determined. For example, the current environment noise level can be determined based on output of the single-photon detector.

At S2, when the environment noise level saturates the single-photon detector within a predetermined duration, a driving voltage of the single-photon detector can be adjusted to reduce the photon detection efficiency of the single-photon detector.

In some embodiments, at step S3, when the environment noise level does not saturate the single-photon detector, the driving voltage of the single-photon detector can be adjusted to increase the photon detection efficiency of the single-photon detector.

In some embodiments of this disclosure, the operation S1 includes the following steps.

At S11, the number of noise pulses in an output signal of the single-photon detector can be determined based on the output signal.

At S12, the environment noise level can be determined based on the number of noise pulses. In some embodiments, the operation S12 can includes the following steps.

At S121, a first number of noise pulses of the single-photon detector can be determined under a current photon detection efficiency.

At S127, the photon detection efficiency of the single-photon detector can be reduced by a predetermined amplitude, and a second number of noise pulses can be determined.

At S128, whether the single-photon detector is saturated can be determined based on a change trend, and the change trend is determined by the first number of noise pulses and the second number of noise pulse.

In some embodiments of this disclosure, the operation S12 can include the following steps.

At S120, based on a mapping relationship between the number of input noise photons and the number of noise pulses, the number of noise photons incident on the single-photon detector can be determined based on the number of noise pulses on the basis of a state where the single-photon detector is not saturated.

At S122, a first value of the photon detection efficiency of the single-photon detector can be determined based on the number of noise photons.

At S123, the driving voltage of the single-photon detector can be adjusted to adjust the photon detection efficiency of the single-photon detector to the first value.

At S124, the number of noise pulses in the output signal of the single-photon detector can be determined based on the output signal after adjusting the photon detection efficiency.

At S125, whether the single-photon detector is saturated can be determined based on the number of noise pulses determined in steps S11 and the number of noise pulses determined in steps S124.

At S126, when the single-photon detector is saturated, the photon detection efficiency of the single-photon detector can be reduced by a predetermined amplitude. When the single-photon detector is not saturated, the photon detection efficiency of the single-photon detector as the first value can be maintained.

In some embodiments of this disclosure, the operation S122 includes the following steps: the first value of the photon detection efficiency of the SPAD (s) can be determined based on both N_noise1 and Deadtime, where the N_noise1 represents the number of noise photons incident on the SPAD (s) , and the Deadtime represents recovery time of the SPAD (s) .

In some embodiments of this disclosure, the operation S3 includes when the environment noise level does not saturate the SPAD (s) , the driving voltage of the SPAD (s) can be adjusted to increase the photon detection efficiency of the SPAD (s) to the first value.

In some embodiments of this disclosure, the operation S3 includes when the environment noise level saturates the SPAD (s) , the driving voltage of the SPAD (s) can be gradually adjusted to gradually decrease the photon detection efficiency of the SPAD (s) , until the difference between the numbers of noise pulses determined based on the output signal of the SPAD (s) before and after the adjustment is less than a predetermined value.

In some embodiments of this disclosure, the operation S11 includes the following steps. The output pulses (e.g., all output pulses) of the SPAD (s) as the number of noise pulses can be taken. The number of noise pulses can be determined based on the output pulses of the SPAD (s) during a period of time other than the time window corresponding to the object.

In some embodiments of this disclosure, saturation of the SPAD (s) corresponds to a situation where the SPAD (s) reaches the maximum count rate.

In some embodiments of this disclosure, the photon detection efficiency of the SPAD (s) has a maximum value and a minimum value, and the control device can perform the following steps. When the SPAD (s) is initiated, the driving voltage of the SPAD (s) can be adjusted to increase the photon detection efficiency thereof to the maximum value. the number of noise pulses in the output signal of the SPAD (s) can be determined based on the output signal.

In some embodiments of this disclosure, saturation of the single-photon detector corresponds to a situation where the single-photon detector reaches the maximum count rate.

In some embodiments of this disclosure, the photon detection efficiency of the single-photon detector can have a maximum value and a minimum value, and the control device can perform the following steps. When the single-photon detector is initiated, the driving voltage of the single-photon detector can be adjusted to increase the photon detection efficiency thereof to the maximum value.

In addition, for example, referring to FIG. 14, FIG. 14 illustrates how the LiDAR executes the technical solution of this disclosure when operating in actual application scenarios.

When the LiDAR is powered on, the initial PDE can be set to a maximum value that the LiDAR can be set, and the total count O1 of noise photons can be collected in this state.

The PDE can be decreased by a fixed step, and the total count O2 of noise photons can be collected in this state.

Furthermore, the change trend of the total count of noise photons from O1 to O2 under the two PDEs before and after adjustment can be determined to assist in determining whether the current noise is sufficient to saturate the SPAD (s) .

When the noise is in an oversaturated or unsaturated state, using the adjusted PDE can further change the total count of noise photons. When the initial PDE is the maximum value, the current (e.g., adjusted) PDE can be smaller than the initial PDE, and the total count of photons increases with the increase of PDE under the unsaturated state, while the total count of photons decreases with the increase of PDE under the saturated state.

When the noise is already saturated, the PDE can be continuously decreased by a step, and the noise saturation determination in step 2 can be repeated.

When the noise is unsaturated, the PDE can be increased by a fixed amplitude, and the increased PDE can be taken as the calculated PDE value.

Whether the calculated PDE is within the maximum configurable range can be determined.

When the PDE is decreased to the minimum value and the SPAD (s) is still in the saturated state, the PDE cannot be further decreased but instead it remains for the predetermined duration first. The PDE is periodically increased by a fixed value and the saturation state of the noise is determined.

When the PDE is increased to the maximum value and the SPAD (s) is still in an unsaturated state, the PDE cannot be further increased but instead the PDE remains for the predetermined duration. The PDE is periodically decreased by a fixed value and the saturation state of the noise is determined.

After the predetermined duration, the LiDAR can collect the output O3 of SPAD (s) under the new PDE and compare O3 with O2. Alternatively, in the next cycle, the LiDAR can compare O4 with O3, and in one cycle after the next cycle, the LiDAR can compare O5 with O4... Based on the comparison results, the noise level can be determined, and PDE can be adjusted based on the above rules. The detailed process of the repeated iteration is not described in detail herein.

In summary, this disclosure provides a method for feedback-adjusting PDE by measuring the environmental noise, so that the LiDAR can balance the detection capability for objects with high reflectivity and objects with low reflectivity under different noise levels. For example, when the environment noise is strong, the PDE is decreased so as to make the detection capability for objects with high reflectivity greater than or equal to the detection capability for objects with low reflectivity under this high noise as far as possible. While, when the environment noise is weak, the PDE is increased to enhance the detection capabilities for both objects with high reflectivity and objects with low reflectivity.

-It should be understood that one or more module or unit in the embodiments described in this disclosure can include one or more physical components in whole or in part. For example, a module or unit can be implemented as a processor, a controller, a computer, or any form of hardware components. As another example, a module or unit can include one or more hardware components and one or more software components. For example, the module can include a processor (e.g., a digital signal processor, microcontroller, field programmable gate array, a central processor, an application-specific integrated circuit, or the like) and a computer program, when the computer program is run on the processor, the function of the module can be realized. The computer program can be stored in a memory (e.g., a random-access memory, a flash memory, a read-only memory, a programmable read-only memory, a register, a hard disk, a removable hard disk, or a storage medium of any other form) , or a server.

-It should be understood that one or more unit in the embodiments described in this disclosure can include one or more physical components in whole or in part. For example, a unit can be implemented as an emitter, a detector, an optic, a processor, a circuit, or any form of hardware component. As another example, a unit can include one or more hardware components and one or more software components. For example, the transmitter unit can include a light emitting circuit, a vertical-cavity surface-emitting laser ( ″VCSEL″ ) , an edge-emitting laser ( ″EEL″ ) , a distributed feedback laser ( ″DFB″ ) , a fiber laser, or the like. The receiver unit can include a light receiving circuit, a SPAD, an avalanche photodiode detector ( ″APD″ ) , a silicon photomultiplier ( ″SiPM″ ) , or the like.

Finally, it should be noted that the above description is only some embodiments of this disclosure and is not intended to limit this disclosure. Although this disclosure has been described in detail with reference to the above-mentioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the embodiments or replace some technical features with equivalent ones. Any modifications, equivalent replacements, improvements, or the like, made within the spirit and principles of this disclosure should be included within the scope of this disclosure.

Claims

A method for dynamically adjusting a single-photon detector, comprising:

determining a current environment noise level; and

when the environment noise level saturates the single-photon detector within a predetermined duration, adjusting a driving voltage of the single-photon detector to reduce a photon detection efficiency of the single-photon detector.
The method of claim 1, wherein determining the current environment noise level comprises:

determining a number of noise pulses in an output signal of the single-photon detector based on the output signal; and

determining the environment noise level based on the number of noise pulses.
The method of claim 2, wherein determining the environment noise level based on the number of noise pulses comprises:

determining a first number of noise pulses of the single-photon detector under a current photon detection efficiency;

reducing the photon detection efficiency of the single-photon detector by a predetermined amplitude, and determining a second number of noise pulses; and

determining whether the single-photon detector is saturated based on a change trend, wherein the change trend is determined by the first number of noise pulses and the second number of noise pulse.
The method of claim 2, wherein determining the environment noise level based on the number of noise pulses comprises:

based on a mapping relationship between a number of input noise photons and the number of noise pulses, determining the number of noise photons incident on the single-photon detector based on the number of noise pulses on the basis of a state where the single-photon detector is not saturated;

determining a first value of the photon detection efficiency of the single-photon detector based on the number of noise photons;

adjusting the driving voltage of the single-photon detector to adjust the photon detection efficiency of the single-photon detector to the first value;

determining the number of noise pulses in the output signal of the single-photon detector based on the output signal after adjusting the photon detection efficiency;

determining whether the single-photon detector is saturated based on the numbers of noise pulses; and

performing one of:

when the single-photon detector is saturated, reducing the photon detection efficiency of the single-photon detector by a predetermined amplitude; or

when the single-photon detector is not saturated, maintaining the photon detection efficiency of the single-photon detector as the first value.
The method of claim 4, wherein determining the first value of the photon detection efficiency of the single-photon detector based on the number of noise photons comprises:

determining the first value of the photon detection efficiency of the single-photon detector based on N_noise1 and Deadtime, wherein the N_noise1 represents the number of noise photons incident on the single-photon detector, and the Deadtime represents dead time of the single-photon detector.
The method of claim 3 or 4, wherein determining the environment noise level based on the number of noise pulses comprises:

when the environment noise level does not saturate the single-photon detector, adjusting the driving voltage of the single-photon detector to increase the photon detection efficiency of the single-photon detector to the first value.
The method of claim 3 or 4, wherein determining the environment noise level based on the number of noise pulses comprises:

when the environment noise level saturates the single-photon detector, adjusting the photon detection efficiency of the single-photon detector can change a reflectivity of an object corresponding to a peak value of the output signal of the single-photon detector.
The method of any one of claims 2-5, wherein determining the number of noise pulses in the output signal of the single-photon detector based on the output signal comprises:

determining the number of noise pulses based on an output of the single-photon detector within a time window; or

determining in advance a ratio or position of a valid signal within the time window and determining the number of noise pulses based on the output of the single-photon detector except the valid signal within the time window; or

determining the number of noise pulses based on the output of the single-photon detector beyond the time window; or

determining the number of noise pulses based on the output of the single-photon detector not used for ranging currently; or

determining the number of noise pulses based on the output of the single-photon detector during non-ranging periods within the time window.
The method of any one of claims 1-5, wherein saturation of the single-photon detector corresponds to a situation where the single-photon detector reaches the maximum count rate.
The method of any one of claims 1-5, further comprising:

when the environment noise level does not saturate the single-photon detector within the predetermined duration, adjusting the driving voltage of the single-photon detector to increase the photon detection efficiency of the single-photon detector.
A LiDAR, comprising:

a transmitter unit configured to transmit a detection light to detect an object;

a receiver unit, wherein the receiver unit comprises an planar array of single-photon detectors configured to receive an echo of the detection light reflected on the object;

a controller coupled to the transmitter unit and the receiver unit and configured to:

determine a current environment noise level; and

when the environment noise level saturates the single-photon detector within a predetermined duration, adjust a driving voltage of the single-photon detector to reduce a photon detection efficiency of the single-photon detector.
The LiDAR of claim 11, wherein the controller is further configured to:

determine the number of noise pulses in an output signal of the single-photon detector based on the output signal; and

determine the environment noise level based on the number of noise pulses.
The LiDAR of claim 12, wherein the controller is further configured to:

determine a first number of noise pulses of the single-photon detector under a current photon detection efficiency;

reduce the photon detection efficiency of the single-photon detector by a predetermined amplitude, and determine a second number of noise pulses; and

determine whether the single-photon detector is saturated based on a change trend, wherein the change trend is determined by the first number of noise pulses and the second number of noise pulse.
The LiDAR of claim 12, wherein the controller is further configured to:

based on a mapping relationship between the number of input noise photons and the number of noise pulses, determine the number of noise photons incident on the single-photon detector based on the number of noise pulses based on a state where the single-photon detector is not saturated;

determine a first value of the photon detection efficiency of the single-photon detector based on the number of noise photons;

adjust the driving voltage of the single-photon detector to adjust the photon detection efficiency of the single-photon detector to the first value;

determine the number of noise pulses in the output signal of the single-photon detector based on the output signal after adjusting the photon detection efficiency;

determine whether the single-photon detector is saturated based on the numbers of noise pulses;

when the single-photon detector is saturated, reduce the photon detection efficiency of the single-photon detector by a predetermined amplitude; and

when the single-photon detector is not saturated, maintain the photon detection efficiency of the single-photon detector as the first value.
The LiDAR of claim 14, wherein the controller is further configured to:

determine the first value of the photon detection efficiency of the single-photon detector based on both N_noise1 and Deadtime, wherein the N_noise1 represents the number of noise photons incident on the single-photon detector, and the Deadtime represents recovery time of the single-photon detector.
The LiDAR of claim 13 or 14, wherein the controller is further configured to: when the environment noise level does not saturate the single-photon detector, adjust the driving voltage of the single-photon detector to increase the photon detection efficiency of the single-photon detector to the first value.
The LiDAR of claim 13 or 14, wherein the controller is further configured to: when the environment noise level saturates the single-photon detector, adjust the photon detection efficiency of the single-photon detector can change a reflectivity of an object corresponding to a peak value of the output signal of the single-photon detector.
The LiDAR of any one of claims 12-15, wherein the controller is further configured to:

determine the number of noise pulses based on an output of the single-photon detector within a time window; or

determine in advance a ratio of a valid signal within the time window and determining the number of noise pulses based on the output except the valid signal within the time window; or

determine the number of noise pulses based on the output of the single-photon detector beyond the time window; or

determine the number of noise pulses based on the output of the single-photon detector not used for ranging currently; or

determine the number of noise pulses based on the output of the single-photon detector during non-ranging periods within the time window.
The LiDAR of any one of claims 11-15, wherein saturation of the single-photon detector corresponds to a situation where the single-photon detector reaches the maximum count rate.
The LiDAR of any one of claims 11-15, wherein the controller is further configured to:

when the environment noise level does not saturate the single-photon detector within a predetermined duration, adjust the driving voltage of the single-photon detector to increase the photon detection efficiency of the single-photon detector.