An integrated ranging and velocity measurement method based on single-photon lidar

Abstract

This invention discloses an integrated ranging and velocity measurement method based on single-photon lidar, belonging to the field of high-precision laser detection technology. The method includes: obtaining the absolute detection time sequence of the photon response in the echo signal by referring to the absolute time of the zero point at the receiver; calculating the fundamental frequency f' of the echo signal and obtaining the pulse period t of the echo signal. r '=1 / f'; based on t r 'With the pulse period t of the transmitted signal r The Doppler equation is used to obtain an estimate of the target's radial velocity, calculate the relative detection time after motion compensation, and obtain an initial estimate of the initial photon flight time. Substituting these values ​​into the maximum likelihood estimation function, gradient optimization is performed to obtain its local optimum, thus achieving integrated measurement of the target's range and radial velocity. This invention enables the simultaneous acquisition of range and radial velocity in the same physical process.

Description

Technical Field

[0001] This invention belongs to the field of high-precision laser detection technology, specifically relating to an integrated ranging and velocity measurement method based on single-photon lidar. Background Technology

[0002] In the field of modern high-precision detection, the requirements for target detection have gradually transitioned from static measurement with only position localization to dynamic measurement that includes distance and velocity information. The velocity parameters of moving objects can also provide important information for many applications. Traditional pulsed lidar can only provide distance measurement information. To obtain the radial velocity of a target, it usually relies on time differentiation of continuous distance data, i.e., differential velocimetry. This method will produce significant motion ambiguity errors and velocity estimation lags when the target is moving at high speed, the sampling frequency is limited, or the signal-to-noise ratio is low, making it difficult to achieve real-time high-precision dynamic perception of fast-moving targets. Therefore, laser integrated ranging and velocimetry technology has emerged. The value of this technological evolution lies not only in the improvement of performance indicators, but also in the reconstruction of the underlying logic of dynamic perception of non-cooperative targets: from the indirect mode of measuring distance first and then velocity, to integrated intelligent perception that measures both distance and velocity and observes both static and dynamic targets. Among them, coherent measurement techniques, represented by amplitude-modulated continuous wave (AMCW) and frequency-modulated continuous wave (FMCW), are relatively mature integrated measurement methods. However, AMCW technology suffers from phase ambiguity, is susceptible to environmental noise interference, and has poor performance in low light; FMCW technology, on the other hand, has drawbacks such as extremely demanding requirements for light sources, excessive system complexity, and high cost, limiting their applications to relatively narrow, specific fields and hindering large-scale engineering promotion. Therefore, there is an urgent need for a cost-effective solution that combines high-precision integrated ranging and velocity measurement capabilities with miniaturization and controllable costs to meet the growing market demand.

[0003] Single-photon lidar (SPL) is currently the most mature and reliable mainstream photoelectric detection device for medium- and long-range ranging applications, and it is widely used due to its high sensitivity, simple system, and small size. For example... Figure 1 As shown, the architecture of a single-photon lidar ranging system is relatively simple, and it is currently the most mature and reliable mainstream technology for long-range ranging applications. It employs TCSPC (Time-Correlated Single-Photon Counting) technology to detect weak echo signals at the photon level. Once a sufficient number of photons have been counted, the time of flight (ToF) can be derived from a histogram. This allows us to obtain the distance to the target. ( Speed ​​of light in vacuum = However, traditional single-photon lidar (SPL) can only provide distance information and cannot directly obtain the crucial radial velocity. In engineering practice, the mainstream approach is to use multiple ranging differential methods to measure velocity. The traditional SPL processing method is essentially a quasi-static measurement; its ranging and velocity measurement processes differ in granularity and range, which leads to significant systematic errors for high-speed moving targets. Therefore, traditional SPL ranging technology, due to its shortcomings such as motion ambiguity, insufficient noise resistance, and low information utilization, is increasingly limiting in modern remote sensing scenarios. Summary of the Invention

[0004] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0005] An integrated ranging and velocity measurement method based on single-photon lidar includes:

[0006] Step 1: Adjust the laser parameters to make it emit a periodic pulse sequence signal; mark the emission time of the first pulse as the zero point, and obtain the absolute detection time sequence T of the photon response in the echo signal by referring to the absolute time of the zero point in the SPAD array and photon response timer at the receiving end;

[0007] Step 2: Neglecting non-ideal factors, perform a non-uniform sampling Fourier transform on the intensity function λ(t) of the echo signal to calculate the power spectrum; solve for the fundamental frequency f' of the echo signal and obtain the pulse period t of the echo signal. r =1 / f'; Based on the pulse period t of the echo signal r 'With the pulse period t of the transmitted signal r Using the Doppler formula to obtain an estimate of the target's radial velocity v. ;

[0008] Step 3: Estimate the absolute detection time series T of the photon response in the echo signal obtained in Step 1 and the radial velocity v of the target. Calculate the relative detection time after motion compensation. and from Obtaining the initial photon flight time initial estimate ; The sequence number of the received photon, with a value ranging from 1 to... , The total number of photons received; Let be the set of relative detection times after motion compensation; where, the th The timestamp of each received photon is T. i ;

[0009] Step 4: Use the initial photon flight time obtained in Step 3. initial estimate The estimated value of the radial velocity v of the target Substitute into the maximum likelihood estimation function ;right Perform gradient optimization to obtain The local optimal solution is obtained, enabling integrated measurement of the distance z and radial velocity v of the target.

[0010] The present invention has the following beneficial effects:

[0011] This invention modifies the recording method of the photon response timer in a single-photon lidar ranging system and upgrades it to a high-precision clock (resolution at the nanosecond level or higher). It treats the periodic emitted pulse sequence of the single-photon lidar as a quantum statistical amplitude-modulated continuous light (AMCW), measuring velocity by observing the Doppler frequency shift between the emitted pulse repetition frequency and the received echo photon detection sequence. Combined with the original time-of-flight ranging method, it constructs a maximum likelihood joint estimate of distance and radial velocity, achieving simultaneous acquisition of distance and radial velocity information in the same physical process. This fundamentally avoids the accumulated errors and dynamic distortions caused by differential velocimetry in existing technologies, enabling high-precision measurements even under extremely low signal-to-noise ratio conditions. This invention is compatible with traditional single-photon lidar ranging systems in terms of core hardware, allowing for the upgrade of existing mature equipment at a very low cost. It facilitates a leap in single-photon lidar detection technology from a single quasi-static ranging function to integrated ranging and velocimetry with dynamic sensing capabilities. Attached Figure Description

[0012] Figure 1 This is a schematic diagram of the time-of-flight ranging method using single-photon lidar.

[0013] Figure 2 This is a schematic diagram of the integrated ranging and velocity measurement method based on single-photon lidar of the present invention. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0015] This invention proposes an integrated ranging and velocity measurement method based on single-photon lidar. The core idea is to treat a series of periodically emitted pulses at equal intervals as a statistically significant modulation signal. The radial velocity of the target is determined by the minute changes in the pulse period in the echo signal generated after the signal hits the moving target. Two main challenges are presented: first, environmental noise significantly affects the echo signal, resulting in a large amount of noise and missing intrinsic signal pulses (false detections and missed detections), requiring specific algorithms for analysis and judgment; second, maintaining the original distance measurement capability while simultaneously performing ranging and velocity measurement, i.e., using the detection data from the same pulse train to simultaneously analyze the target's relative distance and radial velocity information. Compared to existing single-photon lidar ranging technologies, this invention represents a system upgrade while retaining the main hardware, significantly expanding the detection capability and noise immunity of the single-photon lidar measurement system in low signal-to-noise ratio environments, achieving a leap forward in integrated intelligent perception of dynamic targets. Furthermore, compared to FMCW technology, this invention offers more robust adaptability to various application environments, significantly reduces development costs, and promotes the further expansion of high-precision laser detection technology in engineering applications.

[0016] like Figure 2 As shown, the laser operates at a fixed interval period t r The emission of an ultrashort pulse and the detection of the echo signal reflected from the target will alter the receiving period due to the Doppler effect. Radial velocity can be measured by observing the Doppler frequency shift between the repetition frequency of the emitted ultrashort pulse and the received echo photon detection sequence. The integrated ranging and velocimetry method based on single-photon lidar of this invention includes the following specific steps:

[0017] Step 1: Adjust the laser parameters to make it emit a periodic pulse sequence signal, wherein the pulse interval within each subframe maintains a fixed period t. r The number of repetitions within a pulse train is denoted as n. r The total duration of each pulse sequence is t. a =n r t r The pulse period of the echo signal reflected after the transmitted pulse sequence signal hits the detection target changes to t due to the Doppler effect. r Marking the first pulse emission time t=0 as time zero, the SPAD (Single Photon Avalanche Photodiode) array and photon response timer at the receiver need to adjust the TCSPC technology, changing the timestamp recording method of each received photon from relative time referenced to the most recent emission pulse to absolute time referenced to t=0 (i.e., marking the first pulse emission time as time zero), where t is the recording time. This yields the absolute detection time sequence T={T1, T2, T3, ..., T...} of the photon response in the echo signal. N}, T1, T2, T3, ..., T N Here, n represents the timestamp of each received photon, and N is the total number of received photons. The echo signal includes both signal and noise photons. The number of periods, n... r It needs to reach a certain level (e.g., 10) 4 To ensure a sufficient number of signal photons for power spectrum calculation, the total duration t of a single acquisition... a The pulse period t should not be too large (e.g., 0.01s) to avoid significant state changes in the target due to motion during the acquisition time; therefore, the pulse period t is... r For very small values ​​(e.g., on the order of μs), the system acquisition end needs to be upgraded with a more accurate clock module (e.g., on the order of ns) and a larger and faster data storage module.

[0018] Step 2: Under ideal observation conditions, i.e., non-ideal factors (such as dead time and afterpulse) are negligible, photon detection follows a non-homogeneous Poisson process. A non-uniform sampling Fourier transform is performed on the intensity function λ(t) of the echo signal to calculate the power spectrum, and the fundamental frequency f' of the echo signal is calculated. From this, the pulse period t of the echo signal is obtained. r =1 / f ; The Doppler effect generated by the target motion causes the pulse period t of the echo signal to be 1 / f ; r 'With the pulse period t of the transmitted signal r Satisfying Relationships Based on this, an estimate of the target's radial velocity v is obtained. .in, This is the speed of light in a vacuum. This method cleverly bypasses the limitations of traditional pulse-by-pulse processing, making full use of the different manifestations of signal periodicity and noise randomness in the frequency domain, enabling single-photon lidar technology to achieve accurate speed measurement at extremely low signal-to-noise ratios.

[0019] Step 3: The radial velocity v causes a change in the detection time distribution within each pulse cycle. This is used to obtain an initial estimate of the radial velocity v. Then, the relative detection time after motion compensation is calculated. mod ( The sequence number of the received photon, with a value ranging from 1 to... , The total number of photons received. For the first The timestamp of each received photon; mod is the modulo operation), and a method for static targets is applied from... Obtaining the initial photon flight time initial estimate The specific method for static targets is as follows: based on the formula Estimating the initial photon flight time initial estimate .in, To make the function Parameters for obtaining the maximum value The value, This is the pulse energy response function.

[0020] Step 4: Combine the above initial state parameters with the initial photon flight time obtained in Step 3. initial estimate The estimated value of the radial velocity v of the target Substitute into the maximum likelihood estimation function:

[0021] ;

[0022] ;

[0023] in, For t = T i Pulse energy at any moment The maximum likelihood probability is... Let be the pulse energy response function. The observed echo signal pulse number, with a value ranging from 0 to... The natural number.

[0024] Then, L-BFGS-B (a memory-constrained quasi-Newton boundary constraint algorithm) is used to... Perform gradient optimization to achieve statistical significance. The local optimal solution is found. This allows for the determination of the distance z of the detected target. Integrated measurement of radial velocity v.

[0025] Specifically, in step 1, the method of detecting weak photon-level echo signals using TCSPC technology has been changed. It no longer directly calculates the photon flight time by stacking multiple relative times to form a histogram. Instead, in subsequent steps, the radial velocity v of the target is first calculated, and then the relative detection time is calculated. Motion compensation calculations are performed to avoid motion ambiguity. Simultaneously, recording the absolute detection time series supports subsequent power spectrum calculation of radial velocity v, thus avoiding the lag in radial velocity v estimation caused by differential velocities based on ranging results in traditional single-photon lidar technology. The change in data acquisition structure in step 1 is fundamental to ensuring simultaneous measurement of distance z and radial velocity v in a single physical process. This change requires upgrading to a more accurate clock module and a larger, faster data storage module.

[0026] Specifically, in step 2, since the modulation target in a single-photon lidar is a pulse sequence rather than a sine wave, the Doppler frequency shift occurs not only at the fundamental repetition frequency but also at higher harmonics. Due to the randomness of photon detection time and the influence of background noise, the received echo pulse signal cannot completely correspond to the transmitted pulse. Therefore, in environments with lower signal-to-noise ratios, more harmonic orders need to be included in the power spectrum calculation to ensure the detectability of the complex exponential modulus corresponding to the correct frequency shift. Typically, when the signal-to-noise ratio is as low as 0.01, it is necessary to calculate the superposition of harmonic power spectra of more than 200 orders to ensure the accuracy of the radial velocity v measurement.

[0027] Specifically, in step 4, the maximum likelihood joint estimation needs to balance accuracy and computational efficiency. Therefore, when using L-BFGS-B for gradient optimization, it is necessary to match an appropriate step size and convergence threshold to avoid getting caught in overly complex iterative calculations. At the same time, from an efficiency perspective, gradient optimization only needs to reach a local optimum, and there is no need to deliberately pursue the global optimum.

[0028] It is worth noting that in practical engineering applications, the motion state of the target being measured may undergo sudden changes, such as target attitude flipping or loss of field of view due to occlusion. The measurement algorithm itself cannot directly determine such situations, causing the algorithm to fail to converge due to the lack of normal periodic data or to incorrectly identify noise signals as true values ​​during the solution process. Therefore, additional systems (such as tangential plane tracking systems) are often needed to label the validity of the processed data, or to implement a safety mechanism for contextual comprehensive evaluation of the reasonableness of the solution results, such as setting state evaluations for stable tracking and target loss for the tracking system. At the same time, the detection threshold for echo signal intensity also needs to be adjusted as needed during application. If the detection threshold is too high, it may block most of the real signal echo, resulting in too few effective samples and causing algorithm errors; if the detection threshold is too low, a large amount of environmental noise will be included, causing a sharp increase in computational load, which may slow down the processing time or even completely drown out the real signal and cause errors.

[0029] The above description is merely an embodiment of the present invention and does not limit the scope of the invention. Any equivalent structural or procedural transformations made based on the description and drawings of this invention, or direct or indirect applications in other related system fields, are similarly included within the protection scope of this invention. Contents not described in detail in this specification are prior art known to those skilled in the art.

Claims

1. An integrated ranging and velocity measurement method based on single-photon lidar, characterized in that, include: Step 1: Adjust the laser parameters to make it emit a periodic pulse sequence signal; The time of the first pulse transmission is marked as time zero. The absolute detection time sequence T of the photon response in the echo signal is obtained by referring to the absolute time of time zero in the SPAD array and photon response timer at the receiving end. Step 2: Neglecting non-ideal factors, perform a non-uniform sampling Fourier transform on the intensity function λ(t) of the echo signal to calculate the power spectrum; solve for the fundamental frequency f' of the echo signal and obtain the pulse period t of the echo signal. r '=1 / f'; Based on the pulse period t of the echo signal r 'With the pulse period t of the transmitted signal r Using the Doppler formula to obtain an estimate of the target's radial velocity v. ; Step 3: Estimate the absolute detection time series T of the photon response in the echo signal obtained in Step 1 and the radial velocity v of the target. Calculate the relative detection time after motion compensation. and from Obtaining the initial photon flight time initial estimate ; The sequence number of the received photon, with a value ranging from 1 to... , The total number of photons received; Let be the set of relative detection times after motion compensation; where, the th The timestamp of each received photon is T. i ; Step 4: Use the initial photon flight time obtained in Step 3. initial estimate The estimated value of the radial velocity v of the target Substitute into the maximum likelihood estimation function ;right Perform gradient optimization to obtain The local optimal solution is obtained, enabling integrated measurement of the distance z and radial velocity v of the target.

2. The integrated ranging and velocity measurement method based on single-photon lidar according to claim 1, characterized in that, In step 1, the pulse interval within each subframe remains at a fixed period t. r The number of repetitions within the pulse train is n. r The total duration of each pulse sequence is t. a =n r t r The pulse period of the echo signal reflected after the transmitted pulse sequence signal hits the detection target changes to t due to the Doppler effect. r '.

3. The integrated ranging and velocity measurement method based on single-photon lidar according to claim 2, characterized in that, Step 2 includes: based on the periodicity of the emitted pulse sequence signal photon pulses, the echo signal at frequency f = k / t r A coherent superposition occurs at point ', resulting in a detectable harmonic peak; the pulse period t of the echo signal is calculated. r The radial velocity v of the target is calculated based on the Doppler formula; where k is the harmonic series, and its value is a natural number.

4. The integrated ranging and velocity measurement method based on single-photon lidar according to claim 3, characterized in that, In step 2, non-ideal factors include dead time and afterpulse.

5. The integrated ranging and velocity measurement method based on single-photon lidar according to claim 3, characterized in that, In step 2, the Doppler formula is: ； in, It is the speed of light in a vacuum.

6. The integrated ranging and velocity measurement method based on single-photon lidar according to claim 5, characterized in that, In step 3, the relative detection time after motion compensation is: against ; in, The sequence number of the received photon, with a value from 1 to N. For the first The timestamp of the received photon; mod is the modulo operation.

7. The integrated ranging and velocity measurement method based on single-photon lidar according to claim 6, characterized in that, In step 3, methods for static targets are applied from... Obtaining the initial photon flight time initial estimate .

8. The integrated ranging and velocity measurement method based on single-photon lidar according to claim 7, characterized in that, Methods for static targets include: based on formula Estimating the initial photon flight time initial estimate ;in, To make the function Parameters for obtaining the maximum value The value, This is the pulse energy response function.

9. The integrated ranging and velocity measurement method based on single-photon lidar according to claim 8, characterized in that, In step 4, the maximum likelihood estimation function is: ； ； in, for = Pulse energy at any moment To record time, The maximum likelihood probability is... The pulse number of the observed echo signal.

10. The integrated ranging and velocity measurement method based on single-photon lidar according to claim 4, characterized in that, In step 4, a memory-constrained quasi-Newton boundary constraint algorithm is used to... Perform gradient optimization.

Images