Open Access
Add to BibSonomy
Abstract
Ranging ambiguity is the major challenge in most LiDAR techniques with amplitude modulation, which limits the performance of range detection due to the tradeoff between the ranging precision and the unambiguous range. Here we propose a novel disambiguation method using a laser with chirped amplitude modulation (sweeping modulation frequency), which can in theory infinitely expand the unambiguous range and completely solve the ranging ambiguation problem. The usage of the earlier proposed Chirped Amplitude-Modulated Phase-Shift (CAMPS) technique enables us to detect the phase-shift of chirped signals with high precision. Incorporating this technique with the proposed disambiguation method, the absolute distance well beyond the conventional unambiguous range can easily be found with merely <1% frequency sweep range. When certain conditions are met, the Non-Mechanical Spectrally Scanned LiDAR (NMSL) system employing the CAMPS method and the Dispersion-Tuned Swept Laser (DTSL) can also realize disambiguation in non-mechanical line-scanning measurement.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
The conceptualization of the Fourth Industry Revolution prophesies a highly automated future, where smart machines can solve most issues in the industry and daily life without human intervention [1]. For most key techniques of the coming industrial revolution, such as autonomous vehicles [2–6], intelligent mobile robots [7–9], and smart factories [10–12], the light detection and ranging (LiDAR) is an important technique that enables the machine to precisely recognize the world [2–4]. It can provide the distance and surface shape information of the detected object by sending and detecting an amplitude- or frequency-modulated laser. Conventional LiDAR techniques include three main methods: Time-of-Flight (ToF), Amplitude-Modulated Continuous-Wave (AMCW, or indirect ToF), and Frequency-Modulated Continuous-Wave (FMCW), but each has its own shortcomings. The ToF technique has a long detection range that can reach kilometer order but the resolution is limited by the speed of photon detectors and electronics to centimeters order [13–16]; the AMCW technique has a better precision that can reach millimeter order due to the use of microwave phase detection technique, but with a moderate detection range usually on meter order [17–19]; the FMCW technique can achieve very high precision of sub-millimeter, but its ranging distance is limited by the coherence length of the wavelength-sweeping narrow-linewidth laser source [20–22]. FMCW LiDAR using a single-frequency laser as the laser source can largely extend the ranging distance to ∼km, but its accuracy is also reduced to the meter level due to the low wavelength sweep efficiency of such lasers [23,24]. Another powerful method proposed in recent years is the dual-comb LiDAR, which can reach µm order resolution [25–27]. However, this method requires multiple frequency combs that are phase-locked to each other, thus not as widely employed as others due to its system complexity.
Due to the difficulty of high-speed detectors with limited size required for the ToF LiDAR to match the telescopes and the stringent requirements for broadband-tunable narrow-linewidth lasers in FMCW LiDAR, AMCW becomes an attractive choice due to its system simplicity and advantage in cost [28]. However, one major problem that the AMCW (and ToF) LiDAR systems suffer from is the ranging ambiguity. Because the light source has a periodically changing amplitude, the AMCW LiDAR system can only acquire the relative distance in a certain range that determined by the modulation frequency of the light, while the absolute distance of the detected objects cannot be correctly recognized. This range is defined as the unambiguous range or ambiguity-free range, which is in a tradeoff with the longitudinal resolution, so simply reducing the modulation frequency of light is not a solution to this dilemma. A common method for disambiguation is employing dual- or multi-frequency modulation on the light, where several different modulation frequencies are employed alternatingly or simultaneously for the expansion of the unambiguous range [29–35]. However, multi-frequency modulation usually either causes difficulties for phase detection [29–33] or requires multiple laser sources [34,35], which makes the system burdensome and costly. Multi-frequency disambiguation method using a single laser and modulator can be achieved by adding the signals at different frequencies together or by periodically switching the modulation frequency. However, for the former approach, the amplitude of each signal is reduced so the SNR is degraded, and for the latter, the sudden change of modulation frequency will cause problems in phase detection due to the generated harmonics [35]. Additionally, the multi-frequency method can only extend the unambiguous range to the least common multiple of the unambiguous range of each modulation frequency, which does not completely solve the ambiguity problem. There are also other methods for disambiguation, one of them is called random-modulation CW (RMCW) LiDAR, where the light is coded with a pseudo-random signal and the absolute range can be found with the cross-correlation of the received signal and reference signal [36,37]. However, the unambiguous range is limited by the finite length of the pseudo-random code, and the resolution is usually on meter order due to the limitation of code rate and modulation speed [38]. Another approach is the chaotic LiDAR, which utilizes the nonlinear dynamics of semiconductor lasers or fiber lasers for disambiguation [38–40], but complex systems are required to make a chaotic laser with desired broad bandwidth [41].
In this research, we propose a novel disambiguation method using a continuously sweeping (chirped) modulation frequency, which can in theory completely solve the ambiguity problem with only one laser source and modulator. As the sweeping modulation frequency also provides a continuously sweeping phase result for a fixed measuring distance, the phase-shift change due to frequency change can be traced beyond 2π. One problem with this method in practice is the difficulty of detecting the instantaneous phase-shift between chirped signals. Conventional methods to measure the phase-shift of amplitude-modulated signals cannot provide accurate results if the modulation period itself is changing. Recently, a novel technique for LiDAR named Chirped Amplitude-Modulated Phase-Shift (CAMPS) is proposed [42,43]. It was originally proposed for the Non-Mechanical Spectrally-Scanned LiDAR (NMSL) system (also known as the CAMPS LiDAR system), which is an overall non-mechanical spectrally scanned LiDAR that uses a Dispersion-Tuned Swept Laser (DTSL) as the laser source. Due to the property of DTSL that the repetition rate changes with the output wavelength, the laser used in the NMSL system also has a chirped repetition rate. The CAMPS technique uses a dynamic beating algorithm that is designed to remove the noise with frequencies within the sweep range, so that the phase-shift of chirped signals can be calculated with high precision. The setup of the NMSL system and the flowchart of the CAMPS technique are shown in Fig. 1.
Fig. 1. (a) Setup of the Non-Mechanical Spectrally Scanned LiDAR (NMSL / CAMPS LiDAR) system and (b) flowchart of the Chirped Amplitude-Modulated Phase-Shift (CAMPS) algorithm (DTSL: dispersion-tuned swept laser, EDFA: Erbium-doped fiber amplifier, BS: beam splitter, APD: avalanche photodiode).
For comparison with the conventional multi-frequency method for LiDAR disambiguation, the proposed method is named the CAMPS method, or sweeping-frequency method. Note that the previously published CAMPS technique is a data processing algorithm to improve the precision of phase detection of chirped signals and thus achieve non-mechanical scanning, where the system still suffers from the ranging ambiguity problem, whilst the CAMPS method proposed in this paper focuses on a general disambiguation technique for any intensity-modulation LiDAR system. The disambiguation using dual-frequency and CAMPS methods will be explained and compared in the following sections, then the performance of CAMPS in single-spot distant measurement will be presented with experimental data. Finally, the CAMPS will be brought to a non-mechanical spectrally scanned LiDAR system for disambiguation of line-scanning measurement, its performance and restrictions will be analyzed and discussed.
2. Theory
2.1 Disambiguation with dual-frequency method
In the dual-frequency modulation, two frequencies are used for distance measurement, so the absolute distance D of a single spot on a stationary detected object can be expressed by a set of simultaneous equations:
Considering the possible error during measurement, its value can be identified as:
However, calculated D in this method would always be smaller than ${L_{\textrm{lcm}}}$, which means it only extends the ambiguous-free range to the least common multiple of the unambiguous range of the two used frequencies, so the ambiguity problem is not solved completely. Adding more frequencies of modulation would also not be useful in finding the absolute distance, because one additional equation also brings one additional unknown factor, which means a unique solution can never be acquired from this equation system.
2.2 Disambiguation with CAMPS method
In the CAMPS method, similar to Eq. (1), the absolute distance can be acquired as [42]:
where t is the time during one frequency sweep, ${f_\textrm{m}}(t )$ is the temporally varying modulation frequency, and c is the speed of light. The modulation frequency ${f_\textrm{m}}(t )$ changes linearly with time in the CAMPS method, which may seem similar to an FMCW LiDAR, however, there is some key differences. The CAMPS method uses RF electronics for modulation and measure the range with the phase-shift of modulation, so the modulation frequency is usually on the order of megahertz or gigahertz, whilst the FMCW LiDAR uses a light source where the optical frequency is changing, which is on the order of hundreds of terahertz, and distance is measured with the frequency beating. The optical frequency is too fast for optical detectors, so the phase information cannot be utilized in FMCW LiDAR. When the modulation frequency sweeps from ${f_1}$ to ${f_2}$, the unambiguous distance range that corresponds to $2\pi $ phase change varies from ${L_1}$ to ${L_2}$, so the fixed distance of the detected object may result in a linearly changing phase that depends on the instantaneous frequency. A schematic diagram of the conventional AMCW methods and the CAMPS method are shown in Fig. 2(a) and 2(b), respectively, and a typical phase-shift result of the CAMPS method on a stationary object during a sweep is shown in Fig. 2(c).Fig. 2. Schematic diagram of the (a) conventional AMCW methods with different modulation frequencies and (b) the CAMPS method. (b) Phase results of a single-spot distant measurement acquired with the conventional AMCW method when the frequency is constantly 726 MHz and 720 MHz, and the CAMPS method when the frequency sweeps from 726 to 720 MHz.
As the modulation frequency is changing, the instantaneous frequency at the beginning of the sweep ${f_1}$ and the instantaneous frequency at the end of the sweep ${f_2}$ may have a different value for N. The advantage of the CAMPS method over the dual-frequency method is that the obtained phase is continuous, so we can recognize how many $2\pi $ phase-shift there is when the frequency changes. This means we can solve the equation system (1) with a unique solution by introducing an additional equation $\Delta N = {N_2} - {N_1}$, where $\Delta N$ is known.
For simplicity, here we define the value of N at the beginning of the sweep as the N of the whole sweep, and unwrap $\varphi (t )$ to make it varies monotonously and not only confined in the region of [-π, π), then N can be expressed as:
where $f = {f_\textrm{m}}(t )$, $\varphi (f )= \varphi ({{f_\textrm{m}}(t )} )$, and D is unknown. Because D is constant, we have andHere we already get the absolute distance D, but it is only a theoretical expression in an ideal situation without noise or measurement error. To find the correct D with minimal error in a practical situation, N need to be calculated first, which can be expressed as:
As N is supposed to be an integer, the final solution for N is
Once again using the data shown in Fig. 2 as an example, the unwrapped phase versus modulation frequency is shown in Fig. 3. Through linear regression, the slope of $\varphi $ can be easily acquired and then be used for the calculation of the order number N. The error of the CAMPS method in disambiguation mainly comes from the error of linear regression. The predicted slopes when the value of N changes from 0 to 200 are also shown in Fig. 3. One can notice the difference between two adjacent ambiguation orders are very small, so any effects that can affect the accuracy of linear regression may cause errors in the calculation of N, such as the nonlinearity in frequency sweeping or intensity-induced phase drift. With the increase of N, the angle between two orders becomes even smaller, which indicates that the accuracy is more vulnerable to the above-mentioned systematic errors at high ambiguation orders.
Fig. 3. (Blue line) The unwrapped frequency-corresponding phase when ${f_1}$=726 MHz, ${f_2}$=720 MHz, and $N$=63, and (Gray lines) the predicted change of phase when N has different values, each having a unique slope (up to 200 are shown).
2.3 Simulation of CAMPS method
Unlike the theoretical analysis where the received signal is a perfect sinusoidal signal, in practical measurement, the received signal may have a low signal-to-noise ratio (SNR) and degrade the accuracy of the disambiguation. In this section, we perform a simulation to study the influence of SNR and frequency sweep range on the result of the CAMPS method. The detected distance is defined as 15 m and the central modulation frequency is set to 500 MHz, so that the order number of the unambiguous range is $N = 50$. For each SNR condition, the test is repeated 100 times with different randomly generated Gaussian noise. If the unrounded N is within the range from 49.5 to 50.5 (equal to 50 after rounding), then the result is regarded correct, otherwise it is considered as incorrect.
The value of N calculated by the CAMPS method when the frequency sweep range is 0.1% of the nominal modulation frequency (500 kHz) and the SNR of the received signal is 10/15/20 dB, respectively, are shown in Fig. 4. (a). The figure shows that when the SNR is 10 dB, almost half of the dots are outside the range of 49.5∼50.5, which means the accuracy of disambiguation is only about 50%. When the SNR is increased to 15 dB, all the points are within the correct range. We define the number of points that lie in the correct range as the disambiguation accuracy. Figure 4. (b) shows the disambiguation accuracy of CAMPS LiDAR at different SNR levels when the frequency sweep range is 0.1% and 1%. It is clear that the accuracy increases with the raise of the SNR of the received signal, and when the frequency sweep range is 1%, an SNR of only a few dB can provide enough information for the CAMPS method to achieve fairly high accuracy. The relationship between accuracy, SNR, and frequency sweep range is shown in Fig. 4. (c), which indicates that with a larger frequency sweep range the tolerance of accuracy to SNR also increases.
Fig. 4. (a) Calculated N when frequency sweep range is 0.1% and SNR is 10/15/20 dB, (b) accuracy at different SNR levels when frequency sweep range is 0.1% and 0.1%, and (c) accuracy at different SNR levels and frequency sweep ranges.
3. Experimental results
3.1 Single-spot distant measurement
The CAMPS method is applicable for disambiguation to any ToF/AMCW LiDAR, here the test is performed in the NMSL system using a DTSL as the laser source. The NMSL system is the first overall non-mechanical spectrally scanned LiDAR system that uses a nonmechanically operating laser source together with a spatially dispersive element. As there are no mechanical components in the whole system, the NMSL system has good stability and can achieve an excellent ranging precision of sub-millimeter level at a high lateral scanning rate of 10-100 kHz. The DTSL is used as the light source in the NMSL system, which is a wavelength-swept mode-locked laser that can generate an optical pulse train with a sweeping wavelength and repetition rate [45–50]. The invention of the CAMPS technique enables the DTSL to be used for distance measurement and non-mechanical spectrally beam scanning. The DTSL is an actively mode-locked laser where a large amount of chromatic dispersion is introduced to the cavity to make the free spectral range (FSR) for each wavelength different. By modulating the loss of the resonator with an amplitude modulator at a certain frequency, only the wavelength whose FSR is identical to the modulation frequency will be mode-locked and output, and its repetition rate is determined by the modulation frequency. The DTSL can generate a pulse with pulse duration of ∼100 ps. Due to the shape in pulse, compared to a sinusoidally modulated light, it has higher order peaks in frequency domain that correspond to integral multiple times of repetition rate. As the CAMPS method includes a step where a bandpass pass filter is used for noise reduction, the higher order peaks in frequency domain will also be eliminated, so the pulse train generated by DTSL can be regarded as a typical CW laser with sinusoidal amplitude modulation.
For single-spot distance measurement, the diffraction grating in the NMSL system for spectral scanning is temporarily removed. Figure 5 shows the calculated N (before rounded, same below) and D with the experimental data when the object under detection is placed 12, 54, and 123 cm from the scanner. As mentioned above, the unambiguous range is about 20 cm when the pulse repetition rate is 726∼720 MHz. The frequency sweep range is only 6 MHz, which is <1% of the center frequency. The detector (APD450C, Thorlabs) has a bandwidth that ranges from 0.3 MHz to 1600 MHz, and the sweeping rate of DTSL in this experiment is set to 10 kHz, which means the measurement time for one sweep is 0.1 ms. More information of the system can be found in Ref. [42]. The results show that the system can easily distinguish the absolute distance of the object well beyond the unambiguous range. 32 sets of data are presented here, and all of them show precise results with little deviation. Sometimes the calculated N is not a perfect integer because of the error in linear regression of $\varphi $, but it would not affect the answer provided the caused error is smaller than 0.5. For clarity, it should be noted that the calculated D also contains the length of the laser cavity and the length from the laser to the scanner, so the measured distance is subtracted by 7.1 m to get only the distance between the scanner and the detected object.
Fig. 5. (a) Schematic diagram of single-spot distance measurement when the detected object is at different distance D. (b) Calculated N and (c) D when the object under detection is 12/54/123 cm from the scanner.
To compare the error tolerance of the CAMPS method with the dual-frequency method, we performed a measurement on a stationary object with both methods under the exact same circumstance and the same length of data. The noise-reduction filter is also applied to the data acquired by the dual-frequency method to make a fair comparison. The distance between the object and the system is ∼6 cm, because an EDFA is used this time, the total length including fiber length and optical length inside the system is ∼1300 cm, which corresponds to the 63rd order for 720 MHz light and 62nd order for 726 MHz light. The unambiguous range for 720 MHz and 726 MHz modulation frequency is 20.83 cm and 20.66 cm, respectively, resulting in an extended unambiguous range of 25 m (= 20.83 cm × 120 = 20.66 cm × 121).
The calculated N and D acquired with both methods are shown in Fig. 6. The average value of N calculated with 720 MHz light, 726 MHz light, and the chirped modulated light is 62.054, 63.054, and 62.704, respectively, and their root mean square (RMS) error is 0.017, 0.017, and 0.051, respectively, as shown in Fig. 6(a). In terms of the absolute distance, the average estimated distance is 62.002, 62.095, and 61.858 mm, and the RMS error is 0.018, 0.022, and 0.016 mm, as shown in Fig. 6(b). Although the CAMPS method shows lower precision and accuracy in the calculation of N making it more vulnerable to noise, its performance in calculating D has little difference from the dual-frequency method. The shortcoming in the calculation of N can be compensated by averaging multiple scans. Due to the slow scanning rate of most mechanical scanners (∼100 Hz), at a high frequency sweep rate (10-100 kHz), hundreds or even thousands of times averaging for a single location is possible, which can substantially suppress the random noise and increase SNR.
Fig. 6. Calculated (a) N and (b) D obtained with the dual-frequency method (red and green) and the CAMPS method (blue).
3.2 Line-scanning measurement
In the original design of the NMSL system, a diffraction grating is employed to steer the light according to its wavelength [42]. Under certain conditions, the CAMPS can also be utilized for disambiguation of line-scanning measurement, which becomes another key advantage of the CAMPS method. Naturally, there would be some restrictions during the implementation, the conditions of the line-scanning measurement will be categorized into some different situations and discussed one by one based on the results we have concluded in single-spot measurement.
With a modulation frequency that sweeps from 726 MHz to 720 MHz, the output wavelength of DTSL sweeps from 1530 nm to 1570 nm, which leads to a scanning angle of ∼4° using a diffraction grating a groove density of 966.2 groove/mm. As the scanning angle of the NMSL system is relatively small, the light with different wavelengths is almost parallel, so when the object under detection is flat and perpendicular to the light beam, as shown in Fig. 7(a), the distance for each wavelength can be regarded almost the same, thus still follow the assumption that D is constant in single-spot measurement. The phase change $\Delta \varphi $ can be expressed as
whereFig. 7. A flat object under line-scanning detection that is (a) perpendicular to the scanning light beam and (b) tilted.
However, for a tilted object shown in Fig. 7(b), the phase for ${f_1}$ and ${f_2}$ are
Because $N = \textrm{round}\left( {\frac{f}{{2\pi }}\frac{{\Delta \varphi }}{{\Delta f}} - \frac{{\varphi (f )}}{{2\pi }}} \right)$, the extra term in $\Delta {\varphi _{\textrm{tilt}}}$ will lead to an incorrect D. To make sure the calculated N is correct, $\Delta {\varphi _{\textrm{tilt}}}$ must meet the condition:
This inequation gives the threshold that the tilting of the object does not affect the calculation of N in line-scanning measurement. When the object is ∼20 cm away from the scanner (12 mm lateral scanning range), with 720-726 MHz modulation frequency, as long as the tilting of the object is smaller than ${\pm} 4.1^\circ $, the calculated N is supposed to be correct. In DTSL, for a certain wavelength output range, there usually is a fixed proportion between $\Delta f$ and f, so lowering the modulation frequency may increase this error-free angle.
Figure 8 shows the result of line-scanning detection when a flat object is placed 6 and 36 cm from the scanner. The object here is no longer a highly reflective mirror, but a metal plate with diffusive surface, so an EDFA is used here for higher optical power. The length of EDFA (∼5 m) makes the N becomes larger in (b) and is compensated in (c). These results prove that the absolute distance of a flat object with a small tilting angle in line-scanning measurement can be found by the CAMPS method as expected.
Fig. 8. (a) Schematic diagram of line-scanning distance measurement when the detected object is at different distance D. (b) Calculated N and (c) D when a flat object is 6/36 cm from the scanner.
In most situations of detection, the detected object will not be flat, but as long as the linear regression of the object surface is perpendicular to the laser beam, the equation drawn above can still predict the correct N. One simple example is when the object is symmetrical. In Fig. 9, a screw hole with 7 mm depth and ∼3 mm width is placed ∼17 cm away from the scanner perpendicularly and scanned by the CAMPS system. Figure 9(a) illustrates the shape of the object surface, and Fig. 9(b) shows the value of N calculated by Eq. (10), which is predicted correctly.
Fig. 9. (a) The measured surface shape of a symmetrical screw hole with 7 mm depth and 2 mm width, and (b) calculated N. The green line in (b) highlights the correct value of N.
On the other hand, when the object under detection is not overall perpendicular, the accuracy of the Eq. (10) is no longer guaranteed, in that case, we need to focus on the local slope rather than the overall slope of the phase $\varphi $. The prediction of N following this idea can be expressed as
Figure 10 shows the scanned result and calculated N of an asymmetrical notch on a metal block. One of the edges of the notch has a height of 9 mm while the other side is 7 mm, would which causes errors in linear regression. As shown in Fig. 10(b), using the overall expression here leads to a wrong N that is far from the theoretical value highlighted by the green line. With the local expression, the estimated N is much closer to the correct answer, so that the absolute distance can be once again acquired even if the detected object is asymmetrical.
Fig. 10. (a) The measured surface shape of an asymmetrical notch on a metal block, and (b) calculated N. The green line in (b) highlights the correct value of N and the blue dashed line shows the average value of the blue dots.
The problem of the local expression is also obvious. Because this method is much easier to be influenced by the noise and distortion in the received signal, the accuracy of a single sweep is rather low. Only 16 out of 32 sets of data have an error of less than 0.5, but the right N can still be found by taking the average of multiple scans.
As a comprehensive method, both equations should be used when calculating the N if the shape of the object is unknown. If the results are close to each other, the object is supposed to be symmetrical so the overall expression provides an accurate answer; if the results are quite different, the local expression is more likely to provide a correct N, and more scans are required to reduce the effect of noise. There are also methods that can increase the reliability of the CAMPS technique in line-scanning measurement, one of the methods is to make the frequency of light sweep in two different regions. As DTSL is a harmonical actively mode-locked laser, modulation frequencies on different harmonic orders may mode-lock the same output wavelength and leads to the same steered location. By combining dual-frequency and the CAMPS method, the requirement of a small tilting angle can be removed at a cost of a lower scanning rate.
4. Conclusion
In this paper, we expand the usage of the CAMPS technique to the ranging disambiguation, making it a more versatile and comprehensive ranging technique. Through mathematic deduction, we demonstrated that the CAMPS method can in theory infinitely expand the unambiguous range of the LiDAR system, substantially improving the ranging resolution and/or distance for LiDAR systems where ambiguity is a main problem and potentially reducing the difficulty and complexity of long-range LiDAR detection. In the single-spot distance measurement, with less than 1% frequency sweep range, this method can easily find the absolute distance without causing noticeable additional errors in ranging accuracy. Although it is relatively more vulnerable to random noise than the dual-frequency method, multiple times of averaging can be easily achieved to remove most of the noise. When employed in the NMSL system, the CAMPS method can also be used for disambiguation in line-scanning measurement. In that case, as long as the tilt of the object is not too large, the system can still find the right order number of the unambiguous range N. When the object under detection is roughly perpendicular to the light beam but there is a large depth variation on its surface, if the shape of the surface is symmetrical or its shape does not cause errors to the slope in linear regression, a correct N can be found by the equation of overall expression derived in the first section, otherwise, the equation of local expression needs to be employed.
Funding
Core Research for Evolutional Science and Technology (JPMJCR1872); Japan Society for the Promotion of Science (18H05238, 22H00209); ACT-X (JPMJAX21KC).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. C. Bai, P. Dallasega, G. Orzes, and J. Sarkis, “Industry 4.0 technologies assessment: A sustainability perspective,” Int. J. Prod. Econ. 229, 107776 (2020). [CrossRef]
2. B. Schwarz, “Mapping the world in 3D,” Nat. Photonics 4(7), 429–430 (2010). [CrossRef]
3. R. Thakur, “Scanning LIDAR in Advanced Driver Assistance Systems and Beyond: Building a road map for next-generation LIDAR technology,” IEEE Consum. Electron. Mag. 5(3), 48–54 (2016). [CrossRef]
4. “LiDAR drives forwards,” Nat. Photonics 12, 441 (2018).
5. J. Hecht, “Lidar for Self-Driving Cars,” Opt. Photonics News 29(1), 26–33 (2018). [CrossRef]
6. Y. Li and J. Ibanez-Guzman, “Lidar for Autonomous Driving: The Principles, Challenges, and Trends for Automotive Lidar and Perception Systems,” IEEE Signal Process. Mag. 37(4), 50–61 (2020). [CrossRef]
7. P. Moghadam, W. S. Wijesoma, and D. J. Feng, “Improving path planning and mapping based on stereo vision and lidar,” in Robotics and Vision 2008 10th International Conference on Control, Automation (IEEE, 2008), pp. 384–389.
8. Y. Cheng and G. Y. Wang, “Mobile robot navigation based on lidar,” in 2018 Chinese Control and Decision Conference, (IEEE, 2018), pp. 1243–1246.
9. S. Gatesichapakorn, J. Takamatsu, and M. Ruchanurucks, “ROS based Autonomous Mobile Robot Navigation using 2D LiDAR and RGB-D Camera,” in First International Symposium on Instrumentation, Control, Artificial Intelligence, and Robotics (IEEE, 2019), pp. 151–154.
10. E. Shellshear, R. Berlin, and J. S. Carlson, “Maximizing Smart Factory Systems by Incrementally Updating Point Clouds,” IEEE Comput. Grap. Appl. 35(2), 62–69 (2015). [CrossRef]
11. R. Y. Zhong, X. Xu, and L. Wang, “IoT-enabled Smart Factory Visibility and Traceability Using Laser-scanners,” Procedia Manuf. 10, 1–14 (2017). [CrossRef]
12. L. D. Evjemo, T. Gjerstad, E. I. Grøtli, and G. Sziebig, “Trends in smart manufacturing: Role of humans and industrial robots in smart factories,” Curr. Robot. Rep. 1(2), 35–41 (2020). [CrossRef]
13. A. McCarthy, X. Ren, A. Della Frera, N. R. Gemmell, N. J. Krichel, C. Scarcella, A. Ruggeri, A. Tosi, and G. S. Buller, “Kilometer-range depth imaging at 1550 nm wavelength using an InGaAs/InP single-photon avalanche diode detector,” Opt. Express 21(19), 22098–22113 (2013). [CrossRef]
14. R. Whyte, L. Streeter, M. J. Cree, and A. A. Dorrington, “Application of lidar techniques to time-of-flight range imaging,” Appl. Opt. 54(33), 9654–9664 (2015). [CrossRef]
15. T. Fersch, R. Weigel, and A. Koelpin, “A CDMA modulation technique for automotive time-of-flight LiDAR systems,” IEEE Sensors J. 17(11), 3507–3516 (2017). [CrossRef]
16. Y. Jiang, S. Karpf, and B. Jalali, “Time-stretch LiDAR as a spectrally scanned time-of-flight ranging camera,” Nat. Photonics 14(1), 14–18 (2020). [CrossRef]
17. E. M. Strzelecki, D. A. Cohen, and L. A. Coldren, “Investigation of tunable single frequency diode lasers for sensor applications,” J. Lightwave Technol. 6(10), 1610–1618 (1988). [CrossRef]
18. A. R. Slotwinski, F. E. Goodwin, and D. L. Simonson, “Utilizing GaalAs Laser Diodes as A Source for Frequency Modulated Continuous Wave (FMCW) Coherent Laser Radars,” Laser Diode Technology and Applications 1043, 245–251 (1989). [CrossRef]
19. M. C. Amann, T. M. Bosch, M. Lescure, R. A. Myllylae, and M. Rioux, “Laser ranging: a critical review of unusual techniques for distance measurement,” Opt. Eng 40(1), 10–19 (2001). [CrossRef]
20. T. Spirig, P. Seitz, O. Vietze, and F. Heitger, “The lock-in CCD-two-dimensional synchronous detection of light,” IEEE J. Quantum Electron. 31(9), 1705–1708 (1995). [CrossRef]
21. M. Akasu, “Distance measuring apparatus,” U.S. Patent US5552893A (3 September 1996).
22. S. Foix, G. Alenya, and C. Torras, “Lock-in Time-of-Flight (ToF) Cameras: A Survey,” IEEE Sensors J. 11(9), 1917–1926 (2011). [CrossRef]
23. C. J. Karlsson, FÅA Olsson, D. Letalick, and M. Harris, “All-fiber multifunction continuous-wave coherent laser radar at 1.55 µm for range, speed, vibration, and wind measurements,” Appl. Opt. 39(21), 3716–3726 (2000). [CrossRef]
24. P. Feneyrou, L. Leviandier, J. Minet, G. Pillet, A. Martin, D. Dolfi, J.-P. Schlotterbeck, P. Rondeau, X. Lacondemine, A. Rieu, and T. Midavaine, “Frequency-modulated multifunction lidar for anemometry, range finding, and velocimetry—2. Experimental results,” Appl. Opt. 56(35), 9676–9685 (2017). [CrossRef]
25. I. Coddington, W. C. Swann, L. Nenadovic, and N. R. Newbury, “Rapid and precise absolute distance measurements at long range,” Nat. Photonics 3(6), 351–356 (2009). [CrossRef]
26. J. Lee, S. Han, K. Lee, E. Bae, S. Kim, S. Lee, S.-W. Kim, and Y.-J. Kim, “Absolute distance measurement by dual-comb interferometry with adjustable synthetic wavelength,” Meas. Sci. Technol. 24(4), 045201 (2013). [CrossRef]
27. H. Zhang, H. Wei, X. Wu, H. Yang, and Y. Li, “Absolute distance measurement by dual-comb nonlinear asynchronous optical sampling,” Opt. Express 22(6), 6597–6604 (2014). [CrossRef]
28. B. Behroozpour, P. A. Sandborn, M. C. Wu, and B. E. Boser, “Lidar system architectures and circuits,” IEEE Commun. Mag. 55(10), 135–142 (2017). [CrossRef]
29. S. Perez, E. Garcia, and H. Lamela, “AMCW laser rangefinder for machine vision using two modulation frequencies for wide measurement range and high resolution,” in Testing, Packaging, Reliability, and Applications of Semiconductor Lasers IV, Proc. SPIE3626, 48–52 (1999).
30. S. Hwang, J. Jang, and K. Park, “Solving 2pi ambiguity problem of a laser scanner based on phase-shift measurement method or long distances measurement,” in 12th International Conference on Control, Automation and Systems (IEEE, 2012), pp. 1250–1252.
31. A. A. Dorrington, M. J. Cree, A. D. Payne, R. M. Conroy, and D. A. Carnegie, “Achieving sub-millimetre precision with a solid-state full-field heterodyning range imaging camera,” Meas. Sci. Technol. 18(9), 2809–2816 (2007). [CrossRef]
32. A. P. P. Jongenelen, D. A. Carnegie, A. D. Payne, and A. A. Dorrington, “Maximizing precision over extended unambiguous range for TOF range imaging systems,” inInstrumentation & Measurement Technology Conference Proceedings (IEEE, 2010) pp. 1575–1580.
33. S. P. Poujouly and B. Journet, “A twofold modulation frequency laser range finder,” J. Opt. A: Pure Appl. Opt. 4(6), S356–S363 (2002). [CrossRef]
34. S. Hwang, J. Jang, and K. Park, “Note: Continuous-wave time-of-flight laser scanner using two laser diodes to avoid 2π ambiguity,” Rev. Sci. Instrum. 84(8), 086110 (2013). [CrossRef]
35. C. Zhang, Z. Zhang, Y. Tian, S. Y. Set, and S. Yamashita, “Comprehensive ranging disambiguation for amplitude-modulated continuous-wave laser scanner with focusing optics,” IEEE Trans. Instrum. Meas. 70, 1–11 (2021). [CrossRef]
36. N. Takeuchi, N. Sugimoto, H. Baba, and K. Sakurai, “Random modulation cw lidar,” Appl. Opt. 22(9), 1382–1386 (1983). [CrossRef]
37. J. T. Spollard, L. E. Roberts, C. S. Sambridge, K. McKenzie, and D. A. Shaddock, “Mitigation of phase noise and Doppler-induced frequency offsets in coherent random amplitude modulated continuous-wave LiDAR,” Opt. Express 29(6), 9060–9083 (2021). [CrossRef]
38. F. Y. Lin and J. M. Liu, “Chaotic lidar,” IEEE J. Select. Topics Quantum Electron. 10(5), 991–997 (2004). [CrossRef]
39. L. K. Rumbaugh, E. M. Bollt, W. D. Jemison, and Y. Li, “A 532 nm chaotic lidar transmitter for high resolution underwater ranging and imaging,” in OCEANS (IEEE, 2013), pp. 1–6.
40. H. L. Ho, J. D. Chen, C. A. Yang, C. C. Liu, C. T. Lee, Y. H. Lai, and F. Y. Lin, “High-speed 3D imaging using a chaos lidar system,” Eur. Phys. J. Spec. Top. 231(3), 435–441 (2022). [CrossRef]
41. M. Zhang and Y. Wang, “Review on chaotic lasers and measurement applications,” J. Lightwave Technol. 39(12), 3711–3723 (2021). [CrossRef]
42. Z. Zhang, C. Zhang, T. Shirahata, S. Yamashita, and S. Y. Set, “Overall Non-Mechanical Spectrally Steered LiDAR Using Chirped Amplitude-Modulated Phase-Shift Method,” J. Lightwave Technol. 40(7), 1902–1913 (2022). [CrossRef]
43. Z. Zhang, C. Zhang, T. Shirahata, S. Yamashita, and S.Y. Set, “Novel Data-Processing Method for High-Resolution Spectrally Steered LiDAR using Dispersion-Tuned Swept Laser,” in CLEO: Applications and Technology (Optica Publishing Group, 2022), pp. JW3B–9.
44. A. P. Jongenelen, D. G. Bailey, A. D. Payne, A. A. Dorrington, and D. A. Carnegie, “Analysis of errors in ToF range imaging with dual-frequency modulation,” IEEE Trans. Instrum. Meas. 60(5), 1861–1868 (2011). [CrossRef]
45. S. Yamashita and M. Asano, “Wide and fast wavelength-tunable mode-locked fiber laser based on dispersion tuning,” Opt. Express 14(20), 9299–9306 (2006). [CrossRef]
46. Y. Takubo and S. Yamashita, “High-speed dispersion-tuned wavelength-swept fiber laser using a reflective SOA and a chirped FBG,” Opt. Express 21(4), 5130–5139 (2013). [CrossRef]
47. H. Nagai and S. Yamashita, “Coherence improvement in dispersion-tuned swept laser by pulse modulation,” Electron. Lett. 50(23), 1729–1731 (2014). [CrossRef]
48. Y. Takubo, T. Shirahata, and S. Yamashita, “Optimization of a dispersion-tuned wavelength-swept fiber laser for optical coherence tomography,” Appl. Opt. 55(27), 7749–7755 (2016). [CrossRef]
49. S. Yamashita, “Dispersion-Tuned Swept Lasers for Optical Coherence Tomography,” IEEE J. Select. Topics Quantum Electron. 24(3), 1–9 (2018). [CrossRef]
50. T. Shirahata, S. Y. Set, and S. Yamashita, “Compensation of Phase-Uncertainty-Induced Impairments in Dispersion-Tuned Swept Laser OCT using Digital Coherent Detection,” J. Lightwave Technol. 38(23), 6492–6498 (2020). [CrossRef]
