Open Access
Add to BibSonomy
Abstract
We propose and experimentally demonstrate a parallel pulsed chaos light detection and ranging (LiDAR) system with a high peak power, parallelism, and anti-interference. The system generates chaotic microcombs based on a chip-scale Si3N4 microresonator. After passing through an acousto-optic modulator, the continuous-wave chaotic microcomb can be transformed into a pulsed chaotic microcomb, in which each comb line provides pulsed chaos. Thus, a parallel pulsed chaos signal is generated. Using the parallel pulsed chaos as the transmission signal of LiDAR, we successfully realize a 4-m three-dimensional imaging experiment using a microelectromechanical mirror for laser scanning. The experimental results indicate that the parallel pulsed chaos LiDAR can detect twice as many pixels as direct detection continuous wave parallel chaos LiDAR under a transmission power of -6 dBm, a duty cycle of 25%, and a pulse repetition frequency of 100 kHz. By further increasing the transmission power to 10 dBm, we acquire an 11 cm × 10 cm image of a target scene with a resolution of 30 × 50 pixels. Finally, the anti-jamming ability of the system is evaluated, and the results show that the system can withstand interferences of at least 15 dB.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Light detection and ranging (LiDAR) was first demonstrated in the early 1960s and initially applied in meteorology [1]. Owing to the rapid advancement of intelligent technologies, LiDAR has become an essential component in several applications including drones, robots, AR/VR, and autonomous vehicles [2–6]. Owing to their superior detection range, accuracy, spatial resolution, and ability to operate under different weather and lighting conditions, LiDAR systems demonstrate significant potential for range measurement, three-dimensional (3D) imaging, as well as object tracking and recognition [7–10].
To mitigate the risk of interference, researchers have investigated random-modulation continuous-wave (RMCW) LiDAR systems that use randomly modulated lights with specific waveforms [11–13]. In an RMCW LiDAR system, an external intensity modulator modulates a pseudorandom binary sequence (PRBS) via a continuous-wave (CW) laser. By calculating the cross-correlation between the received signal reflected from the target and transmitted replica, the distance to the target can be determined accurately, whereas undesired signals from other sources contribute only to noise. In the conventional RMCW LiDAR system, the duration of the predesigned PRBS determines the unambiguous range. On the other hand, the precision of the range is significantly affected by the width of each bit in the PRBS, which is determined by the modulation speed of the modulator used. However, transmitting cyclic PRBSs for limited durations and bits renders the RMCW LiDAR system vulnerable to jamming, as malicious jammers can easily record the transmitted PRBS and then retransmit it to generate false echoes. By contrast, chaotic LiDAR exploits the aperiodic and unpredictable nature of optical chaos, thus avoiding the disadvantages of jamming and interference inherent in RMCW LiDAR systems [14–20]. In addition, through light injection or feedback, the nonlinear dynamics of semiconductor lasers can easily generate light chaos without requiring expensive high-speed signal generators or external modulators [21–25]. However, external feedback is required to generate chaotic waveforms. Moreover, chaotic LiDAR systems offer only a limited detection throughput because each probe channel features a long cumulative time window, which is unsuitable for real-time high-speed 3D imaging.
Recently, the concept of chaotic microcombs, also known as parallel chaos, based on microresonators was proposed, and a novel type of parallel LiDAR was realized based on parallel chaos [26–29]. Parallel chaos LiDAR combines conventional RMCW techniques and the chaotic nature of modulation-unstable microresonator frequency combs. It offers the advantages of anti-interference, parallelism, and a simple system structure. In 2023, Lukashchuk et al. first demonstrated that the incoherent and chaotic states of light in an optical microresonator can be harnessed to implement unambiguous and interference-immune massively parallel coherent laser ranging using the intrinsic random amplitude and phase modulation of chaotic comb lines. The comb is amplified up to 2 W [27]. In 2023, Lukashchuk et al. used 40 independent channels of a continuously scanned microresonator frequency comb operating in the chaotic regime in addition to optical dispersive elements to perform random modulation LiDAR with two-dimensional passive beam steering. The comb was amplified up to 500 mW [28]. In 2023, Chen et al. proposed a parallel chaotic LiDAR that was interference-free and featured a simplified system architecture, millimeter-level ranging accuracy, and millimeter-per-second-level velocity resolution based on chaotic microcombs [29]. However, for the schemes mentioned above, the power is beyond the class-1 eye-safety regulation (less than 10 mW average power at 1550 nm) [30]. How to improve the performance of chaotic microcomb LiDAR while meeting safety requirements is an interesting research direction
Herein, we propose a parallel pulsed chaos LiDAR system to make the energy of come line more concentrated. First, chaotic microcombs are generated under appropriate conditions based on a Si3N4 microresonator and then transferred into parallel pulsed chaos through an acousto-optic modulator (AOM), which allows emission at a higher peak power without changing the average power. We use 30 independent channels of parallel pulsed chaos to achieve 3D imaging with a microelectromechanical system (MEMS) mirror as a laser scanner and evaluate the performance of the LiDAR system.
2. Experimental setup
Figure 1 shows a schematic illustration of the 3D parallel pulsed chaos LiDAR system. The system primarily comprises a parallel pulsed chaos source and a transmitting and receiving system. In the parallel pulsed chaos source module, a pump light, which is provided by a tunable laser (TL, Santec TSL-710, 1480-1640 nm), passes through a polarization controller (PC) and an erbium-doped fiber amplifier (EDFA, Conquer KG-EDFA220303) sequentially and then injected into a Si3N4 microresonator. The linewidth of the pump laser is <100kHz. A chaotic microcomb can be generated without feedback or control circuitry by appropriately adjusting the polarization, strength, and wavelength of the pump light. The generated chaotic microcombs are transferred to an AOM via a fiber Bragg grating to suppress the pump light. A radio-frequency pulsed signal generated by an arbitrary waveform generator (Tektronix AWG70000, 1.5 KS/s-50 GS/s, 15 GHz bandwidth) is used to control the AOM to generate pulsed chaotic microcombs. By controlling the repetition frequency and duty cycle, the energy in the pulsed chaotic microcombs can be more concentrated. Subsequently, specific comb lines of the pulsed chaotic microcombs are filtered out in parallel by a demultiplexer and parallel pulsed chaos is obtained.
Fig. 1. Schematic illustration of parallel pulsed chaos LiDAR system. TL: tunable laser; PC: polarization controller; EDFA: erbium-doped fiber amplifier; MR: microresonator; FBG: fiber Bragg grating; AOM: acousto-optic modulator; AWG: arbitrary waveform generator; DEMUX: demultiplexer; FC: fiber coupler; DSO: digital storage oscilloscope.
In transmitting and receiving system, the parallel pulsed chaos can be separated into two components using a fiber coupler. As reference signal, 10% of the chaos is sent to a reference detector (Conquer KG-APR-1G-A), whereas the remaining 90% is amplified by the EDFA, and coupled out to the free-space via a collimator. Then it is dispersed horizontally via a 600 lines/mm diffraction grating and finally impinged on a MEMS mirror. By controlling the tilt angle of the MEMS mirror, the beam can be scanned vertically for 3D imaging. We use an avalanche photo diode (APD, Thorlabs APD430C) as the receiving detector to obtain and detect the light backscattered from the target directly, that is, the echo signal. Finally, A digital storage oscilloscope (Agilent DSO9254A 2.5 GHz bandwidth) is used to simultaneously acquire the reference and echo signals at a sampling rate of 20 GS/s.
3. Generation of parallel pulsed chaos
To generate chaotic microcombs, we set the wavelength of the pump light to 1553.1 nm and the power of the pump light to 30 dBm. The Si3N4 microresonator has a size of 400 µm × 400 µm, a Q factor of 2.2 × 106, a free spectral range of 100 GHz and a linewidth of the resonances of 10 MHz. An optical spectrum analyzer (Yokogawa 6317C) and an APD (Conquer KG-APR-1G-A) to analyze the microwave characteristics of the chaotic microcombs. Figure 2(a) shows the optical spectrum of the chaotic microcombs. It is composed of a series of frequency components with uniform intervals and coherent stable phase relationships, and is ultra-flat in the central region. The key difference between chaotic microcombs and normal coherent soliton microcombs is their inherent chaotic nature: all the comb lines exhibit random intensity and frequency modulations. Using one comb line (u = 1, labeled in Fig. 2(a)) as an example, we show its time series and autocorrelation function (ACF) in Figs. 2(b) and 2(c). As shown, the time series is similar to the noise signal, i.e., unpredictable and aperiodic, and the shape of the ACF is similar to the delta function.
Fig. 2. (a) Optical spectrum of the generated chaotic microcomb. The pump, indicated by the arrow, is suppressed by a fiber Bragg grating. (b), (d) Time series and (c), (e) autocorrelation functions from parallel chaos and parallel pulsed chaos, respectively.
For comparison, Figs. 2(d) and 2(e) show the time series and ACF of the same comb line from parallel pulsed chaos detected by the same APD. As shown, the time series comprises periodic pulses and maintains the chaotic characteristics of the pulse. In addition, the ACF of the parallel pulsed chaos is similar to that of the original comb line. This implies that the parallel pulsed chaos preserves the aperiodic features of chaos well, in which only a significant peak is observed at zero lag. Under a delta-function-like ACF, precise detection without range ambiguity can be achieved [31].
In chaotic LiDAR measurements, the range resolution R, which directly determines the ranging precision, is related to the full width at half maximum of the ACF, as expressed in Eq. (1).
where c is the velocity of light and FWHM is the full width at half-maximum of the ACF. This system achieved a resolution of ∼ 13 cm, which corresponded to an FWHM of 0.9 ns.Figure 3(a) shows the optical spectrum details of the microcomb, which were obtained from a Brillouin optical spectrum analyzer (BOSA, Aragon Photonics BOSA lite +, 20-MHz resolution); as shown, the power of each comb line was approximately equal. To highlight the parallelism of this system, Figs. 3(b)–3(d) show the reference signals, echo signals, and cross correlation from five different channels (indicated in Fig. 3(a)) with a duty cycle of 25% and a pulse repetition frequency (PRF) of 100 kHz. For comparison, Figs. 3(e)–3(g) show the comb line without pulse modulation. The span of the correlation window (integration time) was 1.5 µs. As shown, for all the channels presented, the intensity of the echo signal and the peak value of the cross-correlation function improved significantly after pulse modulation. The echo signal was weak when parallel chaos was transmitted in its CW form but conspicuous after pulse modulation. In addition, after pulse modulation, the peak value of the cross-correlation function was close to the limit value of 1, whereas the maximum peak value before modulation was approximately 0.6. Hence, the proposed system enables the better detection of faint signals, such as object edges, and low-SNR sensing.
Fig. 3. (a) Optical spectrum of chaotic microcomb acquired using a BOSA. (b), (e) Reference signals. (c), (f) Echo signals and (d), (g) cross-correlation functions from parallel pulsed chaos and parallel chaos, respectively.
4. Analyses of ranging performance
Chaotic ranging is based on the correlation properties of a random-intensity-modulated signal. By calculating the cross-correlation between the signal and reference, the delay time of the launching signal in the round trip of free space can be obtained from the position of the correlation peak, thus allowing the distance to be determined.
To quantitatively compare the performance of parallel pulsed chaos LiDAR and parallel chaos LiDAR of the same type, we conducted a ranging experiment by placing a U-shaped standardized reflector (58% reflectance) at approximately 4 m away from the LiDAR system as the target. Ideally, this experiment should be performed at different distances, however, owing to the limited experimental conditions, we fixed the distance and varied the transmission power to conduct an equivalent experiment. By transmitting the parallel pulsed chaos (PRF of 100 kHz and duty cycle of 25%) and CW chaos at different power levels, the point clouds of the scenes and the corresponding pixel histograms were obtained, as shown in Fig. 4. The detection experiment was performed sequentially using 12 operational optical channels, and the integration time for each pixel was 1.5 µs. These detection channels were mapped to pixels in the horizontal direction using a diffraction grating. Beam scanning in the vertical direction was performed sequentially using a MEMS mirror.
Fig. 4. Point clouds of U-shaped reflector obtained during scanning (12 × 36 points) using MEMS mirror from (a1), (c1) parallel CW chaos LiDAR and (b1), (d1) parallel pulsed chaos LiDAR at transmission power levels of (a1), (b1) -5 dBm and (c1), (d1) -6 dBm. (a2), (b2), (c2), (d2) Detection histogram corresponding to (a1), (b1), (c1), and (d1), respectively.
Figures 4(a1) and 4(b1) show the point clouds obtained at a transmission power of -5 dBm, which is approximately the minimum operating power of the parallel chaos LiDAR, and the corresponding pixel histograms are shown in Figs. 4(a2) and 4(b2). As shown, the parallel pulsed chaos LiDAR performed better than the parallel chaos LiDAR. Although the two reconstructed image frames appeared similar, the proposed system detected 299 pixels, which was almost twice that detected by the parallel chaos LiDAR. By further reducing the transmission power to -6 dBm, the scene could not be resolved precisely when parallel chaos LiDAR was applied (Fig. 4(c1)). By contrast, the proposed LiDAR still performed well and almost fully reconstructed the scene (Fig. 4(d1)). In addition, the proposed LiDAR detected 244 pixels in total (Fig. 4(d2)), i.e., 100 pixels more than the pixels detected by the parallel chaos LiDAR (Fig. 4(c2)).
In addition, we compared the cross-correlation peak, SNR (defined as the ratio between the peak amplitude of the correlation function and three times the standard deviation of the amplitude changes in the background) and the precision between the parallel pulsed chaos LiDAR and parallel CW chaos LiDAR at different transmission power levels with the same APD. About -30 dBm optical power is needed to overcome the noise of receiving APD and the saturation power is around -9.6 dBm. Figure 5(a) shows the average cross-correlation peak under 50 repeated measurements. If the cross-correlation peak is less than 0.1, the distance cannot be distinguished. It can be seen the working power threshold of parallel CW chaos LiDAR is approximately -5 dBm and when the transmission power is greater than 5 dBm, the receiving APD in parallel pulsed chaos LiDAR gradually approaches saturation. Figures 5(b) and 5(c) show that the parallel pulsed chaos LiDAR presents higher SNRs and precision at all power levels owing to its higher peak power. When the power was less than -1 dBm, the SNRs of the parallel chaos LiDAR were at least 3 dB higher than those of the parallel chaos LiDAR, and the parallel chaos LiDAR could not be operated well (precision exceeding 5 cm).
Fig. 5. (a) Cross-correlation peak, (b) SNR and (c) precision of LiDAR with parallel CW chaos (blue) and parallel pulsed chaos (red) at different power levels. Parallel pulsed chaos has a duty cycle of 25% and a PRF of 100 kHz.
Additional 3D imaging experiments were conducted at a fixed transmission power of 10 dBm. Based on the parallelism of the parallel pulsed chaos, 30 comb lines between 1529 and 1553 nm were used for this imaging experiment. The detection channels were amplified and mapped to the pixels in the horizontal direction through a diffraction grating, and a MEMS mirror was used as the laser scanner in the vertical direction. Three characters (V, U, and I) were placed approximately 4 m in front of the MEMS mirror as targets. The echo signal was obtained using an APD, and the target distance was calculated using the cross-correlation between the reference and echo signals. Figure 6 shows the entire 11 cm × 10 cm 3D image of the reconstructed scene, which featured an image resolution of 30 × 50 pixels. The integration time for each pixel was 1.5 µs. The three objects in the scene were clearly distinguished, thus demonstrating their feasibility in the real world.
Fig. 6. Point cloud of three characters (V, U, and I) obtained during scanning (30 × 50 points) using MEMS mirror.
Finally, the resistance of the system to frequency-modulated continuous waves (FMCWs) and RMCWs was verified via simulation, as shown in Fig. 7. The FMCW and RMCW signals were deliberately mixed with the echo signal. Figures 7(a) and 7(b) show the cross-correlation functions between the reference and mixed signals based on using the FMCW signal as the interference source under ISRs (defined as the ratio of the interference signal power to the echo signal power) of 10 and 15 dB, respectively, with an integration time of 1.5 µs. Figures 7(c) and 7(d) show the cross-correlation functions under the same conditions, except that the RMCW signal was used as the interference signal instead of the FMCW signal. All the results show a correlation peak with detectable ISR exceeding 15 dB.
Fig. 7. Cross-correlation results under interference of (a), (b) FMCW and (c), (d) RMCW, under ISRs of (a), (c) 10 dB and (b), (d) 15 dB. Integration time was 1.5 µs.
5. Conclusion
In this study, we demonstrated a parallel pulsed chaos LiDAR system that significantly improved peak power without changing average power. We performed massively parallel 3D imaging experiments using a MEMS mirror to scan collimated light. Compared with CW chaotic microcomb LiDAR with direct APD detection, the proposed system can operate at a lower power with a higher SNR and precision while maintaining the characteristics of parallelism and anti-jamming. And it is estimated that the detection distance can reach 100 m. These properties render the system invaluable for practical applications in complex real-world environments. The experimental results indicate that the proposed system can achieve 3D imaging of objects 4 m away and detect approximately twice as many pixels as parallel CW chaos LiDAR at -6 dBm power (below the threshold for operating chaotic microcomb LiDAR systems). In addition, the simulation results showed a correlation peak with detectable ISRs exceeding 15 dB, with an integration time of 1.5 µs. This massively parallel laser ranging offers a new method for the development of integrated low-cost unmanned driving technology.
Funding
Science Fund for Distinguished Young Scholars of Chongqing Municipality (cstc2021jcyj-jqX0027); Natural Science Foundation of Chongqing (CSTB2022NSCQ-MSX0313); the Innovation Research 2035 Pilot Plan of Southwest University (SWUXDPY22012); Innovation Support Program for Overseas Students in Chongqing (cx2021008); National Natural Science Foundation of China (61875167, 62335015).
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. G. G. Goyer and R. Watson, “The Laser and Its Application to Meteorology,” Bull. Am. Meteorol. Soc. 44(9), 564–570 (1963). [CrossRef]
2. S. a. Brent, “LiDAR: Mapping the world in 3D,” Nat. Photonics 4(7), 429–430 (2010). [CrossRef]
3. X. Zhao, P. Sun, Z. Xu, et al., “Fusion of 3D LiDAR and Camera Data for Object Detection in Autonomous Vehicle Applications,” IEEE Sensors J. 20(9), 4901–4913 (2020). [CrossRef]
4. A. Mahjoubfar, D. V. Churkin, S. Barland, et al., “Time stretch and its applications,” Nat. Photonics 11(6), 341–351 (2017). [CrossRef]
5. F. Moosmann and C. Stiller, “Velodyne SLAM,” in 2011 IEEE Intelligent Vehicles Symposium (IV) (2011), pp. 393–398.
6. Q. Li, L. Chen, M. Li, et al., “A Sensor-Fusion Drivable-Region and Lane-Detection System for Autonomous Vehicle Navigation in Challenging Road Scenarios,” IEEE Trans. Veh. Technol. 63(2), 540–555 (2014). [CrossRef]
7. I. Kim, R. J. Martins, J. Jang, et al., “Nanophotonics for light detection and ranging technology,” Nat. Nanotechnol. 16(5), 508–524 (2021). [CrossRef]
8. Y. Na, C. G. Jeon, C. Ahn, et al., “Ultrafast, sub-nanometre-precision and multifunctional time-of-flight detection,” Nat. Photonics 14(6), 355–360 (2020). [CrossRef]
9. J. Liu, Q. Sun, Z. Fan, et al., “TOF LiDAR Development in Autonomous Vehicle,” in 2018 IEEE 3rd Optoelectronics Global Conference (OGC) (2018), pp. 185–190.
10. “LiDAR drives forwards,” Nature Photonics 12(8), 441-441 (2018).
11. X. Ai, R. Nock, J. G. Rarity, et al., “High-resolution random-modulation cw LiDAR,” Appl. Opt. 50(22), 4478–4488 (2011). [CrossRef]
12. M. Quatrevalet, X. Ai, A. Pérez-Serrano, et al., “Atmospheric CO2 sensing with a random modulation continuous wave integrated path differential absorption LiDAR,” IEEE J. Select. Topics Quantum Electron. 23(2), 157–167 (2017). [CrossRef]
13. Z. J. Yang, C. Li, Y. U. Ming, et al., “Compact 405-nm random-modulation continuous wave LiDAR for standoff biological warfare detection,” J. Appl. Remote Sens 9(1), 096042 (2015). [CrossRef]
14. F. Y. Lin and J. M. Liu, “Chaotic LiDAR,” IEEE J. Select. Topics Quantum Electron. 10(5), 991–997 (2004). [CrossRef]
15. W. T. Wu, Y. H. Liao, and F. Y. Lin, “Noise suppressions in synchronized chaos LiDARs,” Opt. Express 18(25), 26155–26162 (2010). [CrossRef]
16. C. H. Cheng, Y. C. Chen, and F. Y. Lin, “Generation of Uncorrelated Multichannel Chaos by Electrical Heterodyning for Multiple-Input–Multiple-Output Chaos Radar Application,” IEEE Photonics J. 8(1), 1–14 (2016). [CrossRef]
17. J. D. Chen, H. L. Ho, H. L. Tsay, et al., “3D chaos LiDAR system with a pulsed master oscillator power amplifier scheme,” Opt. Express 29(17), 27871–27881 (2021). [CrossRef]
18. C. H. Cheng, C. Y. Chen, J. D. Chen, et al., “3D pulsed chaos LiDAR system,” Opt. Express 26(9), 12230–12241 (2018). [CrossRef]
19. J. D. Chen, K. W. Wu, H. L. Ho, et al., “3-D Multi-Input Multi-Output (MIMO) Pulsed Chaos LiDAR Based on Time-Division Multiplexing,” IEEE J. Select. Topics Quantum Electron. 28(5), 1–9 (2022). [CrossRef]
20. H. L. Ho, J. D. Chen, C. A. Yang, et al., “High-speed 3D imaging using a chaos LiDAR system,” Eur. Phys. J. Spec. Top. 231(3), 435–441 (2022). [CrossRef]
21. F. Y. Lin, Y. K. Chao, and T. C. Wu, “Effective Bandwidths of Broadband Chaotic Signals,” IEEE J. Quantum Electron. 48(8), 1010–1014 (2012). [CrossRef]
22. J. Mork, B. Tromborg, and J. Mark, “Chaos in semiconductor lasers with optical feedback: theory and experiment,” IEEE J. Quantum Electron. 28(1), 93–108 (1992). [CrossRef]
23. Y. H. Liao and F. Y. Lin, “Dynamical characteristics and their applications of semiconductor lasers subject to both optical injection and optical feedback,” Opt. Express 21(20), 23568–23578 (2013). [CrossRef]
24. J. Ruan and S. C. Chan, “Chaotic dimension enhancement by optical injection into a semiconductor laser under feedback,” Opt. Lett. 47(4), 858–861 (2022). [CrossRef]
25. J. Song, Z. Q. Zhong, L. X. Wei, et al., “Experimental investigations on nonlinear dynamics of a semiconductor laser subject to optical injection and fiber Bragg grating feedback,” Opt. Commun. 354, 213–217 (2015). [CrossRef]
26. B. Shen, H. Shu, W. Xie, et al., “Harnessing microcomb-based parallel chaos for random number generation and optical decision making,” Nat. Commun. 14(1), 4590 (2023). [CrossRef]
27. A. Lukashchuk, J. Riemensberger, A. Tusnin, et al., “Chaotic microcomb-based parallel ranging,” Nat. Photon. 17(9), 814–821 (2023). [CrossRef]
28. A. Lukashchuk, J. Riemensberger, A. Stroganov, et al., “Chaotic microcomb inertia-free parallel ranging,” APL Photonics 8(5), 056102 (2023). [CrossRef]
29. R. Chen, H. Shu, B. Shen, et al., “Breaking the temporal and frequency congestion of LiDAR by parallel chaos,” Nat. Photon. 17(4), 306–314 (2023). [CrossRef]
30. International Electrotechnical Commission (IEC), “Safety of laser products. Part 1: Equipment classification, requirements and user’s guide,” Standard IEC 60825-1:2001 (International Electrotechnical Commission, 2001, pp. 20-40).
31. N. Takeuchi, N. Sugimoto, H. Baba, et al., “Random modulation cw LiDAR,” Appl. Opt. 22(9), 1382–1386 (1983). [CrossRef]
