Silicon light chip (SiLC) Technologies, a silicon photonics startup, has developed a frequency-modulated continuous wave (FMCW) LiDAR on-chip as a four-dimensional (4D) machine vision solution to enable robots, vehicles, and other machines to obtain the data required to classify objects over short or long ranges with high resolutions and accuracy.
Image Credit: SWKStock/Shutterstock.com
Three-dimensional (3D) machine vision solutions are transforming the world through light detection and ranging (LiDAR) technology. However, the use of these solutions is limited to automotive applications due to cost, size, and performance issues.
Issues with Current 3D Machine Vision Technologies
The existing 3D machine vision technologies depend on highly sensitive detectors and 905 nm high-power lasers using a time of flight (TOF) architecture. These technologies have demonstrated their effectiveness in automotive applications, such as during autonomous vehicle trials.
However, expensive assembly processes have limited the cost-effective scaling and resolution of 3D vision technologies. Multiuser interference/crosstalk has also hampered the extensive adoption of these technologies.
Although TOF LiDAR can function effectively at night, the bright daylight photons from the sun during daytime can interfere with the signal, creating noise in the received signal. The normal functioning of a TOF system can also be affected when it is receiving pulses from several other TOF LiDARs operating in the vicinity, leading to multiuser crosstalk.
Coding techniques were introduced by splitting the transmit signal into several pulses to address this issue. However, splitting the available energy into multiple pulses leads to a decline in the range performance as it is proportional to the peak power in TOF systems.
Additionally, eye safety concerns limit both the range and amount of power of the TOF systems. Modern TOF LiDAR systems typically operate between 850 and 905 nm wavelengths, which are extremely close to the visible spectrum. Solar irradiance is significantly higher in the 850 to 905 nm wavelength band, leading to higher daylight interference.
FMCW-based Coherent LiDARs as a Potential Solution
Coherent LiDARs based on the FMCW allow much higher detection accuracy and sensitivity. A coherent detection method emits and relies on a low-power frequency chirp in place of simple pulses.
Thus, the signal-to-noise ratio in an FMCW LiDAR is proportional to the total number of transmitted photons in place of peak laser power. FMCW LiDAR can also transmit at substantially lower average power than pulsed TOF LiDAR owing to its greater sensitivity.
The reflected chirp contains the measurement point distance in the form of a frequency shift. Moreover, the reflected chirp also adds a Doppler frequency shift when the measurement point has a radial velocity.
A coherent LiDAR can resolve the velocity and range of each pixel instantaneously using a down and an up chirp, which cannot be achieved by a TOF LiDAR. Velocity is an extremely important attribute of objects and people surrounding a machine, such as an autonomous vehicle.
The FMCW LiDAR photonic circuitry combines an outgoing coherent laser light portion with the received light, which provides a unique “unlock” key that effectively blocks any LiDAR interference or background radiation.
Additionally, the coherent mixing of the received signal and local oscillator generates an amplified beat frequency through constructive interference before it is detected by a balanced photodetector.
Thus, properly implemented FMCW LiDARs are free from interference. The velocity information with each pixel/4D sensing offers an autonomous system a safer and more transparent perception of its environment.
The integration of all components on a single photonics chip to realize the cost targets can only be achieved using FMCW in the short-wave infrared (SWIR) band, as the on-chip waveguides are low loss and transparent at these wavelengths. Moreover, the photonics integration enables solid-state beam steering using optical phase arrays.
SiLC Technologies is one of the few LiDAR companies that are focusing on FMCW and the only company that has demonstrated a silicon photonics chip. The company used its proprietary process technology and silicon photonics expertise to integrate all photonics components, including lasers, detectors, and a coherent detection engine, into a single chip to achieve the cost targets.
About SiLC Technologies
SiLC Technologies, Inc. is a machine vision innovator based in Monrovia, Southern California that provides coherent vision and chip scale LiDAR solutions to enable smart cameras, robotics, and autonomous vehicles to see like humans.
Founded in 2018 by veterans of the silicon photonics industry with decades of commercial product manufacturing and development experience, SiLC is striving to make machine perception more like human perception through more data and integration, higher resolution, and longer range.
The silicon photonics integrated circuit semiconductor supplier used their proprietary silicon photonic integration platform to develop a fully integrated photonics solution for machine vision.
SiLC is also using its expertise in silicon photonics to advance the deployment of coherent 4D imaging solutions in several applications, including augmented reality, robotics, security and monitoring, industrial machine vision, and autonomous mobility.
SiLC employs a silicon-based semiconductor fabrication process to manufacture its chips and standard and automated integrated circuit-style photonics assembly and manufacturing processes to develop cost-effective, compact, and robust solutions.
SiLC’s Fully Integrated Coherent LiDAR Chip for Machine Vision
SiLC successfully integrated an FMCW imaging system on a cost-effective silicon platform to develop a fully integrated FMCW LiDAR on chip based on a coherent detection approach for the first time. The optical engine utilizes FMCW at 1550 nm wavelength, which significantly increases the range, performance, and safety of the LiDAR.
FMCW technology at 1550 nm can effectively address the regulatory concerns about eye safety and enable volume deployment with limited multiuser crosstalk. However, the approach has not been used extensively until now owing to higher cost and the number of required components.
SiLC’s silicon photonics integration platform provides an effective solution to this issue by integrating all required high-performance components into a single silicon chip through established semiconductor fabrication processes to develop a compact, low-power, and low-cost solution.
SiLC’s technology platform can also meet the other key performance specifications, including ultra-low phase noise, polarization-independent operation, low back reflection, high optical power handling, and very low loss.
Silicon manufacturing facilitates the high-volume scaling of complex technologies and devices affordably. Thus, SiLC’s approach can promote large-scale LiDAR adoption in industrial and automotive robotics.
The initial performance results of the integrated FMCW LiDAR test chip demonstrated a 112 m detection range with only 4.7 mW of optical peak power, which is significantly lower than the hundreds/thousands of Watts of optical peak power in traditional pulsed LiDAR systems. Moreover, SiLC’s LiDAR solutions do not cause any damage to cameras, sensors, or eyes at peak power levels of up to 100 mW.
Future Outlook of the Fully Integrated FMCW LiDAR on Chip
SiLC’s chip integrates every photonics function required to enable a low-cost, low-power coherent 4D vision sensor with an extremely small footprint. The company has already launched Eyeonic Vision Sensor, the first commercially available chip-integrated FMCW LiDAR sensor in the market for both ultra-long-range and ultra-short-range machine-vision applications such as robotics, security, and mobility.
The sensor can provide instantaneous velocity and ultra-long-range information to enable machines and robotics to accurately perceive and classify their environment. Recently, SiLC has extended the detection range of the Eyeonic vision sensor to 1000 meters. The company has also announced a manufacturing partnership with Cloud Light, a company based in Hong Kong, for large-scale production of Eyeonic Vision Sensor.
More from AZoOptics: What Are Semiconductor Lithography Systems Used For?
References and Further Reading
In The News [Online] Available at https://www.silc.com/in-the-news/ (Accessed on 13 March 2023)
Press Releases [Online] Available at https://www.silc.com/press-releases/ (Accessed on 13 March 2023)
Product [Online] Available at https://www.silc.com/product/ (Accessed on 13 March 2023)
Markets [Online] Available at https://www.silc.com/markets/ (Accessed on 13 March 2023)
Home [Online] Available at https://www.silc.com/ (Accessed on 13 March 2023)
About Us [Online] Available at https://www.silc.com/about-us/ (Accessed on 13 March 2023)
Technology [Online] Available at https://www.silc.com/technology/ (Accessed on 13 March 2023)
Jadhav, A. (2022). SiLC Technologies LiDAR vision chip enters production at Cloud Light [Online] Available at https://www.edgeir.com/silc-technologies-lidar-vision-chip-enters-production-at-cloud-light-20220609 (Accessed on 13 March 2023)
Silicon Photonics Startup, SiLC Technologies, Targets Range-Extended, Eye-Safe LiDAR with Highly Integrated 4D Vision Chip [Online] Available at https://www.prnewswire.com/news-releases/silicon-photonics-startup-silc-technologies-targets-range-extended-eye-safe-lidar-with-highly-integrated-4d-vision-chip-300767238.html (Accessed on 13 March 2023)
SiLC Technologies launches new Silicon Photonics LiDAR chip [Online] Available at https://www.geospatialworld.net/news/silc-technologies-launches-new-silicon-photonics-lidar-chip/ (Accessed on 13 March 2023)
SiLC Rolls Out Chip-Integrated FMCW LiDAR Sensor [Online] Available at https://siliconsemiconductor.net/article/113934/SiLC_Rolls_Out_Chip-Integrated_FMCW_LiDAR_Sensor (Accessed on 13 March 2023)
Roos, G. (2022) SiLC extends vision sensor range to 1 km [Online] Available at https://www.electronicproducts.com/silc-extends-vision-sensor-range-to-1-km/ (Accessed on 13 March 2023)
Demaitre, E. (2019) SiLC takes modular, FMCW approach to lidar for self-driving cars, robots [Online] Available at https://www.therobotreport.com/silc-modular-approach-lidar-self-driving-cars/ (Accessed on 13 March 2023)
Muenster, R. J. (2019) FMCW: The future of lidar [Online] Available at https://www.novuslight.com/fmcw-the-future-of-lidar_N9691.html (Accessed on 13 March 2023)
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.