Lidar technology uses pulses of infrared light to calculate distance and create a detailed 3D map of the surrounding environment. This allows self-driving cars to detect objects in their path and respond quickly. However, traditional LIDAR sensors are often large and expensive, and many rely on moving components that wear out over time. These limitations make the system difficult to use in a wide range of settings.
MIT researchers have now developed an approach that could lead to smaller, more durable LIDAR sensors that operate without moving parts. Their advances focus on new silicon photonics chips, a type of semiconductor device that controls light instead of electrical signals.
Existing LIDAR systems built using silicon photonics chips typically have a narrow field of view. As a result, it has a hard time scanning areas at the edges of the scene. Previous attempts to extend this viewing range often introduced extra noise and reduced measurement accuracy.
The MIT team addressed these issues by creating an array of integrated antennas that significantly limits the unwanted crosstalk that occurs when adjacent antennas interfere with each other. This design allows the chip to scan over a wider field of view while producing less noise than other silicon photonics-based methods.
Compact LiDAR system with wider field of view
This advancement could support the development of higher-performance LIDAR sensors for challenging applications such as autonomous vehicle navigation, aerial mapping, and construction site monitoring.
“The capabilities we demonstrate in this study solve fundamental problems in integrated optical phased array technology and enable future lidar sensors that can achieve significantly higher performance than we have been able to demonstrate to date,” said Elena Notaros, a member of the Electronics Institute and senior author of a paper on this innovation, and the Robert J. Sillman Career Development Associate Professor in the MIT School of Electrical Engineering and Computer Science (EECS).
In addition to lead author and EECS graduate student Henry Crawford-Eng, EECS graduate students Andres Garcia Coleto, Benjamin M. Mazur, Daniel M. DeSantis, and Tal Sneh also participated in this study. The results of this research have recently nature communications.
How LiDAR maps its surroundings
Many traditional LIDAR systems use large rotating units to direct pulses of light across the scene. When light hits a nearby object, it reflects back toward the sensor. The returning signals provide the information needed to reconstruct a detailed map of the environment.
Silicon photonics-based LIDARs work differently. Rather than rotating mechanical devices, a system known as an integrated optical phased array (OPA) is used to electronically scan a beam of light in multiple directions.
At the heart of OPA is a group of integrated antennas. Each antenna contains small variations at regular intervals along its length. These features, known as waveforms, scatter light from the input source upward and out of the photonic chip.
By changing the phase of the light sent to each antenna, researchers can control the direction of the outgoing beam. Adjusting these phases changes the angle at which the array emits light, allowing the beam to be manipulated without moving any physical components.
Antenna spacing issue
Placing antennas in close proximity will cause severe interference. Adjacent antennas can couple together and scramble the light they produce. Engineers have traditionally prevented this interference by increasing the distance between antennas, but wider spacing introduces a different set of problems.
If the antennas are too far apart, the array will produce several copies of the same beam at different angles. The primary beam can only travel a limited distance before it becomes difficult to distinguish from these additional copies.
“This limits our field of vision, so self-driving cars can only see what’s in front of them within a certain angular range,” García-Colleto explains.
Unwanted beam copies, called grating lobes, can confuse the sensor and cause false positives. It also consumes energy that could be directed to the main beam.
To overcome this tradeoff, MIT researchers have developed antennas with reduced crosstalk that can be placed close together without strong coupling.
Three antenna shapes reduce interference
In traditional OPA, all antennas have the same structure and use the same waveform pattern. When these matched antennas are placed close to each other, they interact very strongly.
The MIT team instead created a repeating set of three antennas with different shapes. They changed the width of the antenna, the size and placement of the waveform. Because the antennas have different shapes, each antenna also has a different propagation coefficient, which describes how light travels through the structure.
“The propagation coefficients of antennas are very different, so when you bring antennas close together, each antenna basically doesn’t ‘see’ its neighbor. Therefore, there is no coupling with neighboring antennas,” García-Colleto said.
Make different antennas behave the same way
Reducing coupling was only part of the challenge. The antennas needed different propagation coefficients, but still emitted light in the same consistent way.
The team designed the antenna based on three essential requirements.
Each antenna had to emit the same amount of light. All antennas also had to emit their beams at the same angle when receiving the same wavelength. Finally, the emission angle needed to vary evenly across the array as the beam was manipulated.
“The challenge is that you need to change the shape of the antenna to reduce crosstalk, but you also have to design the antenna so that it has the same radiation characteristics at the same time. It is possible to design this, but it is very difficult, because typically antennas designed with different shapes tend to behave differently,” Crawford-Eng says.
The researchers started by developing the basic electromagnetic theory that explains how radiation modes combine. They then used this theoretical framework to guide their antenna design and computer simulations.
Based on these calculations, the team manufactured an OPA containing an antenna with reduced crosstalk. The antennas were placed much closer together than in traditional systems, and the completed device was tested experimentally.
Interference reduced from approximately 100% to 1%
Under our experimental conditions, typical OPA produces approximately 100% binding. MIT’s design lowered that coupling to about 1% while still producing a single clean, accurate beam.
The system precisely steered the beam over a wide field of view without producing grating lobes. This combination of broad scanning, low interference, and strong beam quality solves one of the central obstacles facing integrated LIDAR technology.
The researchers now plan to improve the method so that the system can cover an even wider field of view. They are also investigating other possible approaches to wide-field performance that emerged in developing the underlying theory.
“This research addresses a long-standing challenge in integrated optical phased arrays: simultaneously achieving both a wide field of view, which requires dense antenna spacing, and high beam quality, which requires low crosstalk between adjacent antennas.” “Their innovation is an important step forward for chip-scale solid-state beam steering technology,” said Joyce Poon, professor of electrical and computer engineering and director of the Max Planck Institute for Fine Structures at the University of Toronto. Physics not involved in this work.
Semiconductor Research Corporation, National Science Foundation, MIT MathWorks Fellowship, U.S. Department of the Army, and MIT Rolf G. Locher Endowed Fellowship supported portions of the research.
Part of the work was performed using the MIT.nano facility.
