The Lidar (Light detection and ranging) systems on self-driving vehicles
are big and generally expensive. MIT has a Lidar-on-a-chip solution that will
fit on a dime and cost about $10 to manufacture.
Light detection and ranging, or lidar, is a sensing technology based on laser
light. It’s similar to radar, but can have a higher resolution, since the
wavelength of light is about 100,000 times smaller than radio wavelengths.
For robots, this is very important: Since radar cannot accurately image small
features, a robot equipped with only a radar module would have a hard time
grasping a complex object. At the moment, primary applications of lidar are
autonomous vehicles and robotics, but also include terrain and ocean mapping
and UAVs. Lidar systems are integral to almost all autonomous vehicles and
many other robots that operate autonomously in commercial or industrial
environments.
Lidar systems measure how far away each pixel in a 3D space is from the
emitting device, as well as the direction to that pixel, which allows for the
creation of a full 3D model of the world around the sensor. The basic method
of operation of a lidar system is to transmit a beam of light, and then
measure the returning signal when the light reflects off of an object. The
time that the reflected signal takes to come back to the lidar module
provides a direct measurement of the distance to the object. Additional
information about the object, like its velocity or material composition, can
also be determined by measuring certain properties of the reflected signal,
such as the induced Doppler shift. Finally, by steering this transmitted
light, many different points of an environment can be measured to create a
full 3D model.
Most lidar systems—like the ones commonly seen on autonomous vehicles—use
discrete free-space optical components like lasers, lenses, and external
receivers. In order to have a useful field of view, this laser/receiver
module is mechanically spun around, often while being oscillated up and down.
This mechanical apparatus limits the scan rate of the lidar system while
increasing both size and complexity, leading to concerns about long-term
reliability, especially in harsh environments. Today, commercially available
high-end lidar systems can range from $1,000 to upwards of $70,000, which can
limit their applications where cost must be minimized.
Our work at MIT’s Photonic Microsystems Group is trying to take these
large, expensive, mechanical lidar systems and integrate them on a microchip
that can be mass produced in commercial CMOS foundries.
Our lidar chips are produced on 300-millimeter wafers, making their
potential production cost on the order of $10 each at production volumes of
millions of units per year. These on-chip devices promise to be orders of
magnitude smaller, lighter, and cheaper than lidar systems available on the
market today. They also have the potential to be much more robust because of
the lack of moving parts. The non-mechanical beam steering in this device is
1,000 times faster than what is currently achieved in mechanical lidar
systems, and potentially allows for an even faster image scan rate. This can
be useful for accurately tracking small high-speed objects that are only in
the lidar’s field of view for a short amount of time, which could be
important for obstacle avoidance for high-speed UAVs.
Another problem bites the dust: cost.
Fully driverless vehicles, with no steering wheel, will be on US roads by
2021 in the US, sooner elsewhere. Fully driverless trucks will not only be on
highways, but will be the primary mode of long-haul trucking in a 2022-2024
time frame.
