U.S. patent application number 16/235640 was filed with the patent office on 2019-05-23 for lidar system.
The applicant listed for this patent is GM GLOBAL TECHNOLOGY OPERATIONS LLC. Invention is credited to Lutfollah Maleki, Scott Singer.
Application Number | 20190154835 16/235640 |
Document ID | / |
Family ID | 66534599 |
Filed Date | 2019-05-23 |
![](/patent/app/20190154835/US20190154835A1-20190523-D00000.png)
![](/patent/app/20190154835/US20190154835A1-20190523-D00001.png)
![](/patent/app/20190154835/US20190154835A1-20190523-D00002.png)
![](/patent/app/20190154835/US20190154835A1-20190523-D00003.png)
![](/patent/app/20190154835/US20190154835A1-20190523-D00004.png)
![](/patent/app/20190154835/US20190154835A1-20190523-D00005.png)
![](/patent/app/20190154835/US20190154835A1-20190523-D00006.png)
![](/patent/app/20190154835/US20190154835A1-20190523-D00007.png)
![](/patent/app/20190154835/US20190154835A1-20190523-D00008.png)
![](/patent/app/20190154835/US20190154835A1-20190523-D00009.png)
![](/patent/app/20190154835/US20190154835A1-20190523-D00010.png)
View All Diagrams
United States Patent
Application |
20190154835 |
Kind Code |
A1 |
Maleki; Lutfollah ; et
al. |
May 23, 2019 |
LIDAR SYSTEM
Abstract
A light detection and ranging ("LiDAR") system includes a
coherent light source that generates a frequency modulated optical
signal comprising a series of optical chirps. A scanning assembly
transmits the series of optical chirps in a scan pattern across a
scanning region, and receives a plurality of reflected optical
chirps corresponding to the transmitted optical chirps that have
reflected off one or more objects located within the scanning
region. A photodetector mixes the reflected optical chirps with a
local oscillation (LO) reference signal comprising a series of LO
reference chirps. An electronic data analysis assembly processes
digital data derived from the reflected optical chirps and the LO
reference chirps mixed at the photodetector to generate distance
data and optionally velocity data associated with each of the
reflected optical chirps.
Inventors: |
Maleki; Lutfollah;
(Pasadena, CA) ; Singer; Scott; (San Gabriel,
CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
GM GLOBAL TECHNOLOGY OPERATIONS LLC |
Detroit |
MI |
US |
|
|
Family ID: |
66534599 |
Appl. No.: |
16/235640 |
Filed: |
December 28, 2018 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
16313247 |
|
|
|
|
PCT/US2016/055778 |
Oct 6, 2016 |
|
|
|
16235640 |
|
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G02B 26/12 20130101;
G01S 17/58 20130101; G02B 27/14 20130101; G01S 7/4818 20130101;
G01S 17/34 20200101; G01S 17/89 20130101; G01S 7/4817 20130101;
G02B 26/0833 20130101; G01S 7/4917 20130101; G01S 17/931
20200101 |
International
Class: |
G01S 17/89 20060101
G01S017/89; G02B 26/12 20060101 G02B026/12; G02B 27/14 20060101
G02B027/14; G01S 7/481 20060101 G01S007/481 |
Claims
1. A light detection and ranging (LiDAR) system, comprising: a
coherent light source that generates a frequency modulated optical
signal comprising a series of optical chirps, wherein the coherent
light source has a linewidth of about 100 kHz or less; a scanning
assembly that (a) transmits the series of optical chirps in a scan
pattern across a scanning region and (b) receives a plurality of
reflected optical chirps corresponding to the transmitted optical
chirps that have reflected off one or more objects located within
the scanning region; a photodetector that mixes the reflected
optical chirps with a local oscillation (LO) reference signal
comprising a series of LO reference chirps; and an electronic data
analysis assembly that processes digital data derived from the
reflected optical chirps and the LO reference chirps mixed at the
photodetector to generate distance data associated with each of the
reflected optical chirps; wherein the coherent light source is
configured such that a probability of interference between the
transmitted optical chirps and other optical chirps transmitted by
a second LiDAR system having a configuration of the LiDAR system is
less than 1.times.10.sup.-7.
2. The LiDAR system of claim 1, wherein: the coherent light source
comprises a source laser coupled to a resonator, wherein the
resonator provides optical feedback to the source laser for
linewidth reduction; at least one optical property of the resonator
is modulated to enable generation of the optical chirps; and the
resonator comprises a whispering gallery mode (WGM) resonator.
3. The LiDAR system of claim 1, wherein each of the optical chirps
has a chirp linearity of 80% or more.
4. The LiDAR system of claim 1, wherein the optical chirps are
transmitted at an optical power in a range of 10 mW to 3 W.
5. A light detection and ranging (LiDAR) system, comprising: a
coherent light source that generates a frequency modulated optical
signal comprising a series of optical chirps, wherein the coherent
light source has a linewidth of about 100 kHz or less; a scanning
assembly that (a) transmits the series of optical chirps in a scan
pattern across a scanning region and (b) receives a plurality of
reflected optical chirps corresponding to the transmitted optical
chirps that have reflected off one or more objects located within
the scanning region; a photodetector that mixes the reflected
optical chirps with a local oscillation (LO) reference signal
comprising a series of LO reference chirps; and an electronic data
analysis assembly that processes digital data derived from the
reflected optical chirps and the LO reference chirps mixed at the
photodetector to generate distance data associated with each of the
reflected optical chirps.
6. The LiDAR system of claim 5, wherein the coherent light source
has a linewidth of 10 kHz or less.
7. The LiDAR system of claim 5, wherein the coherent light source
has a linewidth of 1 kHz or less.
8. The LiDAR system of claim 5, wherein the coherent light source
has a linewidth of 100 Hz or less.
9. The LiDAR system of claim 5, wherein the coherent light source
has a linewidth of 1 Hz or less.
10. The LiDAR system of claim 5, wherein the coherent light source
is configured such that a probability of interference between the
transmitted optical chirps and other optical chirps transmitted by
a second LiDAR system having a configuration of the LiDAR system is
less than 1.times.10.sup.-7.
11. The LiDAR system of claim 5, wherein the coherent light source
comprises a source laser coupled to a resonator, wherein the
resonator provides optical feedback to the source laser for
linewidth reduction.
12. The LiDAR system of claim 5, wherein each of the optical chirps
has a chirp linearity of 80% or more.
13. The LiDAR system of claim 12, wherein each of the optical
chirps has a chirp bandwidth in a range of 1 GHz to 100 GHz.
14. The LiDAR system of claim 13, wherein the optical chirps are
generated at a chirp rate in a range of 1.times.10.sup.14 Hz/s to
5.times.10.sup.16 Hz/s.
15. The LiDAR system of claim 5, wherein the optical chirps are
transmitted at an optical power in a range of 10 mW to 3 W.
16. The LiDAR system of claim 5, wherein the electronic data
analysis assembly processes the digital data derived from the
reflected optical chirps and the LO reference chirps mixed at the
photodetector to also generate velocity data associated with each
of the reflected optical chirps.
17. The LiDAR system of claim 5, wherein the photodetector also
detects incoherent light originating from one or more external
light sources, wherein the incoherent light is filtered so as not
to cause interference with the reflected optical chirps.
18. A light detection and ranging (LiDAR) sensor, comprising: a
single coherent light source that generates a frequency modulated
optical signal comprising a series of optical chirps, wherein the
coherent light source has a linewidth of 100 kHz or less; a single
scanner that (a) transmits the series of optical chirps in a scan
pattern across a scanning region at an optical power in a range of
10 mW to 3 W and (b) receives a plurality of reflected optical
chirps corresponding to the transmitted optical chirps that have
reflected off one or more objects located within the scanning
region; a photodetector that mixes the reflected optical chirps
with a local oscillation (LO) reference signal comprising a series
of LO reference chirps, wherein the coherent light source generates
the LO reference chirps; and an electronic data analysis assembly
that processes digital data derived from the reflected optical
chirps and the LO reference chirps mixed at the photodetector to
generate distance data associated with each of the reflected
optical chirps.
19. The LiDAR sensor of claim 18, wherein each of the optical
chirps has a chirp linearity of 80% or more.
20. The LiDAR sensor of claim 18, wherein the single coherent light
source, the single scanner, the photodetector, and the electronic
data analysis assembly are provided in a volume between 10 cm.sup.3
and 500 cm.sup.3.
Description
RELATED APPLICATIONS
[0001] This application is a continuation of U.S. patent
application Ser. No. 16/313,247, filed on Dec. 26, 2018, and
entitled "LIDAR SYSTEM", which is a U.S. National Stage Application
under 35 U.S.C. .sctn. 371 of PCT Application No.
PCT/US2016/055778, filed on Oct. 6, 2016, and entitled "LIDAR
SYSTEM". The entireties of these applications are incorporated
herein by reference.
BACKGROUND
[0002] Light detection and ranging ("LiDAR") systems have been
developed for use in autonomous or semi-autonomous vehicles,
drones, and other applications that require laser scanning
capabilities. One type of LiDAR system is a time of flight ("ToF")
LiDAR system in which a transmitter emits short pulses of light
that reflect off a target object and a receiver receives the
reflected pulses. The delay between emission and reception provides
a measure of the distance between the transmitter and the target
object. ToF LiDAR systems may also determine the velocity of the
target object by comparing two frames of data (i.e., the object
must be identified at least twice) and calculating the shift in
position of the object. A number of different companies have
attempted to develop and/or market ToF LiDAR systems for automotive
applications, including Velodyne LiDAR, Ibeo Automotive Systems
GmbH, and Quanergy Systems, Inc.
[0003] Alternatives to ToF LiDAR systems have also been developed.
For example, a frequency modulated continuous wave (FMCW) LiDAR
system includes a laser source to generate a frequency modulated
(FM) optical signal that comprises a continuous series of optical
"chirps" in which the instantaneous frequency within each chirp
varies with time. A transmitter is used to transmit the optical
chirps across an environment, and a receiver is used to receive the
optical chirps that have reflected off objects located in the
environment. Measurement of the frequency shift and/or phase shift
of each reflected optical chirp relative to a reference optical
chirp provides a measure of the distance and/or velocity of the
reflective objects relative to the system. An example of an FMCW
LiDAR system is described in U.S. Pat. No. 9,310,471 assigned to
HRL Laboratories, LLC. The configuration and performance
characteristics of this LiDAR system (including a wide laser
linewidth of 400 kHz, poor chirp linearity, and limited range,
range resolution and pixel rate) are not ideal for certain
implementations. Other types of FMCW LiDAR systems have also been
developed that use relatively large laser sources and, as a result,
the systems are large, complex, and expensive. As such, FMCW LiDAR
systems have seen limited implementation relative to ToF LiDAR
systems.
[0004] While conventional ToF and FMCW LiDAR systems have been used
in various implementations, there is a need for a LiDAR system that
provides performance characteristics and other advantages that are
not achievable with these conventional systems. In particular,
there is a need for a LiDAR system with low optical power that can
be used to detect objects located at a great distance and/or
traveling at a high speed relative to the LiDAR system and that can
be packaged as a solid state sensor for use in automotive and other
applications.
SUMMARY
[0005] The present invention is directed to a light detection and
ranging ("LiDAR") system that provides performance characteristics
and other advantages that are not achievable with conventional ToF
and FMCW LiDAR systems.
[0006] In a first aspect of the invention, a LiDAR system is
provided that is relatively immune from interference caused by
other LiDAR systems operating nearby. The LiDAR system includes a
coherent light source that generates a frequency modulated optical
signal comprising a series of optical chirps. The coherent light
source has a narrow linewidth, e.g., a linewidth of about 100 kHz
or less, 50 kHz or less, 10 kHz or less, 1 kHz or less, 100 Hz or
less, or 1 Hz or less. The LiDAR system also includes a scanning
assembly that (a) transmits the series of optical chirps in a scan
pattern across a scanning region and (b) receives a plurality of
reflected optical chirps corresponding to the transmitted optical
chirps that have reflected off one or more objects located within
the scanning region. In addition, the LiDAR system includes a
photodetector that mixes the reflected optical chirps with a local
oscillation (LO) reference signal comprising a series of LO
reference chirps. Further, the LiDAR system includes an electronic
data analysis assembly that processes digital data derived from the
reflected optical chirps and the LO reference chirps mixed at the
photodetector to generate distance data and optionally velocity
data associated with each of the reflected optical chirps. In this
implementation, the probability of interference between the
transmitted optical chirps and optical chirps transmitted by other
LiDAR systems is less than 1.times.10.sup.-7.
[0007] In a second aspect of the invention, the LiDAR system is
provided in the form of a sensor that is compact and lightweight
and, in addition, provides solid state reliability and
environmental tolerance. The LiDAR sensor includes a single
coherent light source that generates a frequency modulated optical
signal comprising a series of optical chirps. The LiDAR sensor also
includes a single scanner that (a) transmits the series of optical
chirps in a scan pattern across a scanning region and (b) receives
a plurality of reflected optical chirps corresponding to the
transmitted optical chirps that have reflected off one or more
objects located within the scanning region. In addition, the LiDAR
sensor includes a photodetector that mixes the reflected optical
chirps with a local oscillation (LO) reference signal comprising a
series of LO reference chirps generated by the coherent light
source. Optionally, the LiDAR sensor includes an electronic data
analysis assembly that processes digital data derived from the
reflected optical chirps and the LO reference chirps mixed at the
photodetector to generate distance data and optionally velocity
data associated with each of the reflected optical chirps. In this
implementation, the components are provided in a volume between 10
cm.sup.3 and 500 cm.sup.3.
[0008] In a third aspect of the invention, the LiDAR system
generates, transmits and receives light at a rate that enables data
points to be processed at a high output rate (i.e., pixel rate) so
as to increase the point density on each object of interest. In one
embodiment, the LiDAR system includes a coherent light source that
generates a frequency modulated optical signal comprising a series
of optical chirps each of which is highly linear, e.g., the chirp
linearity is 80%, 90%, 95% or more. This high chirp linearity is
achieved without predistortion or feedback correction to ensure
linearization of the frequency chirps, i.e., the coherent light
source inherently provides the desired chirp linearity. The LiDAR
system also includes a scanning assembly that (a) transmits the
series of optical chirps in a scan pattern across a scanning region
and (b) receives a plurality of reflected optical chirps
corresponding to the transmitted optical chirps that have reflected
off one or more objects located within the scanning region. In
addition, the LiDAR system includes a photodetector that mixes the
reflected optical chirps with a local oscillation (LO) reference
signal comprising a series of LO reference chirps. Further, the
LiDAR system includes an electronic data analysis assembly that
processes digital data derived from the reflected optical chirps
and the LO reference chirps mixed at the photodetector to generate
distance data and optionally velocity data associated with each of
the reflected optical chirps. In this implementation, the data
points are processed at an output rate in the range of 50,000 to
500,000 points/second.
[0009] In a fourth aspect of the invention, the LiDAR system is
able to detect objects located a great distance from the system
while transmitting light at a relatively low level of optical
power. In one embodiment, the LiDAR system includes a coherent
light source that generates a frequency modulated optical signal
comprising a series of optical chirps. The LiDAR system also
includes a scanning assembly that (a) transmits the series of
optical chirps in a scan pattern across a scanning region at an
optical power in the range of 10 mW to 3 W and (b) receives a
plurality of reflected optical chirps corresponding to the
transmitted optical chirps that have reflected off one or more
objects located within the scanning region. Each of the objects is
detectable at a distance of 200 meters or more (e.g., up to 1
kilometer) depending on the reflectivity of the object and the
transmitted optical power. In addition, the LiDAR system includes a
photodetector that mixes the reflected optical chirps with a local
oscillation (LO) reference signal comprising a series of LO
reference chirps. Further, the LiDAR system includes an electronic
data analysis assembly that processes digital data derived from the
reflected optical chirps and the LO reference chirps mixed at the
photodetector to generate distance data and optionally velocity
data associated with each of the reflected optical chirps.
[0010] In a fifth aspect of the invention, the LiDAR system is able
to directly measure both distance and velocity of an object in real
time, which provides significantly lower latency. The LiDAR system
includes a coherent light source that generates a frequency
modulated optical signal comprising a series of optical chirps. The
LiDAR system also includes a scanning assembly that (a) transmits
the series of optical chirps in a scan pattern across a scanning
region and (b) receives a plurality of reflected optical chirps
corresponding to the transmitted optical chirps that have reflected
off one or more objects located within the scanning region. In
addition, the LiDAR system includes a photodetector that mixes the
reflected optical chirps with a local oscillation (LO) reference
signal comprising a series of LO reference chirps. Further, the
LiDAR system includes an electronic data analysis assembly that
processes digital data derived from the reflected optical chirps
and the LO reference chirps mixed at the photodetector to generate
both distance data and velocity data associated with each of the
reflected optical chirps. In this implementation, the
point-to-point latency between successive data points is in the
range of 220 .mu.s to 100 ms.
[0011] In a sixth aspect of the invention, the LiDAR system
provides an adaptive scanning function. In one embodiment, the
LiDAR system includes a coherent light source that generates a
frequency modulated optical signal comprising a series of optical
chirps. The LiDAR system also includes a scanning assembly that (a)
transmits the series of optical chirps in a scan pattern across a
scanning region and (b) receives a plurality of reflected optical
chirps corresponding to the transmitted optical chirps that have
reflected off one or more objects located within the scanning
region. In addition, the LiDAR system includes a photodetector that
mixes the reflected optical chirps with a local oscillation (LO)
reference signal comprising a series of LO reference chirps.
Further, the LiDAR system includes an electronic data analysis and
control assembly that processes digital data derived from the
reflected optical chirps and the LO reference chirps mixed at the
photodetector to generate distance data and optionally velocity
data associated with each of the reflected optical chirps. The
electronic data analysis and control assembly also generates the
scan pattern to be used by the scanning assembly based on one or
more scan parameters. The scan parameters may identify the frame
size (e.g., a full frame size or a reduced frame size), the spatial
resolution (e.g., the number of scan lines and number of pixels per
scan line) for the identified frame size, and a non-zero offset for
the reduced frame size (e.g., the displacement of the reduced frame
size relative to the full frame size in at least one dimension such
that the reduced frame size is not centered with respect to the
full frame size). As such, coarse scanning may be used to locate
one or more objects of interest in the full frame size and fine
scanning may be used on certain objects of interest in the reduced
frame size resulting in greater point density on objects of
interest and better spatial resolution at a distance. Adaptive
scanning may also be used beyond adaptive frame placement, e.g., it
is possible to implement adaptive chirp repetition rate, chirp
bandwidth, chirp configuration, dedication of additional processing
power, and other parameters.
[0012] In a seventh aspect of the invention, the LiDAR system is
eye safe by virtue of its laser wavelength in the range of 1400 nm
to 4000 nm (e.g., the C-band of 1530 nm to 1560 nm) in combination
with a lower transmitted optical power. The LiDAR system includes a
coherent light source that generates a frequency modulated optical
signal comprising a series of optical chirps. The LiDAR system also
includes a scanning assembly that (a) transmits the series of
optical chirps in a scan pattern across a scanning region at an
optical power in the range of 10 mW to 3 W and (b) receives a
plurality of reflected optical chirps corresponding to the
transmitted optical chirps that have reflected off one or more
objects located within the scanning region. In addition, the LiDAR
system includes a photodetector that mixes the reflected optical
chirps with a local oscillation (LO) reference signal comprising a
series of LO reference chirps. Further, the LiDAR system includes
an electronic data analysis assembly that processes digital data
derived from the reflected optical chirps and the LO reference
chirps mixed at the photodetector to generate a plurality of data
points each of which comprises distance data and optionally
velocity data associated with one of the reflected optical chirps.
In this implementation, the data points may be processed at an
output rate in the range of 50,000 to 500,000 points/second.
[0013] Additional aspects of the invention, together with the
advantages and features relating thereto, are described in detail
below and/or will be apparent to one skilled in the art based on
the description of the invention. Also, while certain aspects of
the invention are set forth in the independent and dependent claims
that follow the description of the invention, one skilled in the
art will understand that individual features of these claims may be
combined in a number of different ways and that the invention is
not limited to any particular combination of features.
BRIEF DESCRIPTION OF THE DRAWINGS
[0014] Various embodiments of the present invention are described
in detail below with reference to the attached drawing figures,
wherein:
[0015] FIG. 1 is a block diagram of an embodiment of a LiDAR system
in accordance with the present invention;
[0016] FIG. 2 is a block diagram of an embodiment of a laser that
may be used in the LiDAR system of FIG. 1;
[0017] FIG. 3 is a block diagram of another embodiment of a laser
that may be used in the LiDAR system of FIG. 1;
[0018] FIG. 4 is a block diagram of yet another embodiment of a
laser that may be used in the LiDAR system of FIG. 1;
[0019] FIG. 5 illustrates an embodiment of a modulatable resonator
in which electrodes are placed on opposing planar faces of a
discoidal WGM resonator to enable frequency modulation of the laser
output, which may be used in the laser of FIG. 1;
[0020] FIG. 6 illustrates an optical signal in the form of an FMCW
sawtooth waveform that may be generated by the laser of FIG. 1;
[0021] FIG. 7 illustrates an optical signal in the form of an FMCW
triangular waveform that may be generated by the laser of FIG.
1;
[0022] FIG. 8 illustrates an optical signal in the form of an FM
pulsed triangular waveform that may be generated by the laser of
FIG. 1;
[0023] FIG. 9 illustrates an optical signal in the form of an FMCW
sigmoid-shaped waveform that may be generated by the laser of FIG.
1;
[0024] FIG. 10 is a block diagram of an alternative embodiment of a
mixer assembly that may be used in the LiDAR system of FIG. 1;
[0025] FIG. 11 is a block diagram of another alternative embodiment
of a mixer assembly that may be used in the LiDAR system of FIG.
1;
[0026] FIG. 12 is a graph that shows maximum range as a function of
the reflectivity of an object;
[0027] FIG. 13A illustrates a single optical chirp of the optical
signal of FIG. 6 (which serves as the local oscillator reference
signal) along with the corresponding reflected optical chirp, and
FIG. 13B illustrates the Fourier transformation of the time-domain
data;
[0028] FIG. 14A illustrates a single optical chirp of the optical
signal of FIG. 7 (which serves as the local oscillator reference
signal) along with the corresponding reflected optical chirp, and
FIG. 14B illustrates the Fourier transformation of the time-domain
data;
[0029] FIG. 15 illustrates an example method for providing object
detection;
[0030] FIG. 16 illustrates an example data structure of detected
objects in a frame;
[0031] FIG. 17A illustrates an example scan pattern of the MEMS
mirror of FIG. 1, FIG. 17B illustrates the horizontal component of
the scan, and FIG. 17C illustrates the vertical component of the
scan; and
[0032] FIG. 18A illustrates example scan patterns (full scan and
adaptive scan) of the MEMS mirror of FIG. 1, FIG. 18B illustrates
the horizontal component of the scans, and FIG. 18C illustrates the
vertical component of the scans.
DETAILED DESCRIPTION
[0033] The present invention is directed to a LiDAR system that
provides performance characteristics and other advantages that are
not achievable with conventional ToF and FMCW LiDAR systems. All or
a portion of the components of the LiDAR system may be packaged as
a solid state LiDAR sensor that can be used in automotive and other
applications. While the invention will be described in detail below
with reference to various embodiments, it should be understood that
the invention is not limited to the specific configuration or
methodology of these embodiments. In addition, although the
embodiments are described as embodying several different inventive
features, one skilled in the art will appreciate that any one of
these features could be implemented without the others in
accordance with the invention.
A. LiDAR System
[0034] In general terms, the LiDAR system of the present invention
includes a light source for generating light and a scanning
assembly that transmits the generated light across a scanning
region. The scanning assembly also receives light that has
reflected off one or more objects located within the scanning
region. An electronic data analysis assembly is provided that
generates a plurality of data points each of which comprises
distance data associated with a portion of the reflected light. The
distance data indicates the distance of the object that provided
the portion of the reflected light relative to the system. In
certain embodiments, each of the data points also comprises
velocity data that indicates the velocity of the object that
provided the portion of the reflected light relative to the system.
While the LiDAR system of the present invention is preferably
provided as a coherent LiDAR system, such as an FMCW LiDAR system,
certain aspects of the invention may be incorporated into a ToF
LiDAR system as described herein.
[0035] Referring to FIG. 1, a block diagram of an embodiment of a
LiDAR system in accordance with the present invention is shown
generally as reference numeral 100. LiDAR system 100 includes a
number of different optical and electronic components, including: a
laser assembly that, in the illustrated embodiment, comprises a
laser 102 and an associated laser driver/modulator 104; an optical
amplifier 106; a mixer assembly that, in the illustrated
embodiment, comprises a beam splitter 108, a circulator 110, a
fixed mirror 112 and a beam combiner 114; a scanning assembly that,
in the illustrated embodiment, comprises a microelectromechanical
system (MEMS) mirror 116 and an associated mirror driver 118; a
photodetector 120; a direct current (DC) filter 121; an electronic
amplifier 122; an analog-to-digital converter (ADC) 124; and an
electronic data analysis and control assembly that, in the
illustrated embodiment, comprises a field programmable gate array
(FPGA) 126 and a computer 128. Each of these components will be
described below. Of course, it should be understood that the
components of LiDAR system 100 are merely examples and that a
variety of different components and system configurations may be
used to implement the present invention.
[0036] The optical and electronic components of LiDAR system 100
may be manufactured in a variety of different ways, e.g., assembly
of individual optical and/or electronic components on a suitable
circuit board, use of small-scale electro-optical circuit board
technology or fabrication on a silicon chip using photolithographic
techniques or a combination of photolithographic and conventional
optical manufacturing techniques, including flip chip techniques.
All or a portion of the components may be packaged as a solid state
LiDAR sensor (i.e., an integrated sensor with no moving parts) that
can be used in automotive and other applications, as discussed
below. In certain embodiments, all of the components are provided
on a single operational surface (e.g., circuit board or silicon
chip) to form an integrated LiDAR system. In other embodiments, the
components are distributed among two or more operational surfaces.
These operational surfaces may be located in close proximity to
each other or may be located in different locations to form a
distributed LiDAR system. Communication between optical components
can be provided using an optical fiber or other type of optical
waveguide, and communication between electronic components can be
provided using a wire, lead, or other type of electrical
connection. Of course, other types of manufacturing methods are
possible and within the scope of the present invention.
1. Generation of Optical Chirps
[0037] The LiDAR system of the present invention includes a
coherent light source having a narrow linewidth for generating a
frequency modulated (FM) optical signal that comprises a series of
optical "chirps." In the embodiment shown in FIG. 1, the coherent
light source comprises a laser assembly in which a laser 102 is
driven and controlled by a laser driver/modulator 104. Preferably,
laser 102 has a narrow linewidth of 100 kHz or less, more
preferably 50 kHz or less, and most preferably 10 kHz or less,
i.e., 10 kHz, 9 kHz, 8 kHz, 7 kHz, 6 kHz, 5 kHz, 4 kHz, 3 kHz, 2
kHz, 1 kHz or less (or any value therebetween). The linewidth of
laser 102 influences the performance of LiDAR system 100 in
different ways. For example, a narrower linewidth provides a higher
signal to noise ratio (SNR), as discussed below, and a higher SNR
increases the distance (i.e., maximum range) at which objects can
be detected for a fixed optical power. In contrast, a laser light
source with a wider linewidth provides a lower SNR, and can only
perform at shorter distances. As such, the performance of LiDAR
system 100 is optimized when using a laser having a linewidth as
narrow as possible. Most preferably, laser 102 has an ultra-narrow
linewidth of 1 kHz or less, i.e., 1 kHz, 500 Hz, 250 Hz, 100 Hz, 50
Hz, 10 Hz, 5 Hz, 1 Hz or less (or any value therebetween). In
addition, as discussed below, the linearity of the chirp affects
the performance of LiDAR system 100. It should be noted that the
use of a narrower linewidth enables the chirp to be generated at a
lower degree of linearity to achieve the same SNR.
[0038] One skilled in the art will understand that the linewidth of
a laser has an inverse relationship to the length of the laser's
optical cavity, i.e., the longer the length of the laser's optical
cavity, the narrower the linewidth. A laser, independent of its
linewidth, can also be combined with an external optical cavity to
extend the optical cavity length and further narrow the linewidth.
Examples include a laser stabilized to an external Fabrey-Perot
cavity, or other types of optical cavities with a high quality
factor and finesse. Locking of the laser to the external optical
cavity may be implemented electronically, as in the well-known
Pound-Drever-Hall (PDH) scheme, or by direct optical injection
locking. Thus, laser 102 may comprise any type of laser, either by
itself or in combination with an external optical cavity, which
provides the desired linewidth as described above.
[0039] In the embodiment illustrated in FIG. 1, laser 102 comprises
a source laser, such as a diode laser, in which the linewidth is
narrowed via optical injection locking to a whispering gallery mode
(WGM) resonator. The WGM resonator supports a whispering gallery
mode that serves to entrap and/or accumulate light within the
resonator. The whispering gallery mode corresponds to a single
frequency produced by the source laser. Such a resonator can have a
variety of configurations, such as spherical, toroidal,
cylindrical, ring, or similar axis-symmetric configurations, and
can be constructed from various optically transparent materials
such as glass, quartz, calcium fluoride, fused silica, lithium
niobate, silicon, silicon nitride, and other materials that are
transparent to the selected optical wavelength. Light from the
source laser is coupled into the WGM resonator by any suitable
optical coupling device, such as a prism or waveguide. Light from a
counter propagating whispering gallery mode wave is coupled out of
the WGM resonator (by the same or different optical coupling
device) to provide optical injection locking of the source laser,
substantially reducing its linewidth. In this embodiment, the
linewidth of the source laser locked to the WGM resonator may be as
low as 100 Hz or may even be less than 1 Hz. As described below,
the WGM resonator is constructed from a material with one or more
optical properties, such as refractive index, which can be altered
via the application of an electric potential, stress or strain,
heat or a combination thereof, thereby altering the frequency
supported by the whispering gallery mode. This alters the frequency
of the counter propagating whispering gallery mode wave that is
directed back to the source laser via optical injection locking to
alter the output frequency of the source laser.
[0040] A variety of configurations for laser 102 are possible and
within the scope of the invention. One embodiment is shown in FIG.
2. In this embodiment, a source laser 200 provides a light beam
that is transferred to a WGM resonator 202 by a first optical
coupler 204, such as a prism. A collimator or optical mode selector
206 may optionally be interposed between source laser 200 and WGM
resonator 202. The light beam from source laser 200 is coupled into
WGM resonator 202 wherein a frequency represented in the light beam
propagates as a whispering gallery mode wave that is entrapped
within WGM resonator 202. A portion of the entrapped light is
coupled out of WGM resonator 202 by a second optical coupler 208
and propagates to a set of feedback optics 210. In this embodiment,
feedback optics 210 includes a reflector 212 and an optical spatial
filter 214 (e.g., a pinhole) positioned between second optical
coupler 208 and reflector 212 in order to improve the spatial
distribution of the light beam. A collimator or optical mode
selector 216 may optionally be interposed between second optical
coupler 208 and optical spatial filter 214. Light returned from
reflector 212 is coupled back into WGM resonator 202 to provide a
counter propagating wave. At least a portion of this counter
propagating wave is transferred out of WGM resonator 202 by the
first optical coupler 204 and propagates back toward source laser
200. This returning light provides optical injection that locks
source laser 200 to the frequency corresponding to the whispering
gallery mode of WGM resonator 202, thereby providing a laser output
218 with a narrowed linewidth compared to that of source laser 200
without such optical injection locking. It should be understood
that one or more optical properties of WGM resonator 202 may be
altered via the application of an electric potential, stress or
strain, heat or a combination thereof (as controlled by laser
driver/modulator 104), thereby altering the frequency supported by
the whispering gallery mode. This alters the frequency of the
counter propagating whispering gallery mode wave that is directed
back to source laser 200 to alter the output frequency of source
laser 200 via optical injection.
[0041] Another embodiment is shown in FIG. 3 in which the various
components are connected by waveguides such as optical fibers,
channels or other guiding structures comprising optically
conductive materials. In this embodiment, a source laser 300 is
coupled to a WGM resonator 302 via a waveguide 304. A collimator or
optical mode selector 306 may optionally be interposed between
source laser 300 and WGM resonator 302, with optical communication
provided by an additional waveguide 308. Light from source laser
300 is coupled into WGM resonator 302 wherein light corresponding
to the whispering gallery mode is entrapped. At least a portion of
the entrapped light is coupled out of WGM resonator 302 to a
waveguide 310 and propagates toward a reflector 312. A collimator
or optical mode selector 314 may optionally be interposed between
WGM resonator 302 and reflector 312, with optical communication
provided by an additional waveguide 316. Light returned from
reflector 312 is coupled back into WGM resonator 302 to provide a
counter propagating wave. At least a portion of this counter
propagating wave is transferred out of WGM resonator 302 and
propagates back toward source laser 300. This returning light
provides optical injection that locks source laser 300 to the
frequency corresponding to the whispering gallery mode of WGM
resonator 302, thereby providing a laser output 318 with a narrowed
linewidth compared to that of source laser 300 without such optical
injection locking. It should be understood that one or more optical
properties of WGM resonator 302 may be altered via the application
of an electric potential, stress or strain, heat or a combination
thereof (as controlled by laser driver/modulator 104), thereby
altering the frequency supported by the whispering gallery mode.
This alters the frequency of the counter propagating whispering
gallery mode wave that is directed back to source laser 300 to
alter the output frequency of source laser 300 via optical
injection.
[0042] While the embodiments shown in FIGS. 2 and 3 utilize a
reflector to provide a counter propagating wave within the WGM
resonator, the embodiment shown in FIG. 4 does not require the use
of a reflector and requires only a single optical coupler. In this
embodiment, a source laser 400 provides a light beam that is
transferred to a WGM resonator 402 by an optical coupler 404, such
as a prism or waveguide. A collimator or optical mode selector 406
may optionally be interposed between source laser 400 and WGM
resonator 402. The light beam from source laser 400 is coupled into
WGM resonator 402 wherein a frequency represented in the light beam
propagates as a whispering gallery mode wave that is entrapped
within WGM resonator 202. In one example, light scattering within
the material of WGM resonator 402 provides a counter propagating
wave of sufficient intensity to be useful in optical injection
locking of source laser 400. In another example, features are
introduced into and/or on the surface of WGM resonator 402 to
provide a counter propagating wave. Suitable features include
inclusions within the body of WGM resonator 402, pits, channels, or
other features generated on the surface of WGM resonator 402,
and/or an optical grating generated on the surface of WGM resonator
402. At least a portion of this counter propagating wave is
transferred out of WGM resonator 402 by optical coupler 404 and
propagates back toward source laser 400. This returning light
provides optical injection that locks source laser 400 to the
frequency corresponding to the whispering gallery mode of WGM
resonator 402, thereby providing a laser output 408 with a narrowed
linewidth compared to that of source laser 400 without such optical
injection locking. It should be understood that one or more optical
properties of WGM resonator 402 may be altered via the application
of an electric potential, stress or strain, heat or a combination
thereof (as controlled by a control circuit in laser
driver/modulator 104), thereby altering the frequency supported by
the whispering gallery mode. This alters the frequency of the
counter propagating whispering gallery mode wave that is directed
back to source laser 400 to alter the output frequency of source
laser 400 via optical injection.
[0043] Of course, one skilled in the art will understand that laser
102 may comprise other types of source lasers and/or optical
resonators that are capable of narrowing the linewidth of a source
laser, i.e., the present invention is not limited to the use of a
diode laser in which the linewidth is narrowed via optical
injection locking to a WGM resonator. For example, it is possible
to use a source laser with other types of optical resonators, such
as a Fabrey-Perot resonator, resonators formed with Bragg grating
mirrors, fiber resonators, and ring resonators. However, these
optical resonators have a lower quality factor than a WGM resonator
and, thus, will provide a wider linewidth. It is also possible to
use other types of source lasers, such as fiber lasers, lasers
based on solid state crystals (e.g., a YAG laser), gas lasers, or
chemical lasers. These source lasers typically have a linewidth
greater than 1 kHz and can optionally be combined with an external
Fabrey-Perot resonator or similar type cavity to reduce their
linewidth. These lasers, however, are bulky structures and
typically weigh in excess of several kilograms and require a large
amount of power to operate. In contrast, the diode laser and WGM
resonator described above are light, compact and economical and,
thus, are more suitable for use in LiDAR system 100 of FIG. 1.
[0044] Referring back to FIG. 1, laser driver/modulator 104 is
provided to drive and control laser 102 in such a manner as to
generate an FM optical signal comprising a series of optical chirps
as the laser output. In this embodiment, the WGM resonator of laser
102 is frequency modulatable, i.e., the frequency supported by the
whispering gallery mode of the WGM resonator can be altered, which
alters the frequency of the laser output. The WGM resonator is
constructed from a material (e.g., Lithium Tantalate, Lithium
Niobate, Aluminum Nitride, fused silica, calcium fluoride, etc.)
with one or more optical properties, such as refractive index,
which can be altered via the application of an electric potential,
stress or strain, heat or a combination thereof, thereby altering
the frequency supported by the whispering gallery mode. A frequency
chirp produced in this fashion is highly linear, i.e., the
linearity of the chirp is 80% or more and may be as high as 99% or
more, as described below. Alternatively, the laser (and not the
resonator) of laser 102 may be modulated to alter the frequency of
the laser output, such as by changing the current or temperature of
the laser. However, this approach will cause stress to the laser
and can lead to nonlinearities of the frequency chirp that will
require predistortion or feedback correction to ensure
linearization of the frequency chirp. As such, modulation of the
resonator is preferred in certain implementations.
[0045] In one embodiment, laser driver/modulator 104 applies an
electric potential to electrodes coupled to the WGM resonator to
thereby alter the refractive index of the WGM resonator (in this
case, the WGM material is constructed from an electro-optic
material such as lithium niobate, lithium tantalite, or aluminum
nitide). In another embodiment, laser driver/modulator 104 controls
a piezoelectric element coupled to the WGM resonator in order to
alter the refractive index of the WGM resonator upon the
application of stress or strain. In yet another embodiment, laser
driver/modulator 104 controls a resistive heater that alters the
refractive index of the WGM resonator upon the application of
heat.
[0046] An embodiment of a frequency modulatable WGM resonator is
shown in FIG. 5. In this embodiment, a WGM resonator 500 has a
discoidal shape and is in contact with an upper electrode 502 and a
lower electrode 504 applied to opposing planar faces of WGM
resonator 500. Laser driver/modulator 104 (not shown in FIG. 5)
applies an electric potential to electrodes 502, 504 to alter the
refractive index of WGM resonator 500. This alters the frequency
supported by the whispering gallery mode of WGM resonator 500,
which in turn causes the source laser to follow the resonator mode
frequency and be locked to WGM resonator 500 at a different
frequency. Modulation of the refractive index of WGM resonator 500
alters the frequency of the laser output over time to thereby
enable the generation of optical chirps.
[0047] It should be noted that electrical tuning (such as in the
embodiment just described) is preferred to piezoelectric tuning or
thermal tuning in certain implementations. For example, the
embodiment shown in FIG. 5 enables the generation of optical chirps
with a chirp bandwidth in the GHz range and at a chirp rate in
excess of 1.times.10.sup.14 Hz/s, as described below. In contrast,
piezoelectric tuning provides a narrower chirp bandwidth (e.g.,
10-100 MHz) albeit at a comparable chirp rate. Thermal tuning
provides a comparable chirp bandwidth but at a lower chirp rate
(e.g., 1.times.10.sup.11 Hz/s).
[0048] As described below, laser driver/modulator 104 receives
chirp parameters (e.g., chirp configuration, chirp period and chirp
bandwidth) from FPGA 126 and uses such chirp parameters to alter an
optical property (e.g., refractive index) of the WGM resonator over
time such that laser 102 generates an FM optical signal comprising
a series of optical chirps. One skilled in the art will appreciate
that such an arrangement permits generation of a wide variety of
chirp patterns, which can be suitable for different
implementations. Various examples of chirp patterns that may be
generated by laser 102 are shown in FIGS. 6-9.
[0049] In one embodiment, laser 102 generates an optical signal in
which the frequency of the signal is continuously modulated to
provide the FMCW sawtooth waveform shown in FIG. 6. This optical
signal includes a continuous series of optical chirps, such as
optical chirp 600. As can be seen, optical chirp 600 includes an
up-chirp portion 602 in which the frequency of the signal increases
linearly with time. As described below, the use of this chirp
pattern enables direct measurement of the distance of reflective
objects.
[0050] In another embodiment, laser 102 generates an optical signal
in which the frequency of the signal is continuously modulated to
provide the FMCW triangular waveform shown in FIG. 7. This optical
signal includes a continuous series of optical chirps, such as
optical chirp 700. As can be seen, optical chirp 700 includes an
up-chirp portion 702 in which the frequency of the signal increases
linearly with time and a down-chirp portion 704 in which the
frequency of the signal decreases linearly with time. As described
below, the use of this chirp pattern enables direct measurement of
both the distance and velocity of reflective objects.
[0051] In yet another embodiment, laser 102 generates an optical
signal in which the frequency of the signal is modulated to provide
the FM pulsed triangular waveform shown in FIG. 8. This optical
signal includes a pulsed series of optical chirps, such as optical
chirp 800, and is similar to the optical signal shown in FIG. 7
insofar as each of optical chirps includes an up-chirp portion 802
and a down-chirp portion 804 to enable direct measurement of both
the distance and velocity of reflective objects. However, the
frequency of the optical signal is not continuously modulated. This
pulsed chirp pattern may be used in certain coherent LiDAR systems
that are limited by downstream processing or in which the use of a
pulsed chirp is otherwise more advantageous.
[0052] In a further embodiment, laser 102 generates an optical
signal in which the frequency of the signal is continuously
modulated to provide the FMCW sigmoid-shaped waveform shown in FIG.
9. This optical signal includes a continuous series of optical
chirps, such as optical chirp 900. As can be seen, optical chirp
900 includes a down-chirp portion 902 in which the frequency of the
signal decreases nonlinearly with time in a sigmoidal fashion and
an up-chirp portion 904 in which the frequency of the signal
increases nonlinearly with time in a sigmoidal fashion. Each of the
optical chirps can also change in amplitude 906 (shown in dashed
lines) during the duration of the chirp. For example, the amplitude
may increase rapidly during the initial portion of the chirp and
decrease rapidly over the final portion of the chirp. Other types
of nonlinear optical chirps may also be used, such as quadratic,
cubic, staircase, or any other shape in which the frequency is
modulated in such a way that it is monotonic over half of the
modulation period (i.e., chirps in which the frequency never
decreases on any up-chirp portion and never increases on any
down-chirp portion). Nonlinear chirps may be used in covert
operations, or in certain systems that require reduction of
interference and/or reduction of false object detection beyond that
provided by the linear chirp patterns discussed above. It is also
possible to use hybrid chirp patterns in which each optical chirp
is linear through a first portion of the chirp and nonlinear (e.g.,
sigmoidal) through a second portion of the chirp. It should be
appreciated that optical chirp patterns that include any element of
nonlinearity will increase the processing time required to analyze
the reflected optical chirps by FPGA 126 and, thus, may not be
suitable for certain implementations.
[0053] Preferably, laser 102 generates an FM optical signal in
which the variation between optical chirps is minimized or in which
there is a controlled variation between optical chirps. Optical
chirps can be characterized by a number of quantifiable factors,
including chirp linearity, chirp bandwidth, chirp period and chirp
rate (each of which will be described below). Variation within a
population of optical chirps can be expressed as a standard
deviation and/or coefficient of variation (CV) surrounding a
central value for such quantifiable factors. In certain
embodiments, the CV for each of these factors for a statistically
significant group of optical chirps (e.g., greater than 32) is less
than 25%, i.e., 25%, 20%, 15%, 10%, 7.5%, 5%, 2.5%, 1%, 0.5%,
0.25%, 0.1% or less (or any value therebetween). Each of these
factors will now be described.
[0054] With reference to the chirp patterns shown in FIGS. 6-9, it
can be seen that each optical chirp has an instantaneous frequency
that changes linearly with time (FIGS. 6-8) or that changes
nonlinearly with time in such a manner that the nonlinear function
is defined (FIG. 9). The linearity of a generated optical chirp may
be expressed as a correlation between the measured frequency and
the ideal frequency at various points in time. A generated optical
chirp that is perfectly linear (or perfectly nonlinear) will have a
linearity of 100% at each time point (e.g., the optical chirps
shown in FIGS. 6-9), while deviations from ideal behavior will
result in a linearity of less than 100%.
[0055] Preferably, each optical chirp generated by laser 102 is
substantially linear such that the linearity is greater than 80% at
each point in time, i.e., 80%, 82%, 84%, 86%, 88%, 90%, 92%, 94%,
96%, 98%, or 100% (or any value therebetween). It should be
understood that chirp linearity increases the SNR of the received
optical chirps, which can be calculated from the following
equation:
SNR = I signal 2 2 eI signal B Rx + 4 k B TB Rx Z + I RIN + V quant
Z + I dark ( 1 ) ##EQU00001##
where: [0056] SNR=signal to noise ratio [0057] I.sub.signal=signal
intensity (see equation (2) below) [0058] e=charge of electron
[0059] B.sub.Rx=receiver bandwidth [0060] k.sub.B=Boltzmann's
constant [0061] T=temperature [0062] Z=load resistance [0063]
I.sub.RIN=laser relative intensity noise current [0064]
V.sub.quant=quantization voltage noise [0065]
I.sub.dark=photodetector dark current
[0066] With reference to equation (1), the signal intensity
(I.sub.signal) can be calculated from the following equation:
I signal = 2 P LO P Tx ( r Rx R ) 2 e - 2 R .alpha. .rho. .eta.
signal ( 2 ) ##EQU00002##
where: [0067] I.sub.signal=signal intensity [0068] =photodetector
responsivity [0069] P.sub.LO=local oscillator power [0070]
P.sub.Tx=transmitted (output beam) power [0071] r.sub.Rx=radius of
receiver area [0072] R=range to target object [0073] e=charge of
electron [0074] .alpha.=absorption depth of air [0075]
.rho.=reflectivity of target object [0076]
.eta..sub.signal=efficiency of signal (see equation (3) below)
[0077] With reference to equation (2), the efficiency of the signal
(.eta..sub.signal) can be calculated from the following
equation:
.eta..sub.signal=.eta..sub.target.eta..sub.pol.eta..sub.speckle.eta..sub-
.lin.eta..sub.optics.eta..sub.align.eta..sub.beam.eta..sub.Doppler
(3)
where: [0078] .eta..sub.signal=efficiency of signal [0079]
.eta..sub.target=efficiency of beam hitting target object (1 if
entire beam falls on object) [0080] .eta..sub.pol=efficiency of
depolarization from scattering (1 if polarization does not change)
[0081] .eta..sub.speckle=efficiency of speckle/target object
roughness (1 if object perfectly flat) [0082]
.eta..sub.lin=efficiency of chirp linearity (1 if chirp is
perfectly linear) [0083] .eta..sub.optics=efficiency of optics (1
if all optics in system have transmission=1) [0084]
.eta..sub.align=efficiency of interferometer alignment (1 if LO and
return beams perfectly aligned) [0085] .eta..sub.beam=efficiency of
interferometer shape matching (1 if LO and return beam profiles are
exact same shape) [0086] .eta..sub.Doppler=efficiency of Doppler
broadening of laser (1 if the moving components which beam is
incident on (e.g., mirror) have constant velocity across laser
profile)
[0087] Thus, increasing the efficiency of chirp linearity
(.eta..sub.lin) in equation (3) results in a higher signal
intensity (I.sub.signal) in equation (2), which in turn results in
a higher SNR in equation (1). As such, the SNR is optimized when
each optical chirp is highly linear. Of course, the other variables
in equations (1)-(3) should also be chosen to optimize the SNR as
required for a particular implementation. In a preferred
embodiment, each optical chirp generated by laser 102 has a
linearity greater than 95% at each point in time, i.e., 95%, 96%,
97%, 98%, 99%, or 100% (or any value therebetween). Preferably, the
resultant SNR is 10 dB or higher, although LiDAR system 100 can
detect optical signals with a SNR as low as 2 or 3 dB at a maximum
range of 200 meters. In this embodiment, the SNR is in the range of
3 dB to 110 dB or a subrange within this broader range, e.g., 10 dB
to 110 dB or 10 dB to 80 dB. It can be appreciated that a larger
SNR is preferred in order to extend the maximum range of LiDAR
system 100 without increasing the power of the laser.
[0088] In the embodiment in which laser 102 comprises a source
laser optically coupled to a modulatable WGM resonator, the laser
configuration naturally exhibits high linearity in excess of 80%
without the need for predistortion or feedback correction to ensure
linearization of the frequency chirp. This is the case even at a
high output rate (i.e., pixel rate) of 50,000 points/second or
more, as described below. As such, laser 102 can be operated at a
high output rate with reduced complexity. In contrast, many
conventional FMCW LiDAR systems will lose the linearity of the
chirps at an output rate of 50,000 points/second or more unless
predistortion or feedback correction is used to ensure
linearization of the frequency chirp, such as through a measured
and recorded correction factor or through an active feedback system
(e.g., with a Mach-Zehnder interferometer).
[0089] With reference again to the chirp patterns shown in FIGS.
6-9, it can be seen that each optical chirp has a chirp bandwidth
defined by the range of frequencies within each optical chirp.
Preferably, each optical chirp generated by laser 102 has a chirp
bandwidth in the range of 1 GHz to 100 GHz, i.e., 1 GHz, 5 GHz, 10
GHz, 15 GHz, 20 GHz, 25 GHz, 30 GHz, 35 GHz, 40 GHz, 45 GHz, 50
GHz, 60 GHz, 70 GHz, 80 GHz, 90 GHz, 100 GHz (or any value
therebetween). It is also possible that the chirp bandwidth could
be greater than 100 GHz and up to 1 THz, i.e., 100 GHz, 200 GHz,
300 GHz. 400 GHz, 500 GHz, 600 GHz, 700 GHz, 800 GHz, 900 GHz, 1
THz (or any value therebetween). It should be understood that the
range resolution of LiDAR system 100 is a function of the chirp
bandwidth, as shown by the following equation:
.DELTA. R = c 2 B .times. T t ( 4 ) ##EQU00003##
where: [0090] .DELTA.R=range resolution in meters [0091] c=speed of
light (approximately 3.00.times.10.sup.8 meters/second) [0092]
B=chirp bandwidth in Hz [0093] T=chirp period in seconds [0094]
t=integration time in seconds
[0095] Thus, it can be seen that a greater chirp bandwidth results
in a smaller range resolution. Preferably, LiDAR system 100
provides a range resolution of 20 cm or less, i.e., 20 cm, 15 cm,
12.5 cm, 10 cm, 7.5 cm. 5 cm, 2.5 cm, 1 cm, 5 mm, 1 mm or less (or
any value therebetween). In addition, it should be understood that
the range resolution can be improved by oversampling in the data
processing or by using an FM pulsed waveform.
[0096] Referring again to the chirp patterns shown in FIGS. 6-9, it
can be seen that each optical chirp has a chirp period that
comprises the time duration of each optical chirp. Preferably, each
optical chirp generated by laser 102 has a chirp period in the
range of 50 kHz to 500 kHz or a subrange within this broader range,
e.g., 50 kHz to 375 kHz or 325 kHz to 375 kHz. It should be
appreciated that the chirp repetition rate (i.e., the rate at which
the optical chirps are transmitted and received by a scanner, such
as MEMS mirror 116) is a function of the chirp period. For example,
in a continuous wave, the chirp repetition rate will be equal to
the chirp period. Preferably, the output rate (i.e., pixel rate) is
matched to the chirp repetition rate, wherein the output rate is
the number of data points processed per second (as described in
greater detail below with reference to the operation of FPGA 126
and computer 128). Thus, the chirp period determines the chirp
repetition rate and the output rate of LiDAR system 100.
[0097] Referring yet again to the chirp patterns shown in FIGS.
6-9, it can be seen that each optical chirp has a chirp rate
comprising the rate at which the frequency changes within the chirp
(e.g., the slope of the chirp). For example, in the FMCW sawtooth
waveform shown in FIG. 6, the chirp rate is equal to the chirp
bandwidth (B) divided by the chirp period (T), i.e., B/T. As
another example, in the FMCW triangular waveform shown in FIG. 7,
the chirp rate is equal to 2 B/T. Preferably, each optical chirp
generated by laser 102 has a chirp rate in the range of
1.times.10.sup.14 Hz/s to 5.times.10.sup.16 Hz/s. Of course, it
should be understood that the chirp rate is determined by the chirp
bandwidth and chirp period that are chosen for a particular
implementation.
[0098] In summary, the laser assembly of LiDAR system 100 includes
a narrow linewidth laser 102 that, when driven and controlled by
laser drive/modulator 104, generates an FM optical signal
comprising a series of optical chirps. Preferably, each of the
optical chirps is substantially linear with a wide chirp bandwidth,
a short chirp period and a high chirp rate so as to optimize the
performance of LiDAR system 100, although a plurality of other
waveforms are also possible with the scope of the invention. The FM
optical signal generated by laser 102 may optionally be amplified
by optical amplifier 106 to increase the power level of the signal.
Of course, if the power level of the laser output is sufficiently
large for a particular implementation, then optical amplifier 106
is not required. An example of a suitable optical amplifier is a
semiconductor optical amplifier (SOA).
2. Transmission of Optical Chirps and Detection of Reflected
Optical Chirps
[0099] Referring back to FIG. 1, the FM optical signal generated by
laser 102 (which may optionally be amplified by optical amplifier
106) propagates to beam splitter 108, which divides the optical
signal into two separate signals--one of which propagates to
circulator 110 and the other of which propagates to beam combiner
114. It should be understood that these optical signals are
identical and include a series of optical chirps, such as any of
the chirp patterns shown in FIGS. 6-9, or other types of
waveforms.
[0100] Circulator 110 directs the FM optical signal received from
beam splitter 108 to a scanning assembly. In the illustrated
embodiment, the scanning assembly comprises a dual-axis MEMS mirror
116 that is driven and controlled by mirror driver 118 in such a
manner as to scan a region of space in accordance with a designated
scan pattern. As described in detail below, mirror driver 118
receives scan parameters (e.g., field of view, spatial resolution
and offset) from FPGA 126 and uses such scan parameters to set the
mirror position and timing for each point in the scan pattern.
[0101] MEMS mirror 116 enables the LiDAR system 100 to operate as a
monostatic device in which both the transmission and reception
functions are provided in a single device. Specifically, MEMS
mirror 116 controls the transmission of the optical chirps across
the scanning region and the reception of the incoming, reflected
optical chirps. If the FM optical signal is continuous, laser 102
is always in a state of being chirped and MEMS mirror 116
continuously transmits the optical chirps and receives the
reflected optical chirps at a desired chirp repetition rate (which
is a function of the chirp period (T), as discussed above). As
such, any object (such as object 130 shown in FIG. 1) located
within the scanning region may be illuminated for a partial chirp
period, a single chirp period, or multiple chirp periods, depending
on the implementation. The position from which the transmitted FM
optical chirps leave MEMS mirror 116 and the position from which
the reflected optical chirps are received by MEMS mirror 116 are
coaxial and proximal to provide a compact device.
[0102] It should be understood that a variety of devices are
suitable for providing the scanning function, i.e., the present
invention is not limited to the use of a single scanner such as
MEMS mirror 116. In certain embodiments, the transmission and
reception functions are provided by two separate devices. For
example, as shown in FIG. 10, it is possible to use one MEMS mirror
116a to provide the transmission function and another MEMS mirror
116b to provide the reception function. In this case, the
circulator 110 of FIG. 1 would not be required. Note that beam
splitter 108 and beam combiner 114 are the same as those in FIG. 1,
and fixed mirror 112 (not shown in FIG. 10) may be required in
certain implementations to direct light from MEMS mirror 116b to
beam combiner 114. Additional components would be needed to monitor
and maintain control over the alignment of the two MEMS mirrors
116a, 116b, as is known in the art. For this reason, the use of a
single MEMS mirror is preferred in certain implementations.
[0103] As another example, as shown in FIG. 11, it is possible to
use a MEMS mirror 116c to provide the transmission function and a
collector 116d to provide the reception function. In this case, the
circulator 110 of FIG. 1 would not be required. Note that beam
splitter 108 and beam combiner 114 are the same as those in FIG.
11, and fixed mirror 112 (not shown in FIG. 11) may be required in
certain implementations to direct light from collector 116d to beam
combiner 114. Because collector 116d sees the entire field of view,
off-angle external light could enter and be detected by the system
causing interference. This interference would mostly be
attributable to multiple reflections from a reflective object
(i.e., clutter) and would be filtered out by FPGA 126, as discussed
below. In comparison, many conventional ToF LiDAR systems use
non-angle selective receivers that see interference from other
light sources outside of the direction of interest, such as light
from the sun or light from other LiDAR systems. Thus, it should be
understood that the type of interference associated with ToF
systems is different from interference caused by clutter.
[0104] Other devices for providing the transmission and/or
reception functions include a phased array steering device,
rotating or gimbal-mounted mirrors, mirrors mounted on actuators
(e.g., electric motors, solenoids, and or piezoelectric devices) in
a mutually orthogonal fashion, rotating prisms, and/or rotating
lenses. Of course, one skilled in the art will appreciate that
other scanning devices are also possible and within the scope of
the present invention.
[0105] In one embodiment, the optical chirps are transmitted across
the scanning region at an optical power in the range of 10 mW to 3
W or a subrange within this broader range, e.g., 10 mW to 400 mW,
10 mW to 200 mW, 30 mW to 100 mW, or 40 mW to 60 mW. In a preferred
embodiment, the transmitted optical power is 50 mW.
[0106] As mentioned above, any object located within the scanning
region may be illuminated for a partial chirp period, a single
chirp period, or multiple chirp periods, depending on the
implementation. There is a limit on the greatest distance that an
object can be detected, which is referred to as "maximum range."
The maximum range is a function of the reflectivity of the object
and the SNR of the reflected optical chirps, as shown by the
following equation:
SNR .varies. .rho. R 2 ( 5 ) ##EQU00004##
where: [0107] SNR=signal to noise ratio of reflected optical chirps
[0108] .rho.=reflectivity of target object [0109] R=range to target
object
[0110] Thus, it can be appreciated from equation (5) that the
maximum range is proportional to the square root of the
reflectivity of the object at an equivalent SNR, as follows:
R.sub.max.varies. {square root over (.rho.)}(at equivalent SNR)
(6)
where: [0111] R.sub.max=maximum range to target object [0112]
.rho.=reflectivity of target object
[0113] In accordance with these principles, the graph of FIG. 12
shows an example of the maximum range as a function of the
reflectivity of an object. For purposes of this example, it is
assumed that the transmitted optical power is 50 mW and the minimum
detectable SNR is 3 dB. In this case, it can be seen that a 10%
reflective object can be detected at distances up to at least 200
meters, a 20% reflective object can be detected at distances up to
about 283 meters, a 30% reflective object can be detected at
distances up to about 346 meters, a 40% reflective object can be
detected at distances up to about 400 meters, a 50% reflective
object can be detected at distances up to about 447 meters, a 60%
reflective object can be detected at distances up to about 490
meters, a 70% reflective object can be detected at distances up to
about 529 meters, an 80% reflective object can be detected at
distances up to about 566 meters, a 90% reflective object can be
detected at distances up to about 600 meters, and a 100% reflective
object can be detected at distances up to about 632 meters. Of
course, it can be appreciated that the maximum range for any object
is dependent on the transmitted optical power. For example, the
maximum range can be increased up to 1 kilometer for a 100%
reflective object if the transmitted optical power is increased
above 50 mW. It can be appreciated that the maximum range in
relation to the transmitted optical power for a particular object
reflectivity may be derived from equations (1)-(3) above.
[0114] There is also a limit on the shortest distance that an
object can be detected, which is referred to as "minimum range." In
one embodiment, the minimum range is between 1 millimeter and 1
meter, and is dependent on the range resolution.
[0115] Further, there is a limit to the relative velocity of a
reflective object in order to enable detection of the reflected
optical chirp, which is referred to as "maximum velocity." In one
embodiment, the maximum velocity is in the range of 200 mph (89.4
m/s) to 450 mph (201.2 m/s), and is dependent on the transmitted
optical power (i.e., increasing the power increases the SNR and
thus indirectly allows velocity determination to occur) and chirp
bandwidth (i.e., increasing the bandwidth increases the maximum
velocity that can be measured). In addition, the velocity
resolution is between 0.2 mph (0.09 m/s) and 2 mph (0.9 m/s), and
is dependent on the integration time (i.e., decreasing the
integration time increases the velocity resolution that can be
measured).
[0116] In one embodiment, the source laser has a wavelength in the
range of 1400 nm to 4000 nm or a subrange within this broader
range, e.g., 1400 nm to 1800 nm, 1500 nm to 1800 nm, or 1530 nm to
1560 nm (i.e., the optical communications C-band). The accuracy of
the laser wavelength is preferably in the range of 0.01 nm to 30 nm
or a subrange within this broader range, e.g., 0.01 nm to 0.1 nm,
0.1 nm to 30 nm, or 0.1 nm to 10 nm. It can be appreciated that
these wavelengths enable penetration through sub-optimal
environmental conditions, such as clouds, dust, smoke, snow, and
rain. Of course, it is also possible to use source lasers having
other wavelengths, such as 895 nm, 1064 nm, 2 micrometers and any
other wavelengths. Because LiDAR system 100 has lower noise
sensitivity than conventional LiDAR systems at wavelengths in
excess of about 850 nm, the required transmitted optical power for
detection at an equivalent distance is lower. As such, LiDAR system
100 provides better eye safety than conventional LiDAR systems. Of
course, it should be understood that the permissible transmitted
optical power at these wavelengths to insure eye safety is a
function of the field of view and frame rate and preferably
satisfies the following equation:
P .times. f w .ltoreq. 533 mJ ( 7 ) ##EQU00005##
where: [0117] P=optical power in W [0118] f=frame rate in Hz [0119]
w=solid angle in steridians
[0120] Referring back to FIG. 1, each reflected optical chirp
received by MEMS mirror 116 propagates to circulator 110, which in
turn directs it to fixed mirror 112. Fixed mirror 112 is suitably
positioned to intercept each reflected optical chirp and direct it
to beam combiner 114. Of course, fixed mirror 112 would not be
required in a waveguide-based system. Beam combiner 114 combines
each reflected optical chirp with an optical chirp in the light
beam received from beam splitter 108. It can be appreciated that
each optical chirp in the light beam received from beam splitter
108 can serve as a local oscillation (LO) reference chirp that is
used to characterize a reflected optical chirp (discussed below).
As such, laser 102 acts as the source of each LO reference chirp to
reduce the complexity and size of LiDAR system 100, and a separate
local oscillator is not required.
[0121] After combination of a reflected optical chirp with an LO
reference chirp, the combined optical signal propagates to
photodetector 120. In this embodiment, photodetector 120 comprises
a PIN photodiode that coherently mixes each reflected optical chirp
with a corresponding LO reference chirp at the surface of the
photodiode and converts the mixed optical signal to an electrical
signal. Because the reflected optical chirp and LO reference chirp
are coincident on the surface of the photodiode, the total
intensity of the mixed optical signal can be expressed as
follows:
I=I.sub.LO+I.sub.reflected+2 {square root over
((I.sub.LOI.sub.reflected))}cos[(f.sub.LO-f.sub.reflected)t]
(8)
where: [0122] I=intensity of mixed optical signal in W [0123]
I.sub.LO=intensity of LO reference chirp in W [0124]
I.sub.reflected=intensity of reflected optical chirp in W [0125]
f.sub.LO=frequency of LO reference chirp in Hz [0126]
f.sub.reflected=frequency of reflected optical chirp in Hz
[0127] In equation (8), the first two terms (i.e.,
I.sub.LO+I.sub.reflected) represent the DC component of the optical
intensity and the third term represents the AC component of the
optical intensity. Because the reflected optical chirp and
corresponding LO reference chirp are coherent with one another,
their electric fields combine within the third term (i.e., the
third term is the product of the two electric fields). Importantly,
any incoherent contributions to the optical intensity originating
from external light sources (e.g., the sun, extreme lighting, light
caused by intentional interference) only add to the DC component of
the optical intensity.
[0128] It should be understood that photodetector 120 may comprise
other types of optical detectors, such as an avalanche photodiode
(APD), a single-photon avalanche diode (SPAD), a photo-multiplier
tube (PMT), a nanowire photon detector, or a quantum well
photodetector. Of course, the use of a PIN photodiode is preferred
due to its lower cost and relatively low noise generation. It
should be understood that conventional ToF LiDAR systems typically
cannot use a PIN photodiode because of the need for a more
sensitive photodetector to detect the relatively faint reflected
signal at longer ranges. As such, the ability to use a PIN
photodiode in the LiDAR system of the present invention provides
cost and functional advantages that are not possible with
conventional ToF LiDAR systems. Furthermore, since the signal out
of the photodetector has a frequency below a GHz, a low bandwidth
photodetector will suffice. In ToF LiDAR systems, the bandwidth of
the photodetector is typically multiples of GHz in order to provide
adequate performance.
[0129] In this embodiment, a DC filter 121 is provided to
electrically filter out the DC components of the signal. As such,
the electrical signal comprises the AC component of the optical
intensity (i.e., the third term of equation (8)) as converted into
electrical current, as follows:
J=2R {square root over
((I.sub.LOI.sub.reflected))}cos[(f.sub.LO-f.sub.reflected)t]
(9)
where: [0130] J=electrical current output of photodiode in amps
[0131] R=photodiode responsivity in amps/W [0132]
I.sub.LO=intensity of LO reference chirp in W [0133]
I.sub.reflected=intensity of reflected optical chirp in W [0134]
f.sub.LO=frequency of LO reference chirp in Hz [0135]
f.sub.reflected=frequency of reflected optical chirp in Hz
[0136] In other embodiments, two photodiodes may be used to
optically filter out the DC components of the optical intensity and
other DC-like noise (in which case DC filter 121 is not required).
Because the DC components of the optical intensity are filtered out
either electronically (by DC filter 121) or optically (by two
photodiodes), including those attributable to external light
sources, LiDAR system 100 has reduced to no interference from
external light sources. In contrast, conventional ToF LiDAR systems
are highly susceptible to interference caused by external light
sources. This issue becomes more pronounced as the distance between
the transmitter and the reflective object increases, as such
distance necessarily decreases the strength of the reflected
pulses.
[0137] The electrical signal generated by photodetector 120 and
optionally filtered by DC filter 121 may be amplified by electronic
amplifier 122 to increase the power level of the signal. Of course,
if the power level of the electrical signal is sufficient for a
particular implementation, then electronic amplifier 122 is not
required. The electronic signal is then routed to ADC 124, which
converts the electrical signal to digital data. It can be
appreciated that this digital data is derived from each reflected
optical chirp and corresponding LO reference chirp that were
coherently mixed at photodetector 120, i.e., a "pixel" of data.
This pixel data is then analyzed by FPGA 126, as described
below.
3. Data Analysis and Object Detection
[0138] In this embodiment, FPGA 126 or any other type of processing
device, such as an application specific integrated circuit (ASIC),
a central processing unit (CPU), or a graphics processing unit
(GPU), processes each pixel of data received from ADC 124 by
initially applying a fast Fourier transform algorithm to the
time-domain data. Two examples of this process will be described
below with reference to the FMCW sawtooth and triangular waveforms
shown in FIGS. 6 and 7, respectively.
[0139] In one example shown in FIG. 13A, one of the optical chirps
of the FMCW sawtooth waveform shown in FIG. 6 is designated as
optical chirp 1300. It should be understood that optical chirp 1300
is an LO reference chirp generated by laser 102, which is
representative of the transmitted optical chirp. The reflected
optical chirp corresponding to LO reference chirp 1300 is
designated as optical chirp 1302. Reflected optical chirp 1302 is
superimposed on LO reference chirp 1300 in order to illustrate the
frequency difference 1304 (i.e., the beat frequency) at a specific
point in time. The Fourier transformation is shown in FIG. 13B and
provides a power signal peak 1306 that is indicative of the
distance between LiDAR system 100 (technically, photodetector 120)
and the reflective object that provided reflected optical chirp
1302. In this manner, FPGA 126 is able to calculate the distance
for each pixel of data received from ADC 124.
[0140] In another example shown in FIG. 14A, one of the optical
chirps of the FMCW triangular waveform shown in FIG. 7 is
designated as optical chirp 1400. It should be understood that
optical chirp 1400 is an LO reference chirp generated by laser 102,
which is representative of the transmitted optical chirp. The
reflected optical chirp corresponding to LO reference chirp 1400 is
designated as optical chirp 1402. Reflected optical chirp 1402 is
superimposed on LO reference chirp 1400 in order to illustrate the
frequency difference 1404 in the up-chirp portion (i.e., the
up-chirp beat frequency) at a specific point in time and the
frequency difference 1406 in the down-chirp portion (i.e., the
down-chirp beat frequency) at a specific point in time. The Fourier
transformation is shown in FIG. 14B and provides a power signal
peak 1408 that is indicative of the distance between LiDAR system
100 (technically, photodetector 120) and the reflective object that
provided reflected optical chirp 1402, as well as a power signal
peak 1410 that is indicative of the relative velocity between LiDAR
system 100 (technically, photodetector 120) and the reflective
object that provided reflected optical chirp 1402. In this manner,
FPGA 126 is able to calculate both the distance and velocity for
each pixel of raw data received from ADC 124.
[0141] It can be appreciated that the use of an FMCW triangular
waveform is preferred over an FMCW sawtooth waveform insofar as
each pixel includes both distance data and velocity data. Of
course, distance data and velocity data can also be obtained in a
similar manner when using the FM pulsed waveform shown in FIG. 8 or
the sigmoid-shaped waveform shown in FIG. 9. The ability to
directly determine the velocity of an object based on an analysis
of a reflected optical chirp in relation to an LO reference chirp
provides an advantage over a conventional ToF system, in which two
frames of data are required to calculate velocity because the
object must be identified at least twice in order to calculate the
shift in position of the object.
[0142] After the Fourier transformation, FPGA 126 may optionally
compare the transformed data to a calibration file in order to
subtract any possible undesired noise spikes. FPGA 126 then further
processes the data for derivatization of the distance and/or
velocity of the reflective object that provided the reflected
optical chirp. Finally, the position (i.e., spatial coordinates) of
each reflective object is calculated based on the derived distance
of the reflective object and the MEMS mirror position at the time
of transmission of the optical chirp. FPGA 126 then provides each
pixel of position and/or velocity data to computer 128. It should
be noted that if a collector is used as the receiver (see FIG. 11),
any clutter would be seen as a separate power signal peak of the
Fourier transformation and, in this case, FPGA 126 would compare
adjacent pixels or possibly multiple frames to identify and filter
the clutter.
[0143] Referring back to FIG. 1, FPGA 126 outputs the position
and/or velocity data for each pixel to computer 128. Computer 128
may comprise a CPU, a GPU, or any other type of processing device,
and may process the position and/or velocity data received from
FPGA 126 in a variety of different ways. For example, in one
embodiment, computer 128 simply outputs the raw position and/or
velocity data received from FPGA 126 to an external control system
for object detection and identification and any other processing
required for a particular application. The data transfer may be
accomplished via Ethernet, Low-Voltage Differential Signaling
(LVDS), Universal Serial Bus (USB), Controller Area Network (CAN)
bus, or any other communications protocol. The external control
system may comprise, for example, a system that creates and/or
displays a 3D point cloud, fuses LiDAR data with data from other
sensors, identifies objects for avoidance, and the like.
[0144] In another embodiment, computer 128 processes the raw
position and/or velocity data received from FPGA 126 and outputs
processed data to an external control system as described above in
the form of a point cloud, i.e., a collection of data points each
of which provides point attributes for position and/or velocity. In
some embodiments, the point cloud may also provide other types of
point attributes, such as the overall intensity of the reflected
optical chirp. In addition, the point cloud may provide point
attributes that are derived from other attributes, such as
information on the color, composition or texture of a reflective
object derived from the intensity of the reflected optical chirp
(albeit this implementation would require the use of two or more
lasers). The external control system may then use this point cloud
data for object detection and identification and any other
processing required for a particular application.
[0145] In yet another embodiment, computer 128 processes the raw
position and velocity data received from FPGA 126 to provide the
object detection function and then outputs an object list to an
external control system as described above for object
identification and any other processing required for a particular
application. An example method for providing object detection is
shown in FIG. 15. In step 1500, computer 128 records the position
and velocity associated with a pixel or "point" from the scan. In
step 1502, computer 128 determines if the point is a possible
object. If NO, the method returns to step 1500. If YES, computer
128 determines at step 1504 if the point is near an existing
object, i.e., a previously identified object. If YES, computer 128
adds the point to the existing object at step 1506. If NO, computer
128 creates a new object that includes the point at step 1508.
Then, at step 1510, computer 128 determines if the object (either
the existing object or a newly created object) has N points. If NO,
the method returns to step 1500 and the process is repeated for an
additional point. If YES, computer 128 reports the object in step
1512. In this embodiment, the point-to-point latency (i.e., the
time required to identify each point) is in the range of 220 .mu.s
to 100 ms or a subrange within this broader range, e.g., 220 .mu.s
to 45 ms, 220 .mu.s to 16 ms, 500 .mu.s to 16 ms, 1 ms to 16 ms, or
220 .mu.s to 1 ms.
[0146] Because computer 128 records velocity data for each point
from the scan, only N points need to be identified in order to
detect an object. N is a designated threshold needed to make a
decision on object detection, e.g., N may be one (1), three (3),
eight (8) or some other integer based on Johnson's criteria or
equivalent. For example, if it takes 3 .mu.s to identify each point
and N=8 from Johnson's criteria, and if the processing latency is
about 200 .mu.s, then the total latency for object detection would
be about 224 .mu.s. Even if an object's point list is not complete
until N lines have been scanned in the direction perpendicular to
the fast scan axis (a worst-case scenario), the total latency for
object detection would be about 2 ms to 5 ms. Thus, the LiDAR
system of the present invention enables low-latency object
detection that is not achievable with conventional LiDAR systems.
In one embodiment, the object detection latency is in the range of
1 ms to 80 ms or a subrange within this broader range, e.g., 1 ms
to 40 ms, 2 ms to 40 ms, 2 ms to 16 ms, or 2 ms to 5 ms.
[0147] FIG. 16 shows an example data structure for a single frame
of data processed by computer 128, such as the frame shown in FIGS.
17A-17C. In this example, the frame was generated based on a scan
of 20 horizontal scan lines with 100 points per scan line (as shown
in FIG. 17A), wherein the horizontal angle is scanned at a faster
rate than the vertical angle (as shown by comparing FIG. 17B to
FIG. 17C). That is, all 100 points of a horizontal line are scanned
before incrementing vertically to the next scan line.
[0148] As can be seen in FIG. 16, computer 128 identified three
objects in the frame using the object detection method of FIG. 15,
namely, object A, object B and object C. Three points were
associated with object A (points 4, 5 and 6), four points were
associated with object B (points 11, 111, 211 and 311), and four
points were associated with object C (points 50, 51, 150 and 151).
In this embodiment, each point contains position and velocity data.
As shown in FIG. 17A, object A is positioned horizontally such that
points 4, 5 and 6 were scanned one after another; object B is
positioned vertically such that 100 points were processed between
consecutive points (e.g., 100 points were processed between point
11 and point 111); and object C is positioned both horizontally and
vertically such that points 50 and 51 were scanned one after
another and then 100 points were processed before points 151 and
152 were scanned one after another.
[0149] It should be noted that, in a conventional ToF LiDAR system,
two frames of data are required to calculate velocity because the
object must be identified at least twice in order to calculate the
shift in position of the object. At 80-200 ms per frame, the
latency is at least 160-400 ms, neglecting any additional object
detection and velocity calculation overhead. Thus, a conventional
ToF LiDAR system provides much longer latency for object detection
than the coherent LiDAR system of the present invention. One
skilled in the art will also understand that a coherent LiDAR
system only needs two frames of data to calculate acceleration,
whereas a ToF LiDAR system would need three frames of data to
calculate acceleration.
[0150] As mentioned above, the output rate (i.e., pixel rate) is
matched to the chirp repetition rate, wherein the output rate is
the number of data points processed per second by FPGA 126. In one
embodiment, FPGA 126 supports an output rate in the range of
50,000-500,000 points/second or a subrange within this broader
range, e.g., 50,000-350,000 points/second, 100,000-350,000
points/second, or 300,000-350,000 points/second. In a preferred
embodiment, the output rate is 333,000 points/second. Also, the
output density is in the range of 55,000-300,000 points/solid angle
and, similarly, the detector output density is in the range of
55,000-300,000 points/solid angle/detector. It should be noted that
these parameters are dependent on the linearity of the optical
chirps, which is in excess of 80% in this embodiment. It can be
appreciated that this high chirp linearity enables larger values
for the output rate, output density and detector output density, as
set forth above.
4. Control of Chirp Generation and Scanning Functions
[0151] The LiDAR system of the present invention includes one or
more control assemblies for controlling different aspects of the
system, such as the chirp generation and scanning functions. In one
embodiment, both of those control functions are provided by FPGA
126 and computer 128, as described below. Of course, separate
control assemblies for each of these control functions could be
used within the scope of the present invention.
[0152] In one embodiment, FPGA 126 internally generates the chirp
pattern to be used by laser 102 based on various chirp parameters.
These chirp parameters include, for example, chirp configuration
(e.g., triangular waveform, sawtooth waveform, etc.), chirp period
and chirp bandwidth. FPGA 126 may default to known chirp parameters
or may receive chirp parameters from computer 128. For example,
computer 128 may dynamically change the chirp parameters in
software to enable adaptive range resolution as desired for a
particular implementation. FPGA 126 provides the chirp parameters
to laser driver/modulator 104 to control the chirp generation
function.
[0153] FPGA 126 can also internally generate the scan pattern
(i.e., mirror positions and/or timing) to be used by a scanner,
such as MEMS mirror 116, based on various scan parameters. These
scan parameters include, for example, the field of view for a frame
(i.e., full scan or adaptive scan), spatial resolution (i.e.,
number of scan lines and pixels per scan line) for a frame, and
offset (zero or non-zero offset). FPGA 26 may default to known scan
parameters (e.g., full field of view, a set number of scan lines
and pixels per scan line, and zero offset) or may receive scan
parameters from computer 128. For example, computer 128 may
dynamically change the scan parameters in software to enable both
coarse scanning to locate objects of interest in the full field of
view and fine scanning on those objects of interest. For example,
the coarse scan may encompass 0.2.degree. to 2.0.degree. and the
fine scan may encompass 0.01.degree. to 0.2.degree.. Computer 128
may make this determination based on (a) a determination from a
previous frame (b) an external command or (c) a predefined scan
pattern evolution. It can be appreciated that the adaptive scan
parameters enable the system to "zoom in" on each individual object
of interest and identify the object and its position and velocity
early in the scanning process, as well as provide improved spatial
resolution at greater distances. FPGA 126 provides the scan
parameters to mirror driver 118 to control the scanning
function.
[0154] An example of a full scan pattern in comparison to an
adaptive scan pattern is shown in FIGS. 18A-18C (note that the full
scan pattern is the same as that shown in FIGS. 17A-17C). In each
illustration, the full scan frame is represented in regular solid
lines and the adaptive scan frame is represented in bolded solid
lines. As can be seen in FIG. 18A, the full scan frame is based on
a scan of 20 horizontal scan lines with 100 points per scan line
and the adaptive scan frame is based on a scan of 12 scan lines
with 60 points per scan line. FIGS. 18B and 18C show displacement
of the modified scan pattern in two axes such that the adaptive
scan frame is not centered at zero (i.e., a non-zero offset). Also,
the adaptive scan can have any amplitude up to the amplitude of the
full scan, as indicated by the dashed lines at the top and bottom
of FIG. 18B. As such, the adaptive scanning is more robust than
other types of adaptive scanning that have been used in
conventional LiDAR systems.
[0155] It should be understood that the field of view for a frame
can be any shape and is not limited to the rectangular shape shown
in FIGS. 18A-18C. For example, if a full scan pattern identifies an
object as being possibly a car, then the adaptive scan pattern may
follow the shape of a typical car. Of course, this is just one
example and it can be appreciated that the adaptive scan pattern
may include any subset of the points within a full field of view.
This will reduce the processing time required for object
identification.
[0156] It should also be understood that adaptive scanning may be
used beyond adaptive frame placement, e.g., it is possible to
implement adaptive chirp repetition rate, chirp bandwidth, chirp
configuration, dedication of additional processing power, and other
parameters. In one embodiment, LiDAR system 100 uses adaptive chirp
repetition rate and/or chirp bandwidth to dynamically change the
maximum range, maximum velocity, range resolution, and/or velocity
resolution using the equations set forth above. This approach
improves the ability to identify an object after it has been
initially detected. For example, assume that a wide field of view
scan pattern with certain default parameters, e.g., a maximum range
of two hundred (200) meters and a range resolution of four (4)
centimeters, is used to determine that an object is located forty
(40) meters away. In this case, the chirp pattern can be changed to
"zoom in" on this object in order to obtain a range resolution of
one (1) centimeter at the expense of reducing the maximum range to
fifty (50) meters. This adaptive scanning would enable the system
to map out the texture of the object in order to identify it as,
for example, a person vs. a tree. Of course, other examples of
adaptive scanning will be apparent to one skilled in the art.
B. Performance Characteristics of LiDAR System
[0157] The LiDAR system of the present invention provides
performance characteristics that are not achievable with
conventional ToF and FMCW LiDAR systems. Typical performance
characteristics are summarized in Table 1 below:
TABLE-US-00001 TABLE 1 Performance Characteristic Value/Range
Maximum Range up to 1 km (100% reflective target) up to 200 m (10%
reflective target) Minimum Range 1 mm to 1 m Range Resolution 1 mm
to 20 cm Maximum Velocity 200 mph to 450 mph Velocity Resolution
0.2 mph to 2 mph Number of Lasers/Detectors 1 Output Rate
50,000-500,000 points/second Output Density 55,000-300,000
points/solid angle Detector Output Density 55,000-300,000
points/solid angle/detector # of Returns 1-3 Field of View up to
120.degree. .times. 120.degree. Angular Resolution (H) 0.02.degree.
to 1.degree. Angular Resolution (V) 0.02.degree. to 1.degree. Size
10 cm.sup.3 to 500 cm.sup.3 Transmit Optical Power 10 mW to 3 W
Point-to-Point Latency 220 .mu.s to 100 ms
[0158] It should be noted that the LiDAR system of the present
invention is able to achieve the performance characteristics
described herein with the use of a single laser. In contrast, many
conventional LiDAR systems require the use of multiple lasers
(e.g., sixteen (16) lasers or more) to achieve comparable
performance characteristics or cannot achieve these performance
characteristics at all. The LiDAR system also offers other
advantages over many conventional LiDAR systems. For example, the
LiDAR system is compact, lightweight and has low power consumption
and, in addition, provides solid state reliability and
environmental tolerance. Also, operation of the LiDAR system does
not require GPS synchronization and, as such, can operate in a
broad range of environments (including tunnels)--even if there is
no GPS coverage. In addition, on-board data processing enables the
LiDAR system to directly (simultaneously) measure both distance and
velocity in real time, which provides significantly lower latency.
The LiDAR system also provides an adaptive scanning function, which
results in greater point density on objects of interest and
provides better spatial resolution at a distance, as well as other
advantages discussed above. The LiDAR system is also eye safe by
virtue of its laser wavelength in the range of 1400 nm to 4000 nm
(e.g., the 1530 nm to 1560 nm C-band) in combination with a lower
transmitted optical power.
[0159] Further, the LiDAR system is relatively immune from
interference caused by other LiDAR systems operating nearby. The
probability of interference between the LiDAR system and another
LiDAR system is approximately represented by the following
equation:
P .apprxeq. .lamda. 2 .times. LW c .times. .lamda. accuracy ( 9 )
##EQU00006##
where: [0160] P=probability of interference [0161] .lamda.=laser
wavelength in m [0162] LW=laser linewidth in Hz [0163] c=speed of
light (approximately 3.00.times.10.sup.8 m/sec) [0164]
.lamda..sub.accuracy=laser wavelength accuracy in m
[0165] In most cases, the probability of interference will be less
than 1.times.10.sup.-7. For example, assume that a 1550 nm laser
has a wavelength specified to an accuracy of 0.1 nm. For a laser
linewidth of 1 Hz, the probability that two 1550 nm lasers will be
emitting light at the exact same frequency so as to interfere with
each other can be determined using equation (9) as follows:
P .apprxeq. ( 1.55 .times. 10 - 6 ) 2 .times. ( 1 ) ( 3 .times. 10
8 ) .times. ( 1 .times. 10 - 10 ) .apprxeq. 8 .times. 10 - 11 ( 10
) ##EQU00007##
[0166] If the wavelength of the 1550 nm laser is only specified to
an accuracy of 1 nm, the probability of interference can be
determined using equation (9) as follows:
P .apprxeq. ( 1.55 .times. 10 - 6 ) 2 .times. ( 1 ) ( 3 .times. 10
8 ) .times. ( 1 .times. 10 - 9 ) .apprxeq. 8 .times. 10 - 11 ( 11 )
##EQU00008##
[0167] Of course, it should be understood that the probability of
interference would be orders of magnitude higher if the linewidth
of the laser were increased. For example, a laser linewidth of 400
kHz would result in a difference of almost 6 orders of magnitude.
For this reason, the LiDAR system of the present invention (with
its narrow linewidth laser) is relatively immune from interference.
Other advantages are described herein and/or will be apparent to
one skilled in the art.
C. Applications
[0168] The LiDAR system of the present invention can be used in a
variety of different systems for many applications. Because the
LiDAR system is compact, lightweight and has low power consumption,
it has wide application in systems where space, weight, and/or
power consumption is at a premium. Examples of such applications
include autonomous and semi-autonomous vehicles, remote piloted and
autonomous drones, portable devices, aerospace systems, robotic
systems, imaging systems, assistive devices, biomedical devices,
and virtual reality (VR) systems. Other applications include
agriculture, forestry, terrain mapping, meteorology, microscopy,
warehouse management, construction, and structural sensors. Of
course, other applications will be apparent to one skilled in the
art.
[0169] In one particular application, the LiDAR system is provided
in the form of one or more LiDAR sensors incorporated into an
Advanced Driver Assistance System (ADAS), in which each LiDAR
sensor provides highly accurate information relating to the
position and/or velocity of other vehicles and road obstacles. The
LiDAR sensors are compact and may be positioned in various
locations on the vehicle, such as near the head lights or back
lights, behind the front windshield, or any other location desired
for a particular implementation. In certain implementations, the
LiDAR sensors may be positioned to enable 360.degree. viewing. The
data from the LiDAR system may be combined with data from other
types of sensors, such as a camera, a radar sensor, sonar, or a
thermal imaging sensor.
[0170] In certain embodiments, the ADAS system uses the LiDAR
system information to notify the vehicle operator (i.e., the
driver) of the presence and/or proximity of other vehicles or road
obstacles. Such a notification can be an audible and/or tactile
alarm or warning that, for example, is transmitted through a
speaker of the vehicle's audio system and/or to an earpiece worn by
the vehicle operator. The notification can also be provided by a
visual display that can be seen by the vehicle operator. Such a
visual display can include a dedicated display for this purpose, a
display of the vehicle's navigation system, a display of the
vehicle's instrument cluster, a display integrated into a mirror of
the vehicle, and/or a "heads up" display reflected from the
interior of a vehicle window.
[0171] In other embodiments, the ADAS system uses the LiDAR system
information to provide assistance to a vehicle operator in the form
of automated vehicle responses. For example, the ADAS system can
control the vehicle's brake system, steering system, and/or engine
accelerator to change the vehicle's speed (accelerating,
decelerating, braking, gear ratio, etc.) and/or direction. The
automated vehicle responses can be provided following prompting of
the vehicle operator or in an autonomous fashion. In some
implementations, the automated vehicle responses can override
control of the vehicle provided by the vehicle operator, e.g., when
detected conditions meet certain criteria. Examples of such
criteria include determination that the vehicle operator is
incapacitated, that a detected condition can result in injury to
the vehicle operator and/or a detected individual, vehicle damage
or loss, or require action that is more rapid than can be provided
by the vehicle operator.
[0172] In yet other embodiments, the ADAS system uses the LiDAR
system information to operate the vehicle autonomously. This
autonomous functionality can be at the discretion of an onboard or
remote vehicle operator. If so, the vehicle can be directed by the
vehicle operator during part of its operation (heavy pedestrian
traffic, etc.) and operate autonomously under other conditions. In
other cases, the vehicle operates wholly autonomously. Such an
autonomous vehicle can be configured to carry passengers (i.e.,
persons not involved in operating the vehicle) or can be designed
to operate without a human presence.
D. General
[0173] The description of the invention set forth above provides
many embodiments of the inventive subject matter. Although each
embodiment represents a single combination of inventive elements,
the inventive subject matter is considered to include all possible
combinations of the disclosed elements. Thus, if one embodiment
comprises elements A, B, and C, and a second embodiment comprises
elements B and D, then the inventive subject matter is also
considered to include other remaining combinations of A, B, C, or
D, even if not explicitly disclosed.
[0174] The use of any and all examples or exemplary language (e.g.,
"such as") provided with respect to certain embodiments is intended
merely to better describe the invention and does not pose a
limitation on the scope of the invention. No language in the
description should be construed as indicating any non-claimed
element essential to the practice of the invention.
[0175] As used herein, the singular forms "a," "an" and "the"
include plural reference unless the context clearly dictates
otherwise. Also, the terms "optional" or "optionally" mean that the
subsequently described component or element may or may not be
present, and that the description includes embodiments in which
said component or element is present and embodiments where it is
not. In addition, unless the context dictates otherwise, the term
"coupled to" is intended to include both direct coupling (in which
two elements are directly connected to each other) and indirect
coupling (in which at least one additional element is located
between the two elements).
[0176] The recitation of ranges of values herein is merely intended
to serve as a shorthand method of referring individually to each
separate value falling within the range. Unless otherwise
indicated, each individual value is incorporated into the
description as if it were individually recited herein.
[0177] While the invention has been described and illustrated with
reference to many embodiments, it should be understood that various
modifications could be made to these embodiments without departing
from the scope of the invention. Therefore, the invention is not to
be limited to the specific configuration or methodology of these
embodiments, except insofar as such limitations are included in the
following claims.
* * * * *