IP Library › Granted Patent US 12,461,228
Granted Patent B1
US 12,461,228 · App. 17/898,061 · Granted Nov 4, 2025

Radar-inertial odometry for autonomous ground vehicles

Inventor: Andrew J. Kramer (Seattle, WA)
Assignee: Amazon Technologies, Inc.
G01S13/88B60W60/001G01S13/867B60W2420/408B60W2554/20B60W2554/40
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,461,228
App. No.
17/898,061
Granted
Nov 4, 2025
Kind
B1
Abstract

Autonomous ground vehicles that are outfitted with radar sensors and inertial measurement units accurately determine states of the autonomous ground vehicles, e.g., estimates of the vehicles' positions, orientations, or velocities or accelerations along or about one or more axes, based on data captured by the radar sensors and the inertial measurement units. Where objects are detected in radar scans, the objects are determined to be static (or fixed), or dynamic (or moving), and landmarks representing static objects are identified. Constraints on estimates of states may be calculated based on doppler effects, inertial measurement unit effects, or locations of landmarks, and the states may be determined as solutions to optimization problems based on the data and the calculated constraints. Odometry messages representing the determined states may be generated and stored or utilized for any purpose.

Claims (131)

1 . A method comprising:

receiving, by a plurality of radar sensors provided aboard an autonomous ground vehicle, a first plurality of radar scans, wherein each of the first plurality of radar scans is received in response to transmissions of electromagnetic energy by the plurality of radar sensors at a first time;

determining, by an inertial measurement unit provided aboard the autonomous ground vehicle, a first set of inertial measurement unit data, wherein the first set of inertial measurement unit data is determined at the first time;

calculating at least a first constraint based at least in part on the first plurality of radar scans and the first set of inertial measurement unit data;

determining a first state of the autonomous ground vehicle at the first time based at least in part on the first constraint, the first plurality of radar scans and the first set of inertial measurement unit data;

receiving, by the plurality of radar sensors, a second plurality of radar scans, wherein each of the second plurality of radar scans is received in response to transmissions of electromagnetic energy by the plurality of radar sensors at a second time;

determining, by the inertial measurement unit, a second set of inertial measurement unit data, wherein the second set of inertial measurement unit data is determined at the second time;

determining, by at least one processor unit provided aboard the autonomous ground vehicle, a position of a first object based at least in part on the second plurality of radar scans;

calculating at least a second constraint based at least in part on the second plurality of radar scans and the second set of inertial measurement unit data;

determining, by the at least one processor unit, a second state of the autonomous ground vehicle at the second time based at least in part on the second constraint, the second plurality of radar scans and the second set of inertial measurement unit data, wherein the second state of the autonomous ground vehicle comprises at least:

a position of the autonomous ground vehicle at the second time;

an angle of orientation of the autonomous ground vehicle with respect to at least a first principal axis of the autonomous ground vehicle at the second time;

a linear velocity of the autonomous ground vehicle at the second time with respect to at least the first principal axis; and

an angular velocity of the autonomous ground vehicle at the second time with respect to at least the first principal axis; and

at least one of:

storing information regarding the second state of the autonomous ground vehicle at the second time in at least one memory component of the autonomous ground vehicle; or

transmitting at least a portion of the information regarding the second state of the autonomous ground vehicle to at least one computer system over one or more networks.

2 . The method of claim 1 , wherein the plurality of radar sensors comprises:

a first radar sensor configured to transmit first electromagnetic energy in a first direction and to receive radar scans representing reflections of at least some of the first electromagnetic energy;

a second radar sensor configured to transmit first electromagnetic energy from a second direction and to receive radar scans representing reflections of at least some of the second electromagnetic energy; and

a third radar sensor configured to transmit third electromagnetic energy from a third direction and to receive radar scans representing reflections of at least some of the third electromagnetic energy.

3 . The method of claim 1 , wherein each of the plurality of radar sensors is a millimeter wave radar sensor.

4 . The method of claim 1 , wherein each of the plurality of radar sensors is configured to transmit electromagnetic energy at a frequency within a range of thirty to three hundred gigahertz or at a wavelength within a range of one to ten millimeters.

5 . The method of claim 1 , wherein at least one of the at least one-first constraint or the second constraint comprises a constraint on a velocity of the autonomous ground vehicle calculated according to an error function for each of the first plurality of radar scans based at least in part on:

rotations between the plurality of radar sensors and the inertial measurement unit;

translations between the plurality of radar sensors and the inertial measurement unit;

locations of each of a plurality of detections; and

doppler measurements for each of the plurality of detections.

6 . The method of claim 1 , wherein at least one of the at least one-first constraint or the second constraint comprises a constraint on at least one of:

a difference between a velocity of the autonomous ground vehicle at the first time and a velocity of the autonomous ground vehicle at a third time, wherein the third time precedes the first time, or

a difference between an acceleration of the autonomous ground vehicle at the first time and an acceleration of the autonomous ground vehicle at the third time.

7 . The method of claim 1 , wherein the at least one-first constraint or the second constraint comprises a constraint on a heading of the autonomous ground vehicle calculated based at least in part on a bearing to the first object at the first time and a bearing to the first object at a third time, wherein the third time precedes the first time.

8 . The method of claim 1 , wherein the autonomous ground vehicle is configured for a delivery of an item from a first location to a second location.

9 . The method of claim 1 , wherein the autonomous ground vehicle further comprises:

a frame;

at least one wheel;

a motor disposed in association with the frame, wherein the motor is configured to cause the at least one wheel to rotate at a speed within a predetermined speed range; and

at least one power module for powering the motor.

10 . A method comprising:

determining, by an inertial measurement unit provided aboard the autonomous ground vehicle, a first set of inertial measurement unit data, wherein the first set of inertial measurement unit data is determined at a first time;

receiving, by a first radar sensor, a first plurality of radar scans in response to transmissions of first electromagnetic energy by the first radar sensor in a first direction at approximately the first time;

receiving, by a second radar sensor, a second plurality of radar scans in response to transmissions of second electromagnetic energy by the second radar sensor in a second direction at approximately the first time;

receiving, by a third radar sensor, a third plurality of radar scans in response to transmissions of third electromagnetic energy by the third radar sensor in a third direction at approximately the first time;

generating a first data structure comprising the first plurality of radar scans, the second plurality of radar scans and the third plurality of radar scans;

calculating at least a first constraint based at least in part on the first data structure and the first set of inertial measurement unit data;

determining a first state of the autonomous ground vehicle at the first time based at least in part on the first constraint, the first data structure and the first set of inertial measurement unit data;

determining, by the inertial measurement unit, a second set of inertial measurement unit data, wherein the second set of inertial measurement unit data is determined at a second time;

receiving, by the first radar sensor, a fourth plurality of radar scans in response to transmissions of fourth electromagnetic energy by the first radar sensor in the first direction at approximately the second time;

receiving, by the second radar sensor, a fifth plurality of radar scans in response to transmissions of fifth electromagnetic energy by the second radar sensor in the second direction at approximately the second time;

receiving, by the third radar sensor, a sixth plurality of radar scans in response to transmissions of sixth electromagnetic energy by the third radar sensor in the third direction at approximately the first time;

determining, by at least one processor unit provided aboard the autonomous ground vehicle, a position of a first object based at least in part on the second plurality of radar scans;

calculating at least a second constraint based at least in part on the second plurality of radar scans and the second set of inertial measurement unit data; and

determining, by the at least one processor unit, a second state of the autonomous ground vehicle at the second time based at least in part on the second constraint, the second plurality of radar scans and the second set of inertial measurement unit data.

11 . The method of claim 10 , wherein the autonomous ground vehicle is configured for a delivery of an item from a first location to a second location, and

wherein the autonomous ground vehicle further comprises:

a frame;

at least one wheel;

a motor disposed in association with the frame, wherein the motor is configured to cause the at least one wheel to rotate at a speed within a predetermined speed range; and at least one power module for powering the motor.

12 . A method comprising:

receiving, by a plurality of radar sensors provided aboard an autonomous ground vehicle, a first plurality of radar scans, wherein each of the first plurality of radar scans is received in response to transmissions of electromagnetic energy by the plurality of radar sensors at a first time;

determining, by an inertial measurement unit provided aboard the autonomous ground vehicle, a first set of inertial measurement unit data, wherein the first set of inertial measurement unit data is determined at the first time;

calculating at least a first constraint based at least in part on the first plurality of radar scans and the first set of inertial measurement unit data;

determining a first state of the autonomous ground vehicle at the first time based at least in part on the first constraint, the first plurality of radar scans and the first set of inertial measurement unit data;

determining, by at least one processor unit provided aboard the autonomous ground vehicle, that at least a first radar scan of the first plurality of radar scans corresponds to the first object and that at least a second radar scan of the first plurality of radar scans corresponds to a second object;

determining, by the at least one processor unit, that the first object is stationary according to a random sample consensus method based at least in part on the first radar scan and the second radar scan;

determining, by the at least one processor unit, that the second object is in motion according to the random sample consensus method based at least in part on the first radar scan and the second radar scan;

receiving, by the plurality of radar sensors, a second plurality of radar scans, wherein each of the second plurality of radar scans is received in response to transmissions of electromagnetic energy by the plurality of radar sensors at a second time;

determining, by the inertial measurement unit, a second set of inertial measurement unit data, wherein the second set of inertial measurement unit data is determined at the second time;

determining, by the at least one processor unit, a position of a first object based at least in part on the second plurality of radar scans;

calculating at least a second constraint based at least in part on the second plurality of radar scans and the second set of inertial measurement unit data; and

determining, by the at least one processor unit, a second state of the autonomous ground vehicle at the second time based at least in part on the second constraint, the second plurality of radar scans and the second set of inertial measurement unit data.

13 . The method of claim 12 , wherein the first set of inertial measurement unit data comprises:

at least one of a first velocity or a first acceleration about a first principal axis of the autonomous ground vehicle at the first time;

at least one of a second velocity or a second acceleration about a second principal axis of the autonomous ground vehicle at the first time; and

at least one of a third velocity or a third acceleration about a third principal axis of the autonomous ground vehicle at the first time.

14 . The method of claim 12 , wherein the autonomous ground vehicle is configured for a delivery of an item from a first location to a second location, and

wherein the autonomous ground vehicle further comprises:

a frame;

at least one wheel;

a motor disposed in association with the frame, wherein the motor is configured to cause the at least one wheel to rotate at a speed within a predetermined speed range; and at least one power module for powering the motor.

15 . A method comprising:

receiving, by a plurality of radar sensors provided aboard an autonomous ground vehicle, a first plurality of radar scans, wherein each of the first plurality of radar scans is received in response to transmissions of electromagnetic energy by the plurality of radar sensors at a first time;

capturing, by an inertial measurement unit provided aboard the autonomous ground vehicle, first data regarding at least one of a velocity or an acceleration of the autonomous ground vehicle at the first time;

determining, by at least one processor unit, at least a first constraint on the autonomous vehicle based at least in part on the first plurality of radar scans and the first data;

determining, by the at least one processor unit, a first state of the autonomous ground vehicle at the first time based at least in part on the first data, the first constraint and the first plurality of radar scans;

receiving, by the plurality of radar sensors, a second plurality of radar scans, wherein each of the second plurality of radar scans is received in response to transmissions of electromagnetic energy by the plurality of radar sensors at a second time, and wherein at least a first radar scan of the second plurality of radar scans comprises a detection of a first object;

capturing, by the inertial measurement unit, second data regarding at least one of a velocity or an acceleration of the autonomous ground vehicle at the second time;

determining, by the at least one processor unit, at least a second constraint on the autonomous vehicle based at least in part on the second plurality of radar scans and the second data;

determining, based at least in part on the detection of the first object, a position of a first object; and

determining, by the at least one processor unit, a second state of the autonomous ground vehicle at the second time based at least in part on the second data, the second constraint and the position of the first object, wherein the second state of the autonomous ground vehicle at the second time comprises:

a position of the autonomous ground vehicle at the second time;

an angle of orientation of the autonomous ground vehicle with respect to at least a first principal axis of the autonomous ground vehicle at the second time;

a linear velocity of the autonomous ground vehicle at the second time with respect to at least the first principal axis; and

an angular velocity of the autonomous ground vehicle at the second time with respect to at least the first principal axis.

16 . The method of claim 15 , wherein receiving the first plurality of radar scans comprises:

receiving, by a first radar sensor of the plurality of radar sensors, radar scans in response to transmissions of electromagnetic energy by the first radar sensor at approximately the first time;

receiving, by a second radar sensor of the plurality of radar sensors, radar scans in response to transmissions of electromagnetic energy by the second radar sensor at approximately the first time;

receiving, by a third radar sensor of the plurality of radar sensors, radar scans in response to transmissions of electromagnetic energy by the third radar sensor at approximately the first time, wherein the first plurality of radar scans consists of the radar scans received by the first radar sensor in response to transmissions of electromagnetic energy by the first radar sensor at approximately the first time, the radar scans received by the second radar sensor in response to transmissions of electromagnetic energy by the second radar sensor at approximately the first time, and the radar scans received by the third radar sensor in response to transmissions of electromagnetic energy by the third radar sensor at approximately the first time; and

generating a data structure comprising the first plurality of radar scans,

wherein the second constraint is calculated based at least in part on the data structure.

17 . The method of claim 15 , wherein a second radar scan of the second plurality of radar scans comprises a detection of a second object, and

wherein the method further comprises:

determining, by the at least one processor unit, that at least the first radar scan corresponds to the first object and that at least the second radar scan corresponds to the second object;

determining, by the at least one processor unit, that the first object is stationary according to a random sample consensus method based at least in part on the first radar scan; and

determining, by the at least one processor unit, that the second object is in motion according to the random sample consensus method based at least in part on the second radar scan,

wherein the second state of the autonomous ground vehicle at the second time is not determined based on the second radar scan.

18 . The method of claim 15 , wherein the autonomous ground vehicle is configured for a delivery of an item from a first location to a second location, and

wherein the autonomous ground vehicle further comprises:

a frame;

at least one wheel;

a motor disposed in association with the frame, wherein the motor is configured to cause the at least one wheel to rotate at a speed within a predetermined speed range; and

at least one power module for powering the motor.

19 . A method comprising:

receiving, by a plurality of radar sensors provided aboard an autonomous ground vehicle, a first plurality of radar scans, wherein each of the first plurality of radar scans is received in response to transmissions of electromagnetic energy by the plurality of radar sensors at a first time;

determining, by an inertial measurement unit provided aboard the autonomous ground vehicle, a first set of inertial measurement unit data, wherein the first set of inertial measurement unit data is determined at the first time;

calculating at least a first constraint based at least in part on the first plurality of radar scans and the first set of inertial measurement unit data,

determining a first state of the autonomous ground vehicle at the first time based at least in part on the first constraint, the first plurality of radar scans and the first set of inertial measurement unit data;

receiving, by the plurality of radar sensors, a second plurality of radar scans, wherein each of the second plurality of radar scans is received in response to transmissions of electromagnetic energy by the plurality of radar sensors at a second time;

determining, by the inertial measurement unit, a second set of inertial measurement unit data, wherein the second set of inertial measurement unit data is determined at the second time;

determining, by at least one processor unit provided aboard the autonomous ground vehicle, a position of a first object based at least in part on the second plurality of radar scans;

calculating at least a second constraint based at least in part on the second plurality of radar scans and the second set of inertial measurement unit data, wherein the second constraint comprises one of:

a constraint on a velocity of the autonomous ground vehicle calculated according to an error function for each of the first plurality of radar scans based at least in part on rotations between the plurality of radar sensors and the inertial measurement unit, translations between the plurality of radar sensors and the inertial measurement unit, locations of each of a plurality of detections, and doppler measurements for each of the plurality of detections;

a constraint on a difference between a velocity of the autonomous ground vehicle at the first time and a velocity of the autonomous ground vehicle at a third time, wherein the third time precedes the first time;

a constraint on a difference between an acceleration of the autonomous ground vehicle at the first time and an acceleration of the autonomous ground vehicle at the third time; or

a constraint on a heading of the autonomous ground vehicle calculated based at least in part on a bearing to the first object at the first time and a bearing to the first object at the third time; and

determining, by the at least one processor unit, a second state of the autonomous ground vehicle at the third time based at least in part on the second constraint, the second plurality of radar scans and the second set of inertial measurement unit data.

20 . The method of claim 19 , wherein the autonomous ground vehicle is configured for a delivery of an item from a first location to a second location, and

wherein the autonomous ground vehicle further comprises:

a frame;

at least one wheel;

a motor disposed in association with the frame, wherein the motor is configured to cause the at least one wheel to rotate at a speed within a predetermined speed range; and at least one power module for powering the motor.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 29, 2022
From: KRAMER, ANDREW J.
To: AMAZON TECHNOLOGIES, INC.
Reel/Frame 060931/0332 →
References Cited (202)
US 4865248A · Barth · 1989 [cited by applicant]
US 4954962A · Evans, Jr. et al. · 1990 [cited by applicant]
US 5040116A · Evans, Jr. et al. · 1991 [cited by applicant]
US 5301114A · Mitchell · 1994 [cited by applicant]
US 5386462A · Schlamp · 1995 [cited by applicant]
US 5995898A · Tuttle · 1999 [cited by applicant]
US 6344796B1 · Ogilvie et al. · 2002 [cited by applicant]
US 6426699B1 · Porter · 2002 [cited by applicant]
US 6504530B1 · Wilson et al. · 2003 [cited by applicant]
US 6690997B2 · Rivalto · 2004 [cited by applicant]
US 6694217B2 · Bloom · 2004 [cited by applicant]
US 6919803B2 · Breed · 2005 [cited by applicant]
US 6961711B1 · Chee · 2005 [cited by applicant]
US 6970838B1 · Kamath et al. · 2005 [cited by applicant]
US 7129817B2 · Yamagishi · 2006 [cited by applicant]
US 7133743B2 · Tilles et al. · 2006 [cited by applicant]
US 7188513B2 · Wilson · 2007 [cited by applicant]
US 7337686B2 · Sagi-Dolev · 2008 [cited by applicant]
US 7337944B2 · Devar · 2008 [cited by applicant]
US 7339993B1 · Brooks et al. · 2008 [cited by applicant]
US 7409291B2 · Pasolini et al. · 2008 [cited by applicant]
US 8321128B2 · Park · 2012 [cited by applicant]
US 8532862B2 · Neff · 2013 [cited by applicant]
US 8571743B1 · Cullinane · 2013 [cited by applicant]
US 8688306B1 · Nemec et al. · 2014 [cited by applicant]
US 8738213B1 · Szybalski et al. · 2014 [cited by applicant]
US 8855847B2 · Uehara · 2014 [cited by applicant]
US 9079587B1 · Rupp et al. · 2015 [cited by applicant]
US 9120484B1 · Ferguson et al. · 2015 [cited by applicant]
US 9147296B2 · Ricci · 2015 [cited by applicant]
US 9201421B1 · Fairfield et al. · 2015 [cited by applicant]
US 9294474B1 · Alikhani · 2016 [cited by applicant]
US 9381916B1 · Zhu et al. · 2016 [cited by applicant]
US 9440647B1 · Sucan et al. · 2016 [cited by applicant]
US 9558664B1 · Gaebler et al. · 2017 [cited by applicant]
US 9718564B1 · Beckman et al. · 2017 [cited by applicant]
US 11353578B2 · Wang · 2022 [cited by examiner]
US 20010045449A1 · Shannon · 2001 [cited by applicant]
US 20020016726A1 · Ross · 2002 [cited by applicant]
US 20020087375A1 · Griffin et al. · 2002 [cited by applicant]
US 20020111914A1 · Terada et al. · 2002 [cited by applicant]
US 20020116289A1 · Yang · 2002 [cited by applicant]
US 20020123930A1 · Boyd et al. · 2002 [cited by applicant]
US 20030040980A1 · Nakajima et al. · 2003 [cited by applicant]
US 20030097047A1 · Woltermann et al. · 2003 [cited by applicant]
US 20030141411A1 · Pandya et al. · 2003 [cited by applicant]
US 20030162523A1 · Kapolka et al. · 2003 [cited by applicant]
US 20040172167A1 · Pasolini et al. · 2004 [cited by applicant]
US 20060136237A1 · Spiegel et al. · 2006 [cited by applicant]
US 20060178140A1 · Smith et al. · 2006 [cited by applicant]
US 20070005609A1 · Breed · 2007 [cited by applicant]
US 20070016496A1 · Bar et al. · 2007 [cited by applicant]
US 20070073552A1 · Hileman · 2007 [cited by applicant]
US 20070150375A1 · Yang · 2007 [cited by applicant]
US 20070170237A1 · Neff · 2007 [cited by applicant]
US 20070244614A1 · Nathanson · 2007 [cited by applicant]
US 20070293978A1 · Wurman et al. · 2007 [cited by applicant]
US 20080059068A1 · Strelow et al. · 2008 [cited by applicant]
US 20080150679A1 · Bloomfield · 2008 [cited by applicant]
US 20080154659A1 · Bettes et al. · 2008 [cited by applicant]
US 20080161986A1 · Breed · 2008 [cited by applicant]
US 20080167817A1 · Hessler et al. · 2008 [cited by applicant]
US 20080301009A1 · Plaster et al. · 2008 [cited by applicant]
US 20090005985A1 · Basnayake · 2009 [cited by applicant]
US 20090062974A1 · Tamamoto et al. · 2009 [cited by applicant]
US 20090063166A1 · Palmer · 2009 [cited by applicant]
US 20090106124A1 · Yang · 2009 [cited by applicant]
US 20090149985A1 · Chirnomas · 2009 [cited by applicant]
US 20090236470A1 · Goossen et al. · 2009 [cited by applicant]
US 20090247186A1 · Ji et al. · 2009 [cited by applicant]
US 20090299903A1 · Hung et al. · 2009 [cited by applicant]
US 20090314883A1 · Arlton et al. · 2009 [cited by applicant]
US 20100012573A1 · Dendel et al. · 2010 [cited by applicant]
US 20100057360A1 · Ohkubo · 2010 [cited by applicant]
US 20100121573A1 · Imafuku et al. · 2010 [cited by applicant]
US 20100125403A1 · Clark et al. · 2010 [cited by applicant]
US 20100157061A1 · Katsman et al. · 2010 [cited by applicant]
US 20100241355A1 · Park · 2010 [cited by applicant]
US 20100250134A1 · Bornstein et al. · 2010 [cited by applicant]
US 20100312461A1 · Haynie et al. · 2010 [cited by applicant]
US 20110035149A1 · McAndrew et al. · 2011 [cited by applicant]
US 20110066377A1 · Takaoka · 2011 [cited by applicant]
US 20110071759A1 · Pande et al. · 2011 [cited by applicant]
US 20110112764A1 · Trum · 2011 [cited by applicant]
US 20110125403A1 · Smith · 2011 [cited by applicant]
US 20110208496A1 · Bando et al. · 2011 [cited by applicant]
US 20110264311A1 · Lee et al. · 2011 [cited by applicant]
US 20120039694A1 · Suzanne · 2012 [cited by applicant]
US 20120083960A1 · Zhu et al. · 2012 [cited by applicant]
US 20120109419A1 · Mercado · 2012 [cited by applicant]
US 20120219397A1 · Baker · 2012 [cited by applicant]
US 20130073477A1 · Grinberg · 2013 [cited by applicant]
US 20130081245A1 · Vavrina et al. · 2013 [cited by applicant]
US 20130148123A1 · Hayashi · 2013 [cited by applicant]
US 20130214925A1 · Weiss · 2013 [cited by applicant]
US 20130218799A1 · Lehmann et al. · 2013 [cited by applicant]
US 20130231824A1 · Wilson et al. · 2013 [cited by applicant]
US 20130238170A1 · Klinger · 2013 [cited by applicant]
US 20130261792A1 · Gupta et al. · 2013 [cited by applicant]
US 20130262251A1 · Wan et al. · 2013 [cited by applicant]
US 20130262252A1 · Lakshman et al. · 2013 [cited by applicant]
US 20130262276A1 · Wan et al. · 2013 [cited by applicant]
US 20130262336A1 · Wan et al. · 2013 [cited by applicant]
US 20130264381A1 · Kim et al. · 2013 [cited by applicant]
US 20130297099A1 · Rovik · 2013 [cited by applicant]
US 20140022055A1 · Levien et al. · 2014 [cited by applicant]
US 20140030444A1 · Swaminathan et al. · 2014 [cited by applicant]
US 20140032034A1 · Raptopoulos et al. · 2014 [cited by applicant]
US 20140052661A1 · Shakes et al. · 2014 [cited by applicant]
US 20140081507A1 · Urmson et al. · 2014 [cited by applicant]
US 20140095009A1 · Oshima et al. · 2014 [cited by applicant]
US 20140136282A1 · Fedele · 2014 [cited by applicant]
US 20140136414A1 · Abhyanker · 2014 [cited by applicant]
US 20140156133A1 · Cullinane et al. · 2014 [cited by applicant]
US 20140172290A1 · Prokhorov et al. · 2014 [cited by applicant]
US 20140188376A1 · Gordon · 2014 [cited by applicant]
US 20140200737A1 · Lortz et al. · 2014 [cited by applicant]
US 20140244678A1 · Zamer et al. · 2014 [cited by applicant]
US 20140254896A1 · Zhou et al. · 2014 [cited by applicant]
US 20140278052A1 · Slavin et al. · 2014 [cited by applicant]
US 20150006005A1 · Yu et al. · 2015 [cited by applicant]
US 20150069968A1 · Pounds · 2015 [cited by applicant]
US 20150102154A1 · Duncan et al. · 2015 [cited by applicant]
US 20150120602A1 · Huffman et al. · 2015 [cited by applicant]
US 20150129716A1 · Yoffe · 2015 [cited by applicant]
US 20150153175A1 · Skaaksrud · 2015 [cited by applicant]
US 20150153735A1 · Clarke et al. · 2015 [cited by applicant]
US 20150158599A1 · Sisko · 2015 [cited by applicant]
US 20150175276A1 · Koster · 2015 [cited by applicant]
US 20150183528A1 · Walsh et al. · 2015 [cited by applicant]
US 20150185034A1 · Abhyanker · 2015 [cited by applicant]
US 20150193005A1 · Censo et al. · 2015 [cited by applicant]
US 20150227882A1 · Bhatt · 2015 [cited by applicant]
US 20150241878A1 · Crombez et al. · 2015 [cited by applicant]
US 20150246727A1 · Masticola et al. · 2015 [cited by applicant]
US 20150259078A1 · Filipovic et al. · 2015 [cited by applicant]
US 20150268665A1 · Ludwick et al. · 2015 [cited by applicant]
US 20150291032A1 · Kim et al. · 2015 [cited by applicant]
US 20150317597A1 · Shucker et al. · 2015 [cited by applicant]
US 20150332206A1 · Trew et al. · 2015 [cited by applicant]
US 20150345966A1 · Meuleau · 2015 [cited by applicant]
US 20150345971A1 · Meuleau et al. · 2015 [cited by applicant]
US 20150346727A1 · Ramanujam · 2015 [cited by applicant]
US 20150348112A1 · Ramanujam · 2015 [cited by applicant]
US 20150348335A1 · Ramanujam · 2015 [cited by applicant]
US 20150363986A1 · Hoyos et al. · 2015 [cited by applicant]
US 20150367850A1 · Clarke et al. · 2015 [cited by applicant]
US 20150370251A1 · Siegel et al. · 2015 [cited by applicant]
US 20160009413A1 · Lee et al. · 2016 [cited by applicant]
US 20160025973A1 · Guttag et al. · 2016 [cited by applicant]
US 20160033966A1 · Farris et al. · 2016 [cited by applicant]
US 20160068156A1 · Horii · 2016 [cited by applicant]
US 20160085565A1 · Arcese et al. · 2016 [cited by applicant]
US 20160093212A1 · Barfield et al. · 2016 [cited by applicant]
US 20160104099A1 · Villamar · 2016 [cited by applicant]
US 20160114488A1 · Medina et al. · 2016 [cited by applicant]
US 20160144734A1 · Wang et al. · 2016 [cited by applicant]
US 20160144959A1 · Meffert · 2016 [cited by applicant]
US 20160144982A1 · Sugumaran · 2016 [cited by applicant]
US 20160207627A1 · Hoareau et al. · 2016 [cited by applicant]
US 20160209220A1 · Laetz · 2016 [cited by applicant]
US 20160209845A1 · Kojo et al. · 2016 [cited by applicant]
US 20160216711A1 · Srivastava et al. · 2016 [cited by applicant]
US 20160247404A1 · Srivastava et al. · 2016 [cited by applicant]
US 20160257401A1 · Buchmueller et al. · 2016 [cited by applicant]
US 20160266578A1 · Douglas et al. · 2016 [cited by applicant]
US 20160288796A1 · Yuan · 2016 [cited by applicant]
US 20160299233A1 · Levien et al. · 2016 [cited by applicant]
US 20160300400A1 · Namikawa · 2016 [cited by applicant]
US 20160301698A1 · Katara et al. · 2016 [cited by applicant]
US 20160307449A1 · Gordon et al. · 2016 [cited by applicant]
US 20160334229A1 · Ross et al. · 2016 [cited by applicant]
US 20160364989A1 · Speasl et al. · 2016 [cited by applicant]
US 20160370193A1 · Maischberger et al. · 2016 [cited by applicant]
US 20170010613A1 · Fukumoto · 2017 [cited by applicant]
US 20170032315A1 · Gupta et al. · 2017 [cited by applicant]
US 20170096222A1 · Spinelli et al. · 2017 [cited by applicant]
US 20170098378A1 · Soundararajan et al. · 2017 [cited by applicant]
US 20170101017A1 · Streett · 2017 [cited by applicant]
US 20170123421A1 · Kentley · 2017 [cited by examiner]
US 20170164319A1 · Skaaksrud et al. · 2017 [cited by applicant]
US 20170167881A1 · Rander et al. · 2017 [cited by applicant]
US 20170301232A1 · Xu et al. · 2017 [cited by applicant]
US 20180134286A1 · Yi et al. · 2018 [cited by applicant]
US 20180307223A1 · Peeters et al. · 2018 [cited by applicant]
US 20190042859A1 · Schubert et al. · 2019 [cited by applicant]
US 20200249698A1 · Lu et al. · 2020 [cited by applicant]
US 20210209785A1 · Unnikrishnan · 2021 [cited by examiner]
EP 2228779A2 · 2010 [cited by applicant]
WO 2015134376A1 · 2015 [cited by applicant]
Arandjelovic, R. et al. “NetVLAD: CNN Architecture for Weakly Supervised Place Recognition.” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016, 17 pages. [cited by applicant]
Barnes, Dan, and Ingmar Posner. “Under the Radar: Learning to Predict Robust Keypoints for Odometry Estimation and Metric Localisation in Radar.” 2020 IEEE International Conference on Robotics and Automation (ICRA). IEE… [cited by applicant]
Cen, Sarah H., and Paul Newman. “Precise Ego-Motion Estimation with Millimeter-Wave Radar Under Diverse and Challenging Conditions.” 2018 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2018. URL:… [cited by applicant]
DHL Trend Research, “Self-Driving Vehicles in Logistics,” Dec. 2014, Markus Kuckelhaus et al. (downloaded from http://www.dhl.com/content/dam/downloads/g0/about_us/logistics_insights/dhl_self_driving_vehicles.pdf with a… [cited by applicant]
DHL Trend Research, “Unmanned Aerial Vehicles in Logistics: a DHL perspective on implications and use cases for the logistics industry,” 2014, Markus Kückelhaus et al., URL: http://www.dhl.com/content/dam/downloads/g0/a… [cited by applicant]
Go, Yeong-Ju and Jong-Soo Choi. “An Acoustic Source Localization Method Using a Drone-Mounted Phased Microphone Array.” Drones 5.3 (2021): 75, 18 pages, URL: https://www.mdpi.com/2504-446X/5/3/75. [cited by applicant]
Marcus Wohlsen, “The Next Big Thing You Missed: Amazon's Delivery Drones Could Work—They Just Need Trucks,” Wired: Business, Jun. 10, 2014, URL: https://www.wired.com/2014/06/the-next-big-thing-you-missed-delivery-drone… [cited by applicant]
Mike Murphy, “Google wants to deliver packages from self-driving trucks,” published Feb. 9, 2016, URL: https://qz.com/613277/google-wants-to-deliver-packages-from-self-driving-trucks/, 4 pages. [cited by applicant]
Sandoval, “Google patents secure rolling box to receive packages from drones,” Geekwire.com, Jan. 27, 2016, URL: http://www.geekwire.com/2016/google-pondering-drone-delivery-even-about-boxes-it-flies-to-front-doors/, 11… [cited by applicant]
URL: https://web.archive.org/web/20160804001046/https://www.starship.xyz/, download date: Aug. 4, 2016, 21 pages. [cited by applicant]
Wang, Lin and Andrea Cavallaro. “Acoustic Sensing from a Multi-Rotor Drone.” IEEE Sensors Journal 18.11 (2018): 4570-4582, URL: https://www.researchgate.net/publication/324468964_Acoustic_Sensing_From_a_Multi-Rotor_Dron… [cited by applicant]
Yoon, David J., et al. “Unsupervised Learning of Lidar Features for Use in a Probabilistic Trajectory Estimator.” IEEE Robotics and Automation Letters 6.2 (2021): 2130-2138. URL: https://arxiv.org/pdf/2102.11261.pdf. [cited by applicant]
Cited By (1)
US 12,681,495