RADARODO: Ego-Motion Estimation from Doppler and Spatial Data in RADAR Images

Chris D. Monaco; Sean N. Brennan

doi:10.1109/TIV.2020.2973536

RADARODO: Ego-Motion Estimation from Doppler and Spatial Data in RADAR Images

Chris D. Monaco, Sean N. Brennan

Research output: Contribution to journal › Article › peer-review

17 Scopus citations

Abstract

Accurate and robust ego-motion estimation is critical for aiding autonomous vehicle localization in environments devoid of map landmarks or accurate GNSS measurements. RADAR measurements provide unique capabilities for ego-motion estimation while also posing unique challenges. This paper presents the novel real-time RADAR-based ego-motion estimation algorithm, RADARODO (RADAR Odometry). RADARODO is differentiated from similar techniques by two novel attributes. First, it decouples sensor translational and rotational motion by estimating them from Doppler and spatial data, respectively. Secondly, it directly analyzes RADAR images to estimate rotational motion via correspondence-free non-linear optimization. Unlike other methods, RADARODO estimates ego-motion along with its predicted uncertainty from a single RADAR sensor without depending on a lever-arm offset or the zero side-slip assumption. The real-world results demonstrate accurate ego-motion estimates that are particularly suited for integration within a 'tightly-coupled' sensor fusion framework.

Original language	English (US)
Article number	8995552
Pages (from-to)	475-484
Number of pages	10
Journal	IEEE Transactions on Intelligent Vehicles
Volume	5
Issue number	3
DOIs	https://doi.org/10.1109/TIV.2020.2973536
State	Published - Sep 2020

All Science Journal Classification (ASJC) codes

Automotive Engineering
Control and Optimization
Artificial Intelligence

Access to Document

10.1109/TIV.2020.2973536

Cite this

@article{38e4072bfa8649a6836c3de4f66d989c,

title = "RADARODO: Ego-Motion Estimation from Doppler and Spatial Data in RADAR Images",

abstract = "Accurate and robust ego-motion estimation is critical for aiding autonomous vehicle localization in environments devoid of map landmarks or accurate GNSS measurements. RADAR measurements provide unique capabilities for ego-motion estimation while also posing unique challenges. This paper presents the novel real-time RADAR-based ego-motion estimation algorithm, RADARODO (RADAR Odometry). RADARODO is differentiated from similar techniques by two novel attributes. First, it decouples sensor translational and rotational motion by estimating them from Doppler and spatial data, respectively. Secondly, it directly analyzes RADAR images to estimate rotational motion via correspondence-free non-linear optimization. Unlike other methods, RADARODO estimates ego-motion along with its predicted uncertainty from a single RADAR sensor without depending on a lever-arm offset or the zero side-slip assumption. The real-world results demonstrate accurate ego-motion estimates that are particularly suited for integration within a 'tightly-coupled' sensor fusion framework. ",

author = "Monaco, {Chris D.} and Brennan, {Sean N.}",

note = "Funding Information: Manuscript received May 17, 2019; revised September 11, 2019 and December 6, 2019; accepted January 30, 2020. Date of publication February 12, 2020; date of current version August 27, 2020. This work was partially supported by the Penn State Applied Research Laboratory. (Corresponding author: Chris Monaco.) The authors are with the Department of Mechanical Engineering, The Pennsylvania State University, State College, PA 16801 USA (e-mail: chris.d. monaco@psu.edu; sbrennan@psu.edu). Publisher Copyright: {\textcopyright} 2016 IEEE.",

year = "2020",

month = sep,

doi = "10.1109/TIV.2020.2973536",

language = "English (US)",

volume = "5",

pages = "475--484",

journal = "IEEE Transactions on Intelligent Vehicles",

issn = "2379-8858",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "3",

}

TY - JOUR

T1 - RADARODO

T2 - Ego-Motion Estimation from Doppler and Spatial Data in RADAR Images

AU - Monaco, Chris D.

AU - Brennan, Sean N.

N1 - Funding Information: Manuscript received May 17, 2019; revised September 11, 2019 and December 6, 2019; accepted January 30, 2020. Date of publication February 12, 2020; date of current version August 27, 2020. This work was partially supported by the Penn State Applied Research Laboratory. (Corresponding author: Chris Monaco.) The authors are with the Department of Mechanical Engineering, The Pennsylvania State University, State College, PA 16801 USA (e-mail: chris.d. monaco@psu.edu; sbrennan@psu.edu). Publisher Copyright: © 2016 IEEE.

PY - 2020/9

Y1 - 2020/9

N2 - Accurate and robust ego-motion estimation is critical for aiding autonomous vehicle localization in environments devoid of map landmarks or accurate GNSS measurements. RADAR measurements provide unique capabilities for ego-motion estimation while also posing unique challenges. This paper presents the novel real-time RADAR-based ego-motion estimation algorithm, RADARODO (RADAR Odometry). RADARODO is differentiated from similar techniques by two novel attributes. First, it decouples sensor translational and rotational motion by estimating them from Doppler and spatial data, respectively. Secondly, it directly analyzes RADAR images to estimate rotational motion via correspondence-free non-linear optimization. Unlike other methods, RADARODO estimates ego-motion along with its predicted uncertainty from a single RADAR sensor without depending on a lever-arm offset or the zero side-slip assumption. The real-world results demonstrate accurate ego-motion estimates that are particularly suited for integration within a 'tightly-coupled' sensor fusion framework.

AB - Accurate and robust ego-motion estimation is critical for aiding autonomous vehicle localization in environments devoid of map landmarks or accurate GNSS measurements. RADAR measurements provide unique capabilities for ego-motion estimation while also posing unique challenges. This paper presents the novel real-time RADAR-based ego-motion estimation algorithm, RADARODO (RADAR Odometry). RADARODO is differentiated from similar techniques by two novel attributes. First, it decouples sensor translational and rotational motion by estimating them from Doppler and spatial data, respectively. Secondly, it directly analyzes RADAR images to estimate rotational motion via correspondence-free non-linear optimization. Unlike other methods, RADARODO estimates ego-motion along with its predicted uncertainty from a single RADAR sensor without depending on a lever-arm offset or the zero side-slip assumption. The real-world results demonstrate accurate ego-motion estimates that are particularly suited for integration within a 'tightly-coupled' sensor fusion framework.

UR - http://www.scopus.com/inward/record.url?scp=85085131081&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85085131081&partnerID=8YFLogxK

U2 - 10.1109/TIV.2020.2973536

DO - 10.1109/TIV.2020.2973536

M3 - Article

AN - SCOPUS:85085131081

SN - 2379-8858

VL - 5

SP - 475

EP - 484

JO - IEEE Transactions on Intelligent Vehicles

JF - IEEE Transactions on Intelligent Vehicles

IS - 3

M1 - 8995552

ER -

RADARODO: Ego-Motion Estimation from Doppler and Spatial Data in RADAR Images

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this