Landmark data selection and unmapped obstacle detection in lidar-based navigation

Mathieu Joerger, Guillermo Duenas Arana, Matthew Spenko, Boris Pervan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

This research establishes new methods to quantify lidar-based navigation safety in highly automated vehicle (HAV) applications. Lidar navigation requires feature extraction (FE) and data association (DA). In prior work, an FE and DA risk prediction process was developed assuming that the set of extracted features matched the set of mapped landmarks. This paper addresses these limiting assumptions by first providing the means to select a subset of feature measurements (to be used in the estimator) while accounting for all existing landmarks in the surroundings. This is achieved by employing a probabilistic lower-bound on the mean innovation vector's norm. This measure of landmark separation is used in an analytical integrity risk bound that accounts for all possible association hypotheses. Then, a solution separation algorithm is employed to detect unmapped obstacles and wrong extractions. The integrity risk bound is modified to incorporate the risk of not detecting an unwanted obstacle (UO) when one might be present. Covariance analysis, direct simulation, and preliminary testing show that selecting fewer extracted features can significantly reduce integrity risk, but can also decrease landmark redundancy, thereby reducing UO detection capability.

Original languageEnglish (US)
Title of host publication30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017
PublisherInstitute of Navigation
Pages1886-1903
Number of pages18
ISBN (Electronic)9781510853317
StatePublished - Jan 1 2017
Event30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017 - Portland, United States
Duration: Sep 25 2017Sep 29 2017

Publication series

Name30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017
Volume3

Conference

Conference30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017
CountryUnited States
CityPortland
Period9/25/179/29/17

Fingerprint

Optical radar
Navigation
Feature extraction
Redundancy
Innovation
Testing

ASJC Scopus subject areas

  • Software
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Joerger, M., Arana, G. D., Spenko, M., & Pervan, B. (2017). Landmark data selection and unmapped obstacle detection in lidar-based navigation. In 30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017 (pp. 1886-1903). (30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017; Vol. 3). Institute of Navigation.

Landmark data selection and unmapped obstacle detection in lidar-based navigation. / Joerger, Mathieu; Arana, Guillermo Duenas; Spenko, Matthew; Pervan, Boris.

30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017. Institute of Navigation, 2017. p. 1886-1903 (30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017; Vol. 3).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Joerger, M, Arana, GD, Spenko, M & Pervan, B 2017, Landmark data selection and unmapped obstacle detection in lidar-based navigation. in 30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017. 30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017, vol. 3, Institute of Navigation, pp. 1886-1903, 30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017, Portland, United States, 9/25/17.
Joerger M, Arana GD, Spenko M, Pervan B. Landmark data selection and unmapped obstacle detection in lidar-based navigation. In 30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017. Institute of Navigation. 2017. p. 1886-1903. (30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017).
Joerger, Mathieu ; Arana, Guillermo Duenas ; Spenko, Matthew ; Pervan, Boris. / Landmark data selection and unmapped obstacle detection in lidar-based navigation. 30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017. Institute of Navigation, 2017. pp. 1886-1903 (30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017).
@inproceedings{21050873f4d6439a9465d66106b3e014,
title = "Landmark data selection and unmapped obstacle detection in lidar-based navigation",
abstract = "This research establishes new methods to quantify lidar-based navigation safety in highly automated vehicle (HAV) applications. Lidar navigation requires feature extraction (FE) and data association (DA). In prior work, an FE and DA risk prediction process was developed assuming that the set of extracted features matched the set of mapped landmarks. This paper addresses these limiting assumptions by first providing the means to select a subset of feature measurements (to be used in the estimator) while accounting for all existing landmarks in the surroundings. This is achieved by employing a probabilistic lower-bound on the mean innovation vector's norm. This measure of landmark separation is used in an analytical integrity risk bound that accounts for all possible association hypotheses. Then, a solution separation algorithm is employed to detect unmapped obstacles and wrong extractions. The integrity risk bound is modified to incorporate the risk of not detecting an unwanted obstacle (UO) when one might be present. Covariance analysis, direct simulation, and preliminary testing show that selecting fewer extracted features can significantly reduce integrity risk, but can also decrease landmark redundancy, thereby reducing UO detection capability.",
author = "Mathieu Joerger and Arana, {Guillermo Duenas} and Matthew Spenko and Boris Pervan",
year = "2017",
month = "1",
day = "1",
language = "English (US)",
series = "30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017",
publisher = "Institute of Navigation",
pages = "1886--1903",
booktitle = "30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017",
address = "United States",

}

TY - GEN

T1 - Landmark data selection and unmapped obstacle detection in lidar-based navigation

AU - Joerger, Mathieu

AU - Arana, Guillermo Duenas

AU - Spenko, Matthew

AU - Pervan, Boris

PY - 2017/1/1

Y1 - 2017/1/1

N2 - This research establishes new methods to quantify lidar-based navigation safety in highly automated vehicle (HAV) applications. Lidar navigation requires feature extraction (FE) and data association (DA). In prior work, an FE and DA risk prediction process was developed assuming that the set of extracted features matched the set of mapped landmarks. This paper addresses these limiting assumptions by first providing the means to select a subset of feature measurements (to be used in the estimator) while accounting for all existing landmarks in the surroundings. This is achieved by employing a probabilistic lower-bound on the mean innovation vector's norm. This measure of landmark separation is used in an analytical integrity risk bound that accounts for all possible association hypotheses. Then, a solution separation algorithm is employed to detect unmapped obstacles and wrong extractions. The integrity risk bound is modified to incorporate the risk of not detecting an unwanted obstacle (UO) when one might be present. Covariance analysis, direct simulation, and preliminary testing show that selecting fewer extracted features can significantly reduce integrity risk, but can also decrease landmark redundancy, thereby reducing UO detection capability.

AB - This research establishes new methods to quantify lidar-based navigation safety in highly automated vehicle (HAV) applications. Lidar navigation requires feature extraction (FE) and data association (DA). In prior work, an FE and DA risk prediction process was developed assuming that the set of extracted features matched the set of mapped landmarks. This paper addresses these limiting assumptions by first providing the means to select a subset of feature measurements (to be used in the estimator) while accounting for all existing landmarks in the surroundings. This is achieved by employing a probabilistic lower-bound on the mean innovation vector's norm. This measure of landmark separation is used in an analytical integrity risk bound that accounts for all possible association hypotheses. Then, a solution separation algorithm is employed to detect unmapped obstacles and wrong extractions. The integrity risk bound is modified to incorporate the risk of not detecting an unwanted obstacle (UO) when one might be present. Covariance analysis, direct simulation, and preliminary testing show that selecting fewer extracted features can significantly reduce integrity risk, but can also decrease landmark redundancy, thereby reducing UO detection capability.

UR - http://www.scopus.com/inward/record.url?scp=85047876774&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047876774&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85047876774

T3 - 30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017

SP - 1886

EP - 1903

BT - 30th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS 2017

PB - Institute of Navigation

ER -