Injecting Planning-Awareness into Prediction and Detection Evaluation

Detecting other agents and forecasting their behavior is an integral part of the modern robotic autonomy stack, especially in safety-critical scenarios entailing human-robot interaction such as autonomous driving. Due to the importance of these components, there has been a significant amount of inte...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ivanovic, Boris, Pavone, Marco
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Ivanovic, Boris
Pavone, Marco
description Detecting other agents and forecasting their behavior is an integral part of the modern robotic autonomy stack, especially in safety-critical scenarios entailing human-robot interaction such as autonomous driving. Due to the importance of these components, there has been a significant amount of interest and research in perception and trajectory forecasting, resulting in a wide variety of approaches. Common to most works, however, is the use of the same few accuracy-based evaluation metrics, e.g., intersection-over-union, displacement error, log-likelihood, etc. While these metrics are informative, they are task-agnostic and outputs that are evaluated as equal can lead to vastly different outcomes in downstream planning and decision making. In this work, we take a step back and critically assess current evaluation metrics, proposing task-aware metrics as a better measure of performance in systems where they are deployed. Experiments on an illustrative simulation as well as real-world autonomous driving data validate that our proposed task-aware metrics are able to account for outcome asymmetry and provide a better estimate of a model's closed-loop performance.
doi_str_mv 10.48550/arxiv.2110.03270
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2110_03270</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2110_03270</sourcerecordid><originalsourceid>FETCH-LOGICAL-a670-9d972e25532130fddfead2e880c1099913bd0dcfcd26c954e3d43dd3077a1af03</originalsourceid><addsrcrecordid>eNotj01OwzAQhb1hgQoHYIUvkDK24zpeViVApUrtovto8IyRUXCREwrcnrRl9X709KRPiDsF87qxFh6w_KTjXKupAKMdXIt2nd85jCm_yV2POU-mWn5j4czDIFMeD3JXmNI0OWSJmeQjj3xJ7RH7LzzZG3EVsR_49l9nYv_U7lcv1Wb7vF4tNxUuHFSevNOsrTVaGYhEkZE0Nw0EBd57ZV4JKMRAehG8rdlQbYgMOIcKI5iZuL_cnjm6z5I-sPx2J57uzGP-AOR9RiA</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Injecting Planning-Awareness into Prediction and Detection Evaluation</title><source>arXiv.org</source><creator>Ivanovic, Boris ; Pavone, Marco</creator><creatorcontrib>Ivanovic, Boris ; Pavone, Marco</creatorcontrib><description>Detecting other agents and forecasting their behavior is an integral part of the modern robotic autonomy stack, especially in safety-critical scenarios entailing human-robot interaction such as autonomous driving. Due to the importance of these components, there has been a significant amount of interest and research in perception and trajectory forecasting, resulting in a wide variety of approaches. Common to most works, however, is the use of the same few accuracy-based evaluation metrics, e.g., intersection-over-union, displacement error, log-likelihood, etc. While these metrics are informative, they are task-agnostic and outputs that are evaluated as equal can lead to vastly different outcomes in downstream planning and decision making. In this work, we take a step back and critically assess current evaluation metrics, proposing task-aware metrics as a better measure of performance in systems where they are deployed. Experiments on an illustrative simulation as well as real-world autonomous driving data validate that our proposed task-aware metrics are able to account for outcome asymmetry and provide a better estimate of a model's closed-loop performance.</description><identifier>DOI: 10.48550/arxiv.2110.03270</identifier><language>eng</language><subject>Computer Science - Computer Vision and Pattern Recognition ; Computer Science - Learning ; Computer Science - Robotics ; Computer Science - Systems and Control</subject><creationdate>2021-10</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2110.03270$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2110.03270$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Ivanovic, Boris</creatorcontrib><creatorcontrib>Pavone, Marco</creatorcontrib><title>Injecting Planning-Awareness into Prediction and Detection Evaluation</title><description>Detecting other agents and forecasting their behavior is an integral part of the modern robotic autonomy stack, especially in safety-critical scenarios entailing human-robot interaction such as autonomous driving. Due to the importance of these components, there has been a significant amount of interest and research in perception and trajectory forecasting, resulting in a wide variety of approaches. Common to most works, however, is the use of the same few accuracy-based evaluation metrics, e.g., intersection-over-union, displacement error, log-likelihood, etc. While these metrics are informative, they are task-agnostic and outputs that are evaluated as equal can lead to vastly different outcomes in downstream planning and decision making. In this work, we take a step back and critically assess current evaluation metrics, proposing task-aware metrics as a better measure of performance in systems where they are deployed. Experiments on an illustrative simulation as well as real-world autonomous driving data validate that our proposed task-aware metrics are able to account for outcome asymmetry and provide a better estimate of a model's closed-loop performance.</description><subject>Computer Science - Computer Vision and Pattern Recognition</subject><subject>Computer Science - Learning</subject><subject>Computer Science - Robotics</subject><subject>Computer Science - Systems and Control</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj01OwzAQhb1hgQoHYIUvkDK24zpeViVApUrtovto8IyRUXCREwrcnrRl9X709KRPiDsF87qxFh6w_KTjXKupAKMdXIt2nd85jCm_yV2POU-mWn5j4czDIFMeD3JXmNI0OWSJmeQjj3xJ7RH7LzzZG3EVsR_49l9nYv_U7lcv1Wb7vF4tNxUuHFSevNOsrTVaGYhEkZE0Nw0EBd57ZV4JKMRAehG8rdlQbYgMOIcKI5iZuL_cnjm6z5I-sPx2J57uzGP-AOR9RiA</recordid><startdate>20211007</startdate><enddate>20211007</enddate><creator>Ivanovic, Boris</creator><creator>Pavone, Marco</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20211007</creationdate><title>Injecting Planning-Awareness into Prediction and Detection Evaluation</title><author>Ivanovic, Boris ; Pavone, Marco</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a670-9d972e25532130fddfead2e880c1099913bd0dcfcd26c954e3d43dd3077a1af03</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Computer Science - Computer Vision and Pattern Recognition</topic><topic>Computer Science - Learning</topic><topic>Computer Science - Robotics</topic><topic>Computer Science - Systems and Control</topic><toplevel>online_resources</toplevel><creatorcontrib>Ivanovic, Boris</creatorcontrib><creatorcontrib>Pavone, Marco</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Ivanovic, Boris</au><au>Pavone, Marco</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Injecting Planning-Awareness into Prediction and Detection Evaluation</atitle><date>2021-10-07</date><risdate>2021</risdate><abstract>Detecting other agents and forecasting their behavior is an integral part of the modern robotic autonomy stack, especially in safety-critical scenarios entailing human-robot interaction such as autonomous driving. Due to the importance of these components, there has been a significant amount of interest and research in perception and trajectory forecasting, resulting in a wide variety of approaches. Common to most works, however, is the use of the same few accuracy-based evaluation metrics, e.g., intersection-over-union, displacement error, log-likelihood, etc. While these metrics are informative, they are task-agnostic and outputs that are evaluated as equal can lead to vastly different outcomes in downstream planning and decision making. In this work, we take a step back and critically assess current evaluation metrics, proposing task-aware metrics as a better measure of performance in systems where they are deployed. Experiments on an illustrative simulation as well as real-world autonomous driving data validate that our proposed task-aware metrics are able to account for outcome asymmetry and provide a better estimate of a model's closed-loop performance.</abstract><doi>10.48550/arxiv.2110.03270</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2110.03270
ispartof
issn
language eng
recordid cdi_arxiv_primary_2110_03270
source arXiv.org
subjects Computer Science - Computer Vision and Pattern Recognition
Computer Science - Learning
Computer Science - Robotics
Computer Science - Systems and Control
title Injecting Planning-Awareness into Prediction and Detection Evaluation
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-08T13%3A34%3A08IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Injecting%20Planning-Awareness%20into%20Prediction%20and%20Detection%20Evaluation&rft.au=Ivanovic,%20Boris&rft.date=2021-10-07&rft_id=info:doi/10.48550/arxiv.2110.03270&rft_dat=%3Carxiv_GOX%3E2110_03270%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true