Forecasting Trajectory and Behavior of Road-Agents Using Spectral Clustering in Graph-LSTMs

We present a novel approach for traffic forecasting in urban traffic scenarios using a combination of spectral graph analysis and deep learning. We predict both the low-level information (future trajectories) as well as the high-level information (road-agent behavior) from the extracted trajectory o...

Full description

Saved in:
Bibliographic Details
Published in:IEEE robotics and automation letters 2020-07, Vol.5 (3), p.4882-4890
Main Authors: Chandra, Rohan, Guan, Tianrui, Panuganti, Srujan, Mittal, Trisha, Bhattacharya, Uttaran, Bera, Aniket, Manocha, Dinesh
Format: Article
Language:English
Subjects:
Accuracy
Algorithms
Approximation
autonomous agents
Clustering
Error reduction
Forecasting
Intelligent transportation systems
Machine learning
Prediction algorithms
Predictions
Predictive models
Reagents
Regularization
Roads
Signal processing algorithms
Spectra
Trajectory
Upper bounds
Vehicle dynamics
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c357t-83ce02d55add20bf04bb1133e6cfead448ff0f99f57300a87b3a974123f10cc43
cites cdi_FETCH-LOGICAL-c357t-83ce02d55add20bf04bb1133e6cfead448ff0f99f57300a87b3a974123f10cc43
container_end_page 4890
container_issue 3
container_start_page 4882
container_title IEEE robotics and automation letters
container_volume 5
creator Chandra, Rohan
Guan, Tianrui
Panuganti, Srujan
Mittal, Trisha
Bhattacharya, Uttaran
Bera, Aniket
Manocha, Dinesh
description We present a novel approach for traffic forecasting in urban traffic scenarios using a combination of spectral graph analysis and deep learning. We predict both the low-level information (future trajectories) as well as the high-level information (road-agent behavior) from the extracted trajectory of each road-agent. Our formulation represents the proximity between the road agents using a weighted dynamic geometric graph (DGG). We use a two-stream graph-LSTM network to perform traffic forecasting using these weighted DGGs. The first stream predicts the spatial coordinates of road-agents, while the second stream predicts whether a road-agent is going to exhibit overspeeding, underspeeding, or neutral behavior by modeling spatial interactions between road-agents. Additionally, we propose a new regularization algorithm based on spectral clustering to reduce the error margin in long-term prediction (3-5 seconds) and improve the accuracy of the predicted trajectories. Moreover, we prove a theoretical upper bound on the regularized prediction error. We evaluate our approach on the Argoverse, Lyft, Apolloscape, and NGSIM datasets and highlight the benefits over prior trajectory prediction methods. In practice, our approach reduces the average prediction error by approximately 75% over prior algorithms and achieves a weighted average accuracy of 91.2% for behavior prediction. Additionally, our spectral regularization improves long-term prediction by up to 70%.
doi_str_mv 10.1109/LRA.2020.3004794
format article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_crossref_primary_10_1109_LRA_2020_3004794</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>9126166</ieee_id><sourcerecordid>2420295171</sourcerecordid><originalsourceid>FETCH-LOGICAL-c357t-83ce02d55add20bf04bb1133e6cfead448ff0f99f57300a87b3a974123f10cc43</originalsourceid><addsrcrecordid>eNpNkE1LAzEQhoMoWGrvgpeA562Tj900x1psFVaEfpw8hGw2abfUzZpshf57d2kRTzMMzzvDPAjdExgTAvIpX07HFCiMGQAXkl-hAWVCJExk2fW__haNYtwDAEmpYDIdoM-5D9bo2Fb1Fq-D3lvT-nDCui7xs93pn8oH7B1eel0m062t24g3sYdXTYcGfcCzwzG2NvSzqsaLoJtdkq_W7_EO3Th9iHZ0qUO0mb-sZ69J_rF4m03zxLBUtMmEGQu0TFNdlhQKB7woCGHMZsZZXXI-cQ6clC4V3Xt6IgqmpeCEMkfAGM6G6PG8twn--2hjq_b-GOrupKK80yJTIkhHwZkywccYrFNNqL50OCkCqreoOouqt6guFrvIwzlSWWv_cEloRrKM_QJA7W0Q</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2420295171</pqid></control><display><type>article</type><title>Forecasting Trajectory and Behavior of Road-Agents Using Spectral Clustering in Graph-LSTMs</title><source>IEEE Electronic Library (IEL) Journals</source><creator>Chandra, Rohan ; Guan, Tianrui ; Panuganti, Srujan ; Mittal, Trisha ; Bhattacharya, Uttaran ; Bera, Aniket ; Manocha, Dinesh</creator><creatorcontrib>Chandra, Rohan ; Guan, Tianrui ; Panuganti, Srujan ; Mittal, Trisha ; Bhattacharya, Uttaran ; Bera, Aniket ; Manocha, Dinesh</creatorcontrib><description>We present a novel approach for traffic forecasting in urban traffic scenarios using a combination of spectral graph analysis and deep learning. We predict both the low-level information (future trajectories) as well as the high-level information (road-agent behavior) from the extracted trajectory of each road-agent. Our formulation represents the proximity between the road agents using a weighted dynamic geometric graph (DGG). We use a two-stream graph-LSTM network to perform traffic forecasting using these weighted DGGs. The first stream predicts the spatial coordinates of road-agents, while the second stream predicts whether a road-agent is going to exhibit overspeeding, underspeeding, or neutral behavior by modeling spatial interactions between road-agents. Additionally, we propose a new regularization algorithm based on spectral clustering to reduce the error margin in long-term prediction (3-5 seconds) and improve the accuracy of the predicted trajectories. Moreover, we prove a theoretical upper bound on the regularized prediction error. We evaluate our approach on the Argoverse, Lyft, Apolloscape, and NGSIM datasets and highlight the benefits over prior trajectory prediction methods. In practice, our approach reduces the average prediction error by approximately 75% over prior algorithms and achieves a weighted average accuracy of 91.2% for behavior prediction. Additionally, our spectral regularization improves long-term prediction by up to 70%.</description><identifier>ISSN: 2377-3766</identifier><identifier>EISSN: 2377-3766</identifier><identifier>DOI: 10.1109/LRA.2020.3004794</identifier><identifier>CODEN: IRALC6</identifier><language>eng</language><publisher>Piscataway: IEEE</publisher><subject>Accuracy ; Algorithms ; Approximation ; autonomous agents ; Clustering ; Error reduction ; Forecasting ; Intelligent transportation systems ; Machine learning ; Prediction algorithms ; Predictions ; Predictive models ; Reagents ; Regularization ; Roads ; Signal processing algorithms ; Spectra ; Trajectory ; Upper bounds ; Vehicle dynamics</subject><ispartof>IEEE robotics and automation letters, 2020-07, Vol.5 (3), p.4882-4890</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2020</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c357t-83ce02d55add20bf04bb1133e6cfead448ff0f99f57300a87b3a974123f10cc43</citedby><cites>FETCH-LOGICAL-c357t-83ce02d55add20bf04bb1133e6cfead448ff0f99f57300a87b3a974123f10cc43</cites><orcidid>0000-0003-2141-9276 ; 0000-0001-7047-9801 ; 0000-0003-4846-5029 ; 0000-0003-4843-6375 ; 0000-0002-0182-6985</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/9126166$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>315,786,790,27957,27958,55147</link.rule.ids></links><search><creatorcontrib>Chandra, Rohan</creatorcontrib><creatorcontrib>Guan, Tianrui</creatorcontrib><creatorcontrib>Panuganti, Srujan</creatorcontrib><creatorcontrib>Mittal, Trisha</creatorcontrib><creatorcontrib>Bhattacharya, Uttaran</creatorcontrib><creatorcontrib>Bera, Aniket</creatorcontrib><creatorcontrib>Manocha, Dinesh</creatorcontrib><title>Forecasting Trajectory and Behavior of Road-Agents Using Spectral Clustering in Graph-LSTMs</title><title>IEEE robotics and automation letters</title><addtitle>LRA</addtitle><description>We present a novel approach for traffic forecasting in urban traffic scenarios using a combination of spectral graph analysis and deep learning. We predict both the low-level information (future trajectories) as well as the high-level information (road-agent behavior) from the extracted trajectory of each road-agent. Our formulation represents the proximity between the road agents using a weighted dynamic geometric graph (DGG). We use a two-stream graph-LSTM network to perform traffic forecasting using these weighted DGGs. The first stream predicts the spatial coordinates of road-agents, while the second stream predicts whether a road-agent is going to exhibit overspeeding, underspeeding, or neutral behavior by modeling spatial interactions between road-agents. Additionally, we propose a new regularization algorithm based on spectral clustering to reduce the error margin in long-term prediction (3-5 seconds) and improve the accuracy of the predicted trajectories. Moreover, we prove a theoretical upper bound on the regularized prediction error. We evaluate our approach on the Argoverse, Lyft, Apolloscape, and NGSIM datasets and highlight the benefits over prior trajectory prediction methods. In practice, our approach reduces the average prediction error by approximately 75% over prior algorithms and achieves a weighted average accuracy of 91.2% for behavior prediction. Additionally, our spectral regularization improves long-term prediction by up to 70%.</description><subject>Accuracy</subject><subject>Algorithms</subject><subject>Approximation</subject><subject>autonomous agents</subject><subject>Clustering</subject><subject>Error reduction</subject><subject>Forecasting</subject><subject>Intelligent transportation systems</subject><subject>Machine learning</subject><subject>Prediction algorithms</subject><subject>Predictions</subject><subject>Predictive models</subject><subject>Reagents</subject><subject>Regularization</subject><subject>Roads</subject><subject>Signal processing algorithms</subject><subject>Spectra</subject><subject>Trajectory</subject><subject>Upper bounds</subject><subject>Vehicle dynamics</subject><issn>2377-3766</issn><issn>2377-3766</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><recordid>eNpNkE1LAzEQhoMoWGrvgpeA562Tj900x1psFVaEfpw8hGw2abfUzZpshf57d2kRTzMMzzvDPAjdExgTAvIpX07HFCiMGQAXkl-hAWVCJExk2fW__haNYtwDAEmpYDIdoM-5D9bo2Fb1Fq-D3lvT-nDCui7xs93pn8oH7B1eel0m062t24g3sYdXTYcGfcCzwzG2NvSzqsaLoJtdkq_W7_EO3Th9iHZ0qUO0mb-sZ69J_rF4m03zxLBUtMmEGQu0TFNdlhQKB7woCGHMZsZZXXI-cQ6clC4V3Xt6IgqmpeCEMkfAGM6G6PG8twn--2hjq_b-GOrupKK80yJTIkhHwZkywccYrFNNqL50OCkCqreoOouqt6guFrvIwzlSWWv_cEloRrKM_QJA7W0Q</recordid><startdate>20200701</startdate><enddate>20200701</enddate><creator>Chandra, Rohan</creator><creator>Guan, Tianrui</creator><creator>Panuganti, Srujan</creator><creator>Mittal, Trisha</creator><creator>Bhattacharya, Uttaran</creator><creator>Bera, Aniket</creator><creator>Manocha, Dinesh</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><orcidid>https://orcid.org/0000-0003-2141-9276</orcidid><orcidid>https://orcid.org/0000-0001-7047-9801</orcidid><orcidid>https://orcid.org/0000-0003-4846-5029</orcidid><orcidid>https://orcid.org/0000-0003-4843-6375</orcidid><orcidid>https://orcid.org/0000-0002-0182-6985</orcidid></search><sort><creationdate>20200701</creationdate><title>Forecasting Trajectory and Behavior of Road-Agents Using Spectral Clustering in Graph-LSTMs</title><author>Chandra, Rohan ; Guan, Tianrui ; Panuganti, Srujan ; Mittal, Trisha ; Bhattacharya, Uttaran ; Bera, Aniket ; Manocha, Dinesh</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c357t-83ce02d55add20bf04bb1133e6cfead448ff0f99f57300a87b3a974123f10cc43</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Accuracy</topic><topic>Algorithms</topic><topic>Approximation</topic><topic>autonomous agents</topic><topic>Clustering</topic><topic>Error reduction</topic><topic>Forecasting</topic><topic>Intelligent transportation systems</topic><topic>Machine learning</topic><topic>Prediction algorithms</topic><topic>Predictions</topic><topic>Predictive models</topic><topic>Reagents</topic><topic>Regularization</topic><topic>Roads</topic><topic>Signal processing algorithms</topic><topic>Spectra</topic><topic>Trajectory</topic><topic>Upper bounds</topic><topic>Vehicle dynamics</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Chandra, Rohan</creatorcontrib><creatorcontrib>Guan, Tianrui</creatorcontrib><creatorcontrib>Panuganti, Srujan</creatorcontrib><creatorcontrib>Mittal, Trisha</creatorcontrib><creatorcontrib>Bhattacharya, Uttaran</creatorcontrib><creatorcontrib>Bera, Aniket</creatorcontrib><creatorcontrib>Manocha, Dinesh</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998–Present</collection><collection>IEEE Xplore</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics &amp; Communications Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts – Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><jtitle>IEEE robotics and automation letters</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Chandra, Rohan</au><au>Guan, Tianrui</au><au>Panuganti, Srujan</au><au>Mittal, Trisha</au><au>Bhattacharya, Uttaran</au><au>Bera, Aniket</au><au>Manocha, Dinesh</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Forecasting Trajectory and Behavior of Road-Agents Using Spectral Clustering in Graph-LSTMs</atitle><jtitle>IEEE robotics and automation letters</jtitle><stitle>LRA</stitle><date>2020-07-01</date><risdate>2020</risdate><volume>5</volume><issue>3</issue><spage>4882</spage><epage>4890</epage><pages>4882-4890</pages><issn>2377-3766</issn><eissn>2377-3766</eissn><coden>IRALC6</coden><abstract>We present a novel approach for traffic forecasting in urban traffic scenarios using a combination of spectral graph analysis and deep learning. We predict both the low-level information (future trajectories) as well as the high-level information (road-agent behavior) from the extracted trajectory of each road-agent. Our formulation represents the proximity between the road agents using a weighted dynamic geometric graph (DGG). We use a two-stream graph-LSTM network to perform traffic forecasting using these weighted DGGs. The first stream predicts the spatial coordinates of road-agents, while the second stream predicts whether a road-agent is going to exhibit overspeeding, underspeeding, or neutral behavior by modeling spatial interactions between road-agents. Additionally, we propose a new regularization algorithm based on spectral clustering to reduce the error margin in long-term prediction (3-5 seconds) and improve the accuracy of the predicted trajectories. Moreover, we prove a theoretical upper bound on the regularized prediction error. We evaluate our approach on the Argoverse, Lyft, Apolloscape, and NGSIM datasets and highlight the benefits over prior trajectory prediction methods. In practice, our approach reduces the average prediction error by approximately 75% over prior algorithms and achieves a weighted average accuracy of 91.2% for behavior prediction. Additionally, our spectral regularization improves long-term prediction by up to 70%.</abstract><cop>Piscataway</cop><pub>IEEE</pub><doi>10.1109/LRA.2020.3004794</doi><tpages>9</tpages><orcidid>https://orcid.org/0000-0003-2141-9276</orcidid><orcidid>https://orcid.org/0000-0001-7047-9801</orcidid><orcidid>https://orcid.org/0000-0003-4846-5029</orcidid><orcidid>https://orcid.org/0000-0003-4843-6375</orcidid><orcidid>https://orcid.org/0000-0002-0182-6985</orcidid></addata></record>
fulltext fulltext
identifier ISSN: 2377-3766
ispartof IEEE robotics and automation letters, 2020-07, Vol.5 (3), p.4882-4890
issn 2377-3766
2377-3766
language eng
recordid cdi_crossref_primary_10_1109_LRA_2020_3004794
source IEEE Electronic Library (IEL) Journals
subjects Accuracy
Algorithms
Approximation
autonomous agents
Clustering
Error reduction
Forecasting
Intelligent transportation systems
Machine learning
Prediction algorithms
Predictions
Predictive models
Reagents
Regularization
Roads
Signal processing algorithms
Spectra
Trajectory
Upper bounds
Vehicle dynamics
title Forecasting Trajectory and Behavior of Road-Agents Using Spectral Clustering in Graph-LSTMs
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-09-21T17%3A35%3A14IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Forecasting%20Trajectory%20and%20Behavior%20of%20Road-Agents%20Using%20Spectral%20Clustering%20in%20Graph-LSTMs&rft.jtitle=IEEE%20robotics%20and%20automation%20letters&rft.au=Chandra,%20Rohan&rft.date=2020-07-01&rft.volume=5&rft.issue=3&rft.spage=4882&rft.epage=4890&rft.pages=4882-4890&rft.issn=2377-3766&rft.eissn=2377-3766&rft.coden=IRALC6&rft_id=info:doi/10.1109/LRA.2020.3004794&rft_dat=%3Cproquest_cross%3E2420295171%3C/proquest_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c357t-83ce02d55add20bf04bb1133e6cfead448ff0f99f57300a87b3a974123f10cc43%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2420295171&rft_id=info:pmid/&rft_ieee_id=9126166&rfr_iscdi=true