Deep Local Trajectory Replanning and Control for Robot Navigation

We present a navigation system that combines ideas from hierarchical planning and machine learning. The system uses a traditional global planner to compute optimal paths towards a goal, and a deep local trajectory planner and velocity controller to compute motion commands. The latter components of t...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2019-05
Main Authors: Pokle, Ashwini, Martín-Martín, Roberto, Goebel, Patrick, Chow, Vincent, Ewald, Hans M, Yang, Junwei, Wang, Zhenkai, Sadeghian, Amir, Sadigh, Dorsa, Savarese, Silvio, Vázquez, Marynel
Format: Article
Language:English
Subjects:
Computer simulation
Machine learning
Navigation systems
Pedestrians
Robot control
Trajectory control
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Pokle, Ashwini
Martín-Martín, Roberto
Goebel, Patrick
Chow, Vincent
Ewald, Hans M
Yang, Junwei
Wang, Zhenkai
Sadeghian, Amir
Sadigh, Dorsa
Savarese, Silvio
Vázquez, Marynel
description We present a navigation system that combines ideas from hierarchical planning and machine learning. The system uses a traditional global planner to compute optimal paths towards a goal, and a deep local trajectory planner and velocity controller to compute motion commands. The latter components of the system adjust the behavior of the robot through attention mechanisms such that it moves towards the goal, avoids obstacles, and respects the space of nearby pedestrians. Both the structure of the proposed deep models and the use of attention mechanisms make the system's execution interpretable. Our simulation experiments suggest that the proposed architecture outperforms baselines that try to map global plan information and sensor data directly to velocity commands. In comparison to a hand-designed traditional navigation system, the proposed approach showed more consistent performance.
doi_str_mv 10.48550/arxiv.1905.05279
format article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2225513044</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2225513044</sourcerecordid><originalsourceid>FETCH-LOGICAL-a524-ee8af2ad17cd7fcd747631b964f4f3b91490ba6ab315e947784eaa26762e96f03</originalsourceid><addsrcrecordid>eNotj8tqwzAUREWh0JDmA7oTdG1XunpZy-C-AqaF4H24tqVgYyRXdkL79zW0i2F2Z84Q8sBZLgul2BOm7_6ac8tUzhQYe0M2IATPCglwR3bzPDDGQBtQSmzI_tm5iVaxxZHWCQfXLjH90KObRgyhD2eKoaNlDEuKI_Ux0WNs4kI_8NqfceljuCe3HsfZ7f57S-rXl7p8z6rPt0O5rzJUIDPnCvSAHTdtZ_waabTgjdXSSy8ay6VlDWpsBFfOSmMK6RBXTQ3Oas_Eljz-YacUvy5uXk5DvKSwLp4A1i9cMCnFL6llSuY</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2225513044</pqid></control><display><type>article</type><title>Deep Local Trajectory Replanning and Control for Robot Navigation</title><source>ProQuest - Publicly Available Content Database</source><creator>Pokle, Ashwini ; Martín-Martín, Roberto ; Goebel, Patrick ; Chow, Vincent ; Ewald, Hans M ; Yang, Junwei ; Wang, Zhenkai ; Sadeghian, Amir ; Sadigh, Dorsa ; Savarese, Silvio ; Vázquez, Marynel</creator><creatorcontrib>Pokle, Ashwini ; Martín-Martín, Roberto ; Goebel, Patrick ; Chow, Vincent ; Ewald, Hans M ; Yang, Junwei ; Wang, Zhenkai ; Sadeghian, Amir ; Sadigh, Dorsa ; Savarese, Silvio ; Vázquez, Marynel</creatorcontrib><description>We present a navigation system that combines ideas from hierarchical planning and machine learning. The system uses a traditional global planner to compute optimal paths towards a goal, and a deep local trajectory planner and velocity controller to compute motion commands. The latter components of the system adjust the behavior of the robot through attention mechanisms such that it moves towards the goal, avoids obstacles, and respects the space of nearby pedestrians. Both the structure of the proposed deep models and the use of attention mechanisms make the system's execution interpretable. Our simulation experiments suggest that the proposed architecture outperforms baselines that try to map global plan information and sensor data directly to velocity commands. In comparison to a hand-designed traditional navigation system, the proposed approach showed more consistent performance.</description><identifier>EISSN: 2331-8422</identifier><identifier>DOI: 10.48550/arxiv.1905.05279</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Computer simulation ; Machine learning ; Navigation systems ; Pedestrians ; Robot control ; Trajectory control</subject><ispartof>arXiv.org, 2019-05</ispartof><rights>2019. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/2225513044?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>780,784,25753,27925,37012,44590</link.rule.ids></links><search><creatorcontrib>Pokle, Ashwini</creatorcontrib><creatorcontrib>Martín-Martín, Roberto</creatorcontrib><creatorcontrib>Goebel, Patrick</creatorcontrib><creatorcontrib>Chow, Vincent</creatorcontrib><creatorcontrib>Ewald, Hans M</creatorcontrib><creatorcontrib>Yang, Junwei</creatorcontrib><creatorcontrib>Wang, Zhenkai</creatorcontrib><creatorcontrib>Sadeghian, Amir</creatorcontrib><creatorcontrib>Sadigh, Dorsa</creatorcontrib><creatorcontrib>Savarese, Silvio</creatorcontrib><creatorcontrib>Vázquez, Marynel</creatorcontrib><title>Deep Local Trajectory Replanning and Control for Robot Navigation</title><title>arXiv.org</title><description>We present a navigation system that combines ideas from hierarchical planning and machine learning. The system uses a traditional global planner to compute optimal paths towards a goal, and a deep local trajectory planner and velocity controller to compute motion commands. The latter components of the system adjust the behavior of the robot through attention mechanisms such that it moves towards the goal, avoids obstacles, and respects the space of nearby pedestrians. Both the structure of the proposed deep models and the use of attention mechanisms make the system's execution interpretable. Our simulation experiments suggest that the proposed architecture outperforms baselines that try to map global plan information and sensor data directly to velocity commands. In comparison to a hand-designed traditional navigation system, the proposed approach showed more consistent performance.</description><subject>Computer simulation</subject><subject>Machine learning</subject><subject>Navigation systems</subject><subject>Pedestrians</subject><subject>Robot control</subject><subject>Trajectory control</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNotj8tqwzAUREWh0JDmA7oTdG1XunpZy-C-AqaF4H24tqVgYyRXdkL79zW0i2F2Z84Q8sBZLgul2BOm7_6ac8tUzhQYe0M2IATPCglwR3bzPDDGQBtQSmzI_tm5iVaxxZHWCQfXLjH90KObRgyhD2eKoaNlDEuKI_Ux0WNs4kI_8NqfceljuCe3HsfZ7f57S-rXl7p8z6rPt0O5rzJUIDPnCvSAHTdtZ_waabTgjdXSSy8ay6VlDWpsBFfOSmMK6RBXTQ3Oas_Eljz-YacUvy5uXk5DvKSwLp4A1i9cMCnFL6llSuY</recordid><startdate>20190513</startdate><enddate>20190513</enddate><creator>Pokle, Ashwini</creator><creator>Martín-Martín, Roberto</creator><creator>Goebel, Patrick</creator><creator>Chow, Vincent</creator><creator>Ewald, Hans M</creator><creator>Yang, Junwei</creator><creator>Wang, Zhenkai</creator><creator>Sadeghian, Amir</creator><creator>Sadigh, Dorsa</creator><creator>Savarese, Silvio</creator><creator>Vázquez, Marynel</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20190513</creationdate><title>Deep Local Trajectory Replanning and Control for Robot Navigation</title><author>Pokle, Ashwini ; Martín-Martín, Roberto ; Goebel, Patrick ; Chow, Vincent ; Ewald, Hans M ; Yang, Junwei ; Wang, Zhenkai ; Sadeghian, Amir ; Sadigh, Dorsa ; Savarese, Silvio ; Vázquez, Marynel</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a524-ee8af2ad17cd7fcd747631b964f4f3b91490ba6ab315e947784eaa26762e96f03</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Computer simulation</topic><topic>Machine learning</topic><topic>Navigation systems</topic><topic>Pedestrians</topic><topic>Robot control</topic><topic>Trajectory control</topic><toplevel>online_resources</toplevel><creatorcontrib>Pokle, Ashwini</creatorcontrib><creatorcontrib>Martín-Martín, Roberto</creatorcontrib><creatorcontrib>Goebel, Patrick</creatorcontrib><creatorcontrib>Chow, Vincent</creatorcontrib><creatorcontrib>Ewald, Hans M</creatorcontrib><creatorcontrib>Yang, Junwei</creatorcontrib><creatorcontrib>Wang, Zhenkai</creatorcontrib><creatorcontrib>Sadeghian, Amir</creatorcontrib><creatorcontrib>Sadigh, Dorsa</creatorcontrib><creatorcontrib>Savarese, Silvio</creatorcontrib><creatorcontrib>Vázquez, Marynel</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>AUTh Library subscriptions: ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>ProQuest - Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection><jtitle>arXiv.org</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Pokle, Ashwini</au><au>Martín-Martín, Roberto</au><au>Goebel, Patrick</au><au>Chow, Vincent</au><au>Ewald, Hans M</au><au>Yang, Junwei</au><au>Wang, Zhenkai</au><au>Sadeghian, Amir</au><au>Sadigh, Dorsa</au><au>Savarese, Silvio</au><au>Vázquez, Marynel</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Deep Local Trajectory Replanning and Control for Robot Navigation</atitle><jtitle>arXiv.org</jtitle><date>2019-05-13</date><risdate>2019</risdate><eissn>2331-8422</eissn><abstract>We present a navigation system that combines ideas from hierarchical planning and machine learning. The system uses a traditional global planner to compute optimal paths towards a goal, and a deep local trajectory planner and velocity controller to compute motion commands. The latter components of the system adjust the behavior of the robot through attention mechanisms such that it moves towards the goal, avoids obstacles, and respects the space of nearby pedestrians. Both the structure of the proposed deep models and the use of attention mechanisms make the system's execution interpretable. Our simulation experiments suggest that the proposed architecture outperforms baselines that try to map global plan information and sensor data directly to velocity commands. In comparison to a hand-designed traditional navigation system, the proposed approach showed more consistent performance.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><doi>10.48550/arxiv.1905.05279</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2019-05
issn 2331-8422
language eng
recordid cdi_proquest_journals_2225513044
source ProQuest - Publicly Available Content Database
subjects Computer simulation
Machine learning
Navigation systems
Pedestrians
Robot control
Trajectory control
title Deep Local Trajectory Replanning and Control for Robot Navigation
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-06T06%3A51%3A55IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Deep%20Local%20Trajectory%20Replanning%20and%20Control%20for%20Robot%20Navigation&rft.jtitle=arXiv.org&rft.au=Pokle,%20Ashwini&rft.date=2019-05-13&rft.eissn=2331-8422&rft_id=info:doi/10.48550/arxiv.1905.05279&rft_dat=%3Cproquest%3E2225513044%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-a524-ee8af2ad17cd7fcd747631b964f4f3b91490ba6ab315e947784eaa26762e96f03%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2225513044&rft_id=info:pmid/&rfr_iscdi=true