Driving Style Encoder: Situational Reward Adaptation for General-Purpose Planning in Automated Driving
General-purpose planning algorithms for automated driving combine mission, behavior, and local motion planning. Such planning algorithms map features of the environment and driving kinematics into complex reward functions. To achieve this, planning experts often rely on linear reward functions. The...
Gespeichert in:
Veröffentlicht in: | arXiv.org 2020-09 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | |
container_start_page | |
container_title | arXiv.org |
container_volume | |
creator | Rosbach, Sascha Vinit, James Großjohann, Simon Homoceanu, Silviu Li, Xing Roth, Stefan |
description | General-purpose planning algorithms for automated driving combine mission, behavior, and local motion planning. Such planning algorithms map features of the environment and driving kinematics into complex reward functions. To achieve this, planning experts often rely on linear reward functions. The specification and tuning of these reward functions is a tedious process and requires significant experience. Moreover, a manually designed linear reward function does not generalize across different driving situations. In this work, we propose a deep learning approach based on inverse reinforcement learning that generates situation-dependent reward functions. Our neural network provides a mapping between features and actions of sampled driving policies of a model-predictive control-based planner and predicts reward functions for upcoming planning cycles. In our evaluation, we compare the driving style of reward functions predicted by our deep network against clustered and linear reward functions. Our proposed deep learning approach outperforms clustered linear reward functions and is at par with linear reward functions with a-priori knowledge about the situation. |
doi_str_mv | 10.48550/arxiv.1912.03509 |
format | Article |
fullrecord | <record><control><sourceid>proquest_arxiv</sourceid><recordid>TN_cdi_arxiv_primary_1912_03509</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2323284508</sourcerecordid><originalsourceid>FETCH-LOGICAL-a528-2161f1b07b454ac320eb5098b0f7f15632f46dc6b20bb5cc26fd6b9640d472a93</originalsourceid><addsrcrecordid>eNotkF1LwzAYhYMgOOZ-gFcGvO7Md1vvxpxTGDjc7kvSJJLRJTVNp_v3dh-8FwdeDg-cB4AHjKas4Bw9y_jnDlNcYjJFlKPyBowIpTgrGCF3YNJ1O4QQETnhnI6AfY3u4Pw33KRjY-DC10Gb-AI3LvUyueBlA7_Mr4wazrRs0_kHbYhwabyJssnWfWxDZ-C6kd6fSM7DWZ_CXiaj4RV_D26tbDozueYYbN8W2_l7tvpcfsxnq0xyUmQEC2yxQrlinMmaEmTUsKBQyOYWc0GJZULXQhGkFK9rIqwWqhQMaZYTWdIxeLxgzxKqNrq9jMfqJKM6yxgaT5dGG8NPb7pU7UIfh5VdRehwBeOooP-sx2Jc</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2323284508</pqid></control><display><type>article</type><title>Driving Style Encoder: Situational Reward Adaptation for General-Purpose Planning in Automated Driving</title><source>arXiv.org</source><source>Free E- Journals</source><creator>Rosbach, Sascha ; Vinit, James ; Großjohann, Simon ; Homoceanu, Silviu ; Li, Xing ; Roth, Stefan</creator><creatorcontrib>Rosbach, Sascha ; Vinit, James ; Großjohann, Simon ; Homoceanu, Silviu ; Li, Xing ; Roth, Stefan</creatorcontrib><description>General-purpose planning algorithms for automated driving combine mission, behavior, and local motion planning. Such planning algorithms map features of the environment and driving kinematics into complex reward functions. To achieve this, planning experts often rely on linear reward functions. The specification and tuning of these reward functions is a tedious process and requires significant experience. Moreover, a manually designed linear reward function does not generalize across different driving situations. In this work, we propose a deep learning approach based on inverse reinforcement learning that generates situation-dependent reward functions. Our neural network provides a mapping between features and actions of sampled driving policies of a model-predictive control-based planner and predicts reward functions for upcoming planning cycles. In our evaluation, we compare the driving style of reward functions predicted by our deep network against clustered and linear reward functions. Our proposed deep learning approach outperforms clustered linear reward functions and is at par with linear reward functions with a-priori knowledge about the situation.</description><identifier>EISSN: 2331-8422</identifier><identifier>DOI: 10.48550/arxiv.1912.03509</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Algorithms ; Automation ; Coders ; Computer Science - Artificial Intelligence ; Computer Science - Learning ; Computer Science - Robotics ; Deep learning ; Kinematics ; Machine learning ; Mapping ; Motion planning ; Neural networks ; Predictive control</subject><ispartof>arXiv.org, 2020-09</ispartof><rights>2020. 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><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>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,784,885,27925</link.rule.ids><backlink>$$Uhttps://doi.org/10.48550/arXiv.1912.03509$$DView paper in arXiv$$Hfree_for_read</backlink><backlink>$$Uhttps://doi.org/10.1109/ICRA40945.2020.9196778$$DView published paper (Access to full text may be restricted)$$Hfree_for_read</backlink></links><search><creatorcontrib>Rosbach, Sascha</creatorcontrib><creatorcontrib>Vinit, James</creatorcontrib><creatorcontrib>Großjohann, Simon</creatorcontrib><creatorcontrib>Homoceanu, Silviu</creatorcontrib><creatorcontrib>Li, Xing</creatorcontrib><creatorcontrib>Roth, Stefan</creatorcontrib><title>Driving Style Encoder: Situational Reward Adaptation for General-Purpose Planning in Automated Driving</title><title>arXiv.org</title><description>General-purpose planning algorithms for automated driving combine mission, behavior, and local motion planning. Such planning algorithms map features of the environment and driving kinematics into complex reward functions. To achieve this, planning experts often rely on linear reward functions. The specification and tuning of these reward functions is a tedious process and requires significant experience. Moreover, a manually designed linear reward function does not generalize across different driving situations. In this work, we propose a deep learning approach based on inverse reinforcement learning that generates situation-dependent reward functions. Our neural network provides a mapping between features and actions of sampled driving policies of a model-predictive control-based planner and predicts reward functions for upcoming planning cycles. In our evaluation, we compare the driving style of reward functions predicted by our deep network against clustered and linear reward functions. Our proposed deep learning approach outperforms clustered linear reward functions and is at par with linear reward functions with a-priori knowledge about the situation.</description><subject>Algorithms</subject><subject>Automation</subject><subject>Coders</subject><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Learning</subject><subject>Computer Science - Robotics</subject><subject>Deep learning</subject><subject>Kinematics</subject><subject>Machine learning</subject><subject>Mapping</subject><subject>Motion planning</subject><subject>Neural networks</subject><subject>Predictive control</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><sourceid>GOX</sourceid><recordid>eNotkF1LwzAYhYMgOOZ-gFcGvO7Md1vvxpxTGDjc7kvSJJLRJTVNp_v3dh-8FwdeDg-cB4AHjKas4Bw9y_jnDlNcYjJFlKPyBowIpTgrGCF3YNJ1O4QQETnhnI6AfY3u4Pw33KRjY-DC10Gb-AI3LvUyueBlA7_Mr4wazrRs0_kHbYhwabyJssnWfWxDZ-C6kd6fSM7DWZ_CXiaj4RV_D26tbDozueYYbN8W2_l7tvpcfsxnq0xyUmQEC2yxQrlinMmaEmTUsKBQyOYWc0GJZULXQhGkFK9rIqwWqhQMaZYTWdIxeLxgzxKqNrq9jMfqJKM6yxgaT5dGG8NPb7pU7UIfh5VdRehwBeOooP-sx2Jc</recordid><startdate>20200913</startdate><enddate>20200913</enddate><creator>Rosbach, Sascha</creator><creator>Vinit, James</creator><creator>Großjohann, Simon</creator><creator>Homoceanu, Silviu</creator><creator>Li, Xing</creator><creator>Roth, Stefan</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><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20200913</creationdate><title>Driving Style Encoder: Situational Reward Adaptation for General-Purpose Planning in Automated Driving</title><author>Rosbach, Sascha ; Vinit, James ; Großjohann, Simon ; Homoceanu, Silviu ; Li, Xing ; Roth, Stefan</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a528-2161f1b07b454ac320eb5098b0f7f15632f46dc6b20bb5cc26fd6b9640d472a93</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Algorithms</topic><topic>Automation</topic><topic>Coders</topic><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Learning</topic><topic>Computer Science - Robotics</topic><topic>Deep learning</topic><topic>Kinematics</topic><topic>Machine learning</topic><topic>Mapping</topic><topic>Motion planning</topic><topic>Neural networks</topic><topic>Predictive control</topic><toplevel>online_resources</toplevel><creatorcontrib>Rosbach, Sascha</creatorcontrib><creatorcontrib>Vinit, James</creatorcontrib><creatorcontrib>Großjohann, Simon</creatorcontrib><creatorcontrib>Homoceanu, Silviu</creatorcontrib><creatorcontrib>Li, Xing</creatorcontrib><creatorcontrib>Roth, Stefan</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>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><collection>arXiv Computer Science</collection><collection>arXiv.org</collection><jtitle>arXiv.org</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Rosbach, Sascha</au><au>Vinit, James</au><au>Großjohann, Simon</au><au>Homoceanu, Silviu</au><au>Li, Xing</au><au>Roth, Stefan</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Driving Style Encoder: Situational Reward Adaptation for General-Purpose Planning in Automated Driving</atitle><jtitle>arXiv.org</jtitle><date>2020-09-13</date><risdate>2020</risdate><eissn>2331-8422</eissn><abstract>General-purpose planning algorithms for automated driving combine mission, behavior, and local motion planning. Such planning algorithms map features of the environment and driving kinematics into complex reward functions. To achieve this, planning experts often rely on linear reward functions. The specification and tuning of these reward functions is a tedious process and requires significant experience. Moreover, a manually designed linear reward function does not generalize across different driving situations. In this work, we propose a deep learning approach based on inverse reinforcement learning that generates situation-dependent reward functions. Our neural network provides a mapping between features and actions of sampled driving policies of a model-predictive control-based planner and predicts reward functions for upcoming planning cycles. In our evaluation, we compare the driving style of reward functions predicted by our deep network against clustered and linear reward functions. Our proposed deep learning approach outperforms clustered linear reward functions and is at par with linear reward functions with a-priori knowledge about the situation.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><doi>10.48550/arxiv.1912.03509</doi><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | EISSN: 2331-8422 |
ispartof | arXiv.org, 2020-09 |
issn | 2331-8422 |
language | eng |
recordid | cdi_arxiv_primary_1912_03509 |
source | arXiv.org; Free E- Journals |
subjects | Algorithms Automation Coders Computer Science - Artificial Intelligence Computer Science - Learning Computer Science - Robotics Deep learning Kinematics Machine learning Mapping Motion planning Neural networks Predictive control |
title | Driving Style Encoder: Situational Reward Adaptation for General-Purpose Planning in Automated Driving |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-27T08%3A11%3A30IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_arxiv&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Driving%20Style%20Encoder:%20Situational%20Reward%20Adaptation%20for%20General-Purpose%20Planning%20in%20Automated%20Driving&rft.jtitle=arXiv.org&rft.au=Rosbach,%20Sascha&rft.date=2020-09-13&rft.eissn=2331-8422&rft_id=info:doi/10.48550/arxiv.1912.03509&rft_dat=%3Cproquest_arxiv%3E2323284508%3C/proquest_arxiv%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2323284508&rft_id=info:pmid/&rfr_iscdi=true |