K-D Bonsai: ISA-Extensions to Compress K-D Trees for Autonomous Driving Tasks

Autonomous Driving (AD) systems extensively manipulate 3D point clouds for object detection and vehicle localization. Thereby, efficient processing of 3D point clouds is crucial in these systems. In this work we propose K-D Bonsai, a technique to cut down memory usage during radius search, a critica...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2023-08
Hauptverfasser: Becker, Pedro H E, Arnau, José María, González, Antonio
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 Becker, Pedro H E
Arnau, José María
González, Antonio
description Autonomous Driving (AD) systems extensively manipulate 3D point clouds for object detection and vehicle localization. Thereby, efficient processing of 3D point clouds is crucial in these systems. In this work we propose K-D Bonsai, a technique to cut down memory usage during radius search, a critical building block of point cloud processing. K-D Bonsai exploits value similarity in the data structure that holds the point cloud (a k-d tree) to compress the data in memory. K-D Bonsai further compresses the data using a reduced floating-point representation, exploiting the physically limited range of point cloud values. For easy integration into nowadays systems, we implement K-D Bonsai through Bonsai-extensions, a small set of new CPU instructions to compress, decompress, and operate on points. To maintain baseline safety levels, we carefully craft the Bonsai-extensions to detect precision loss due to compression, allowing re-computation in full precision to take place if necessary. Therefore, K-D Bonsai reduces data movement, improving performance and energy efficiency, while guaranteeing baseline accuracy and programmability. We evaluate K-D Bonsai over the euclidean cluster task of Autoware.ai, a state-of-the-art software stack for AD. We achieve an average of 9.26% improvement in end-to-end latency, 12.19% in tail latency, and a reduction of 10.84% in energy consumption. Differently from expensive accelerators proposed in related work, K-D Bonsai improves radius search with minimal area increase (0.36%).
doi_str_mv 10.48550/arxiv.2302.00361
format Article
fullrecord <record><control><sourceid>proquest_arxiv</sourceid><recordid>TN_cdi_arxiv_primary_2302_00361</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2771816789</sourcerecordid><originalsourceid>FETCH-LOGICAL-a951-da4f5345470b49f9d79ae66373884d6b00a070a080d6e3bab07719a0cdd32cc23</originalsourceid><addsrcrecordid>eNotj01Lw0AURQdBsNT-AFcOuE59me-4i2nVYsWF2YdJMpGpJhNnklL_vWnr4nHhcbjcg9BNDEumOId77Q92vyQUyBKAivgCzQilcaQYIVdoEcIOAIiQhHM6Q2-v0Qo_ui5o-4A3H2m0PgymC3b64MHhzLW9NyHgI5Z7YwJunMfpOLjOtW4MeOXt3nafONfhK1yjy0Z_B7P4zznKn9Z59hJt3583WbqNdMLjqNas4ZRxJqFkSZPUMtFGCCqpUqwWJYAGOZ2CWhha6hKkjBMNVV1TUlWEztHtufbkWvTettr_Fkfn4uQ8EXdnovfuZzRhKHZu9N20qSBTmYqFVAn9A2IcV34</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2771816789</pqid></control><display><type>article</type><title>K-D Bonsai: ISA-Extensions to Compress K-D Trees for Autonomous Driving Tasks</title><source>arXiv.org</source><source>Free E- Journals</source><creator>Becker, Pedro H E ; Arnau, José María ; González, Antonio</creator><creatorcontrib>Becker, Pedro H E ; Arnau, José María ; González, Antonio</creatorcontrib><description>Autonomous Driving (AD) systems extensively manipulate 3D point clouds for object detection and vehicle localization. Thereby, efficient processing of 3D point clouds is crucial in these systems. In this work we propose K-D Bonsai, a technique to cut down memory usage during radius search, a critical building block of point cloud processing. K-D Bonsai exploits value similarity in the data structure that holds the point cloud (a k-d tree) to compress the data in memory. K-D Bonsai further compresses the data using a reduced floating-point representation, exploiting the physically limited range of point cloud values. For easy integration into nowadays systems, we implement K-D Bonsai through Bonsai-extensions, a small set of new CPU instructions to compress, decompress, and operate on points. To maintain baseline safety levels, we carefully craft the Bonsai-extensions to detect precision loss due to compression, allowing re-computation in full precision to take place if necessary. Therefore, K-D Bonsai reduces data movement, improving performance and energy efficiency, while guaranteeing baseline accuracy and programmability. We evaluate K-D Bonsai over the euclidean cluster task of Autoware.ai, a state-of-the-art software stack for AD. We achieve an average of 9.26% improvement in end-to-end latency, 12.19% in tail latency, and a reduction of 10.84% in energy consumption. Differently from expensive accelerators proposed in related work, K-D Bonsai improves radius search with minimal area increase (0.36%).</description><identifier>EISSN: 2331-8422</identifier><identifier>DOI: 10.48550/arxiv.2302.00361</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Computer memory ; Computer Science - Hardware Architecture ; Data structures ; Energy consumption ; Floating point arithmetic ; Object recognition ; Three dimensional models</subject><ispartof>arXiv.org, 2023-08</ispartof><rights>2023. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,784,885,27924</link.rule.ids><backlink>$$Uhttps://doi.org/10.48550/arXiv.2302.00361$$DView paper in arXiv$$Hfree_for_read</backlink><backlink>$$Uhttps://doi.org/10.1145/3579371.3589055$$DView published paper (Access to full text may be restricted)$$Hfree_for_read</backlink></links><search><creatorcontrib>Becker, Pedro H E</creatorcontrib><creatorcontrib>Arnau, José María</creatorcontrib><creatorcontrib>González, Antonio</creatorcontrib><title>K-D Bonsai: ISA-Extensions to Compress K-D Trees for Autonomous Driving Tasks</title><title>arXiv.org</title><description>Autonomous Driving (AD) systems extensively manipulate 3D point clouds for object detection and vehicle localization. Thereby, efficient processing of 3D point clouds is crucial in these systems. In this work we propose K-D Bonsai, a technique to cut down memory usage during radius search, a critical building block of point cloud processing. K-D Bonsai exploits value similarity in the data structure that holds the point cloud (a k-d tree) to compress the data in memory. K-D Bonsai further compresses the data using a reduced floating-point representation, exploiting the physically limited range of point cloud values. For easy integration into nowadays systems, we implement K-D Bonsai through Bonsai-extensions, a small set of new CPU instructions to compress, decompress, and operate on points. To maintain baseline safety levels, we carefully craft the Bonsai-extensions to detect precision loss due to compression, allowing re-computation in full precision to take place if necessary. Therefore, K-D Bonsai reduces data movement, improving performance and energy efficiency, while guaranteeing baseline accuracy and programmability. We evaluate K-D Bonsai over the euclidean cluster task of Autoware.ai, a state-of-the-art software stack for AD. We achieve an average of 9.26% improvement in end-to-end latency, 12.19% in tail latency, and a reduction of 10.84% in energy consumption. Differently from expensive accelerators proposed in related work, K-D Bonsai improves radius search with minimal area increase (0.36%).</description><subject>Computer memory</subject><subject>Computer Science - Hardware Architecture</subject><subject>Data structures</subject><subject>Energy consumption</subject><subject>Floating point arithmetic</subject><subject>Object recognition</subject><subject>Three dimensional models</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</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>eNotj01Lw0AURQdBsNT-AFcOuE59me-4i2nVYsWF2YdJMpGpJhNnklL_vWnr4nHhcbjcg9BNDEumOId77Q92vyQUyBKAivgCzQilcaQYIVdoEcIOAIiQhHM6Q2-v0Qo_ui5o-4A3H2m0PgymC3b64MHhzLW9NyHgI5Z7YwJunMfpOLjOtW4MeOXt3nafONfhK1yjy0Z_B7P4zznKn9Z59hJt3583WbqNdMLjqNas4ZRxJqFkSZPUMtFGCCqpUqwWJYAGOZ2CWhha6hKkjBMNVV1TUlWEztHtufbkWvTettr_Fkfn4uQ8EXdnovfuZzRhKHZu9N20qSBTmYqFVAn9A2IcV34</recordid><startdate>20230830</startdate><enddate>20230830</enddate><creator>Becker, Pedro H E</creator><creator>Arnau, José María</creator><creator>González, Antonio</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>PTHSS</scope><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20230830</creationdate><title>K-D Bonsai: ISA-Extensions to Compress K-D Trees for Autonomous Driving Tasks</title><author>Becker, Pedro H E ; Arnau, José María ; González, Antonio</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a951-da4f5345470b49f9d79ae66373884d6b00a070a080d6e3bab07719a0cdd32cc23</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer memory</topic><topic>Computer Science - Hardware Architecture</topic><topic>Data structures</topic><topic>Energy consumption</topic><topic>Floating point arithmetic</topic><topic>Object recognition</topic><topic>Three dimensional models</topic><toplevel>online_resources</toplevel><creatorcontrib>Becker, Pedro H E</creatorcontrib><creatorcontrib>Arnau, José María</creatorcontrib><creatorcontrib>González, Antonio</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; 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>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>Becker, Pedro H E</au><au>Arnau, José María</au><au>González, Antonio</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>K-D Bonsai: ISA-Extensions to Compress K-D Trees for Autonomous Driving Tasks</atitle><jtitle>arXiv.org</jtitle><date>2023-08-30</date><risdate>2023</risdate><eissn>2331-8422</eissn><abstract>Autonomous Driving (AD) systems extensively manipulate 3D point clouds for object detection and vehicle localization. Thereby, efficient processing of 3D point clouds is crucial in these systems. In this work we propose K-D Bonsai, a technique to cut down memory usage during radius search, a critical building block of point cloud processing. K-D Bonsai exploits value similarity in the data structure that holds the point cloud (a k-d tree) to compress the data in memory. K-D Bonsai further compresses the data using a reduced floating-point representation, exploiting the physically limited range of point cloud values. For easy integration into nowadays systems, we implement K-D Bonsai through Bonsai-extensions, a small set of new CPU instructions to compress, decompress, and operate on points. To maintain baseline safety levels, we carefully craft the Bonsai-extensions to detect precision loss due to compression, allowing re-computation in full precision to take place if necessary. Therefore, K-D Bonsai reduces data movement, improving performance and energy efficiency, while guaranteeing baseline accuracy and programmability. We evaluate K-D Bonsai over the euclidean cluster task of Autoware.ai, a state-of-the-art software stack for AD. We achieve an average of 9.26% improvement in end-to-end latency, 12.19% in tail latency, and a reduction of 10.84% in energy consumption. Differently from expensive accelerators proposed in related work, K-D Bonsai improves radius search with minimal area increase (0.36%).</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><doi>10.48550/arxiv.2302.00361</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2023-08
issn 2331-8422
language eng
recordid cdi_arxiv_primary_2302_00361
source arXiv.org; Free E- Journals
subjects Computer memory
Computer Science - Hardware Architecture
Data structures
Energy consumption
Floating point arithmetic
Object recognition
Three dimensional models
title K-D Bonsai: ISA-Extensions to Compress K-D Trees for Autonomous Driving Tasks
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-08T10%3A38%3A04IST&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=K-D%20Bonsai:%20ISA-Extensions%20to%20Compress%20K-D%20Trees%20for%20Autonomous%20Driving%20Tasks&rft.jtitle=arXiv.org&rft.au=Becker,%20Pedro%20H%20E&rft.date=2023-08-30&rft.eissn=2331-8422&rft_id=info:doi/10.48550/arxiv.2302.00361&rft_dat=%3Cproquest_arxiv%3E2771816789%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=2771816789&rft_id=info:pmid/&rfr_iscdi=true