high-definition map
File:Zoox autonomous prototype vehicle on Lombard St San Francisco dllu.jpg autonomous prototype vehicle, outfitted with sensors for sensing and high-definition mapping.]]
A high-definition map (HD map) is a highly accurate map used primarily in the field of autonomous driving,{{Cite journal |last1=Liu |first1=Rong |last2=Wang |first2=Jinling |last3=Zhang |first3=Bingqi |date=27 August 2019 |title=High Definition Map for Automated Driving: Overview and Analysis |url=https://www.cambridge.org/core/product/identifier/S0373463319000638/type/journal_article |journal=Journal of Navigation |language=en |volume=73 |issue=2 |pages=324–341 |doi=10.1017/S0373463319000638 |s2cid=202906063 |issn=0373-4633 |hdl=1959.4/unsworks_77182 |hdl-access=free |access-date=2024-11-08}}{{Cite journal |last1=Elghazaly |first1=Gamal |last2=Frank |first2=Raphael |last3=Harvey |first3=Scott |last4=Safko |first4=Stefan |date=2023 |title=High-Definition Maps: Comprehensive Survey, Challenges, and Future Perspectives |journal=IEEE Open Journal of Intelligent Transportation Systems |language=en |volume=4 |pages=527–550 |doi=10.1109/OJITS.2023.3295502 |s2cid=259915452 |doi-access=free}} containing details not normally present on traditional maps.{{Cite web |last=Vardhan |first=Harsha |date=2017-09-22 |title=HD Maps: New age maps powering autonomous vehicles |url=https://www.geospatialworld.net/article/hd-maps-autonomous-vehicles/ |access-date=2021-01-20 |website=Geospatial World |language=en-US}}{{Cite news |last=Matthews |first=Kayla |date=September 16, 2019 |title=What are HD maps, and how will they get us closer to autonomous cars? |work=EETimes |url=https://iot.eetimes.com/what-are-hd-maps-and-how-will-they-get-us-closer-to-autonomous-cars/ |access-date=2024-11-08}} HD maps are often captured using an array of sensors, such as LiDARs, radars, digital cameras, and GPS,{{Cite web |first=Freddie |last=Holmes |date=14 March 2019 |title=HD maps—the hidden sensors that help autonomous vehicles see round corners |url=https://www.automotiveworld.com/articles/hd-maps-the-hidden-sensors-that-help-autonomous-vehicles-see-round-corners/ |access-date=2021-01-20 |website=Automotive World}}{{Cite book |last1=Mueck |first1=Markus |url=https://www.worldcat.org/oclc/1021887635 |title=Networking vehicles to everything : evolving automotive solutions |last2=Karls |first2=Ingolf |date=9 January 2018 |isbn=978-1-5015-0724-3 |location=Boston |publisher=De Gruyter |oclc=1021887635 |access-date=2024-11-08}} and they can also be constructed using aerial imagery.{{Cite journal |last1=Javanmardi |first1=Mahdi |last2=Javanmardi |first2=Ehsan |last3=Gu |first3=Yanlei |last4=Kamijo |first4=Shunsuke |date=2017-09-21 |title=Towards High-Definition 3D Urban Mapping: Road Feature-Based Registration of Mobile Mapping Systems and Aerial Imagery |journal=Remote Sensing |language=en |volume=9 |issue=10 |pages=975 |doi=10.3390/rs9100975 |bibcode=2017RemS....9..975J |issn=2072-4292 |doi-access=free}}{{Cite book |last1=Zang |first1=Andi |last2=Xu |first2=Runsheng |last3=Li |first3=Zichen |last4=Doria |first4=David |title=Proceedings of the 1st ACM SIGSPATIAL Workshop on High-Precision Maps and Intelligent Applications for Autonomous Vehicles |chapter=Lane boundary extraction from satellite imagery |date=2017-11-07 |chapter-url=https://dl.acm.org/doi/10.1145/3149092.3149093 |series=AutonomousGIS '17 |language=en |location=Redondo Beach California |publisher=ACM |pages=1–8 |doi=10.1145/3149092.3149093 |isbn=978-1-4503-5497-4 |arxiv=2002.02362 |s2cid=11512991 |access-date=2024-11-08}} Such maps can be precise at a centimetre level.{{Cite book |last=Jiao |first=Jialin |title=2018 IEEE 42nd Annual Computer Software and Applications Conference (COMPSAC) |date=22 June 2018 |isbn=978-1-5386-2666-5 |volume=01 |pages=367–373 |chapter=Machine Learning Assisted High-Definition Map Creation |doi=10.1109/COMPSAC.2018.00058 |chapter-url=https://ieeexplore.ieee.org/document/8377682 |s2cid=52058583 |access-date=2024-11-08}}
High-definition maps for self-driving cars usually include map elements such as road shape, road marking, traffic signs, and barriers.{{Cite journal |last1=Zang |first1=Andi |last2=Chen |first2=Xin |last3=Trajcevski |first3=Goce |date=2018-06-05 |title=High definition maps in urban context |url=https://dl.acm.org/doi/10.1145/3231541.3231546 |journal=SIGSPATIAL Special |language=en |volume=10 |issue=1 |pages=15–20 |doi=10.1145/3231541.3231546 |s2cid=47019015 |issn=1946-7729 |access-date=2024-11-08|url-access=subscription }} Maintaining high accuracy is one of the biggest challenges in building HD maps of real-world roads. With regard to accuracy, there are two main focus points that determine the quality of an HD map:
- Global accuracy (positioning of a feature on the surface of the Earth).
- Local accuracy (positioning of a feature in relation to road elements around it).
In areas with good GPS reception it is possible to achieve a global accuracy of less than {{cvt|3|cm|in}} deviation using satellite signals and correction data from base stations.
In GPS-denied areas, however, inaccuracy rises with distance traveled through the area, being largest in its middle. This means that the maximum GPS error can be expressed as a percentage of the distance traveled through a GPS-denied area: this value is less than 0.5%.{{Cite web |date=2020-10-22 |title=How Accurate Are HD Maps for Autonomous Driving and ADAS Simulation? |url=https://atlatec.de/blog/how-accurate-are-hd-maps-for-autonomous-driving-and-adas-simulation/ |archive-url=https://web.archive.org/web/20210506084424/https://atlatec.de/blog/how-accurate-are-hd-maps-for-autonomous-driving-and-adas-simulation/ |url-status=dead |archive-date=May 6, 2021 |access-date=2021-05-20 |website=Atlatec |language=en-US}}