# Study on Relative Accuracy and Verification Method of High-Definition Maps for Autonomous Driving

^{1}

^{2}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. High-Definition Maps and Accuracy

#### 2.1. Accuracy and Precision in Mapping

#### 2.2. Accuracy in HDMs for AD

#### 2.3. Traditional Method for Accuracy

#### 2.4. Limitations of Traditional Methods

#### 2.4.1. Finding the Same Feature Point Pair

#### 2.4.2. Differences between Lane Heading and Side Directions

#### 2.5. Elemental Classification and Decomposition of HDMs

## 3. Materials and Methods

#### 3.1. Case Study

#### 3.1.1. Equipment and Environment

^{2}and the total road mileage is about 22.4 km. It contains various general roads, parking lots, intersections, road appurtenances, etc. The width of each lane is between 3.5 m and 3.75 m. The road markings are clear and contain obvious curves and straight lines. However, the park is a general urban environment with no significant changes in overall elevation and no overly complex marking and road environment. In addition, thanks to the absence of tall buildings blocking the park, point measurements using GNSS technology are not greatly affected.

#### 3.1.2. Data Acquisition

#### 3.2. Methodology

- Calculate the corresponding closet point in the target point set for each point in the source point set.
- Compute the rigid transformation that minimizes the average distance of the above corresponding point pairs. Calculate the rotation and translation parameters.
- The new point set is obtained by using the rotation and translation parameters found in 2, for the source point set.
- Determines if the iterative computation stop condition is met. If it is, the calculation is stopped, if not, the new point set obtained in 3 is used as the new source point set and input into 1 to continue the iterative calculation.

#### 3.2.1. Lane-Heading Direction

#### 3.2.2. Lane-Side Direction

_{1}with the corresponding left true curve is called the corresponding true coordinate point b

_{1}, and the intersection of point a

_{1}with the corresponding right data curve is called the corresponding data point a

_{2}. The normal of the left true curve is made over point b

_{1}, and the right true curve is intersected at point b

_{2}. The distance between points a

_{1}and a

_{2}is the data lane width m

_{1}at data point a

_{1}, and the distance between points b

_{1}and b

_{2}is the corresponding real lane width n

_{1}at point a

_{1}. The purpose of the ICP alignment is to find points on the real curve that correspond to data points and to calculate the data lane width at the corresponding point, as well as the real lane width. The absolute value of a difference between m

_{1}and n

_{1}is calculated and denoted as a relative error of the lane bypass at data point a

_{1}. The relative error is calculated for all data points to obtain the median lane bypass relative error, and then the limit error is obtained by multiplying the median error by two, which represents the lane bypass relative accuracy. This relative accuracy can be set to 20 cm.

## 4. Results

#### 4.1. Results for Lane-Heading Direction

#### 4.2. Results for Collinear Direction

#### 4.3. Results of Traditional Methods

## 5. Discussion

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## References

- Liu, J.; Zhan, J.; Guo, C.; Lei, T.; Li, Y. Data Logic Structure and Key Technologies on Intelligent High-precision Map. J. Geod. Geoinf. Sci.
**2020**, 3, 1–17. [Google Scholar] - Liu, R.; Wang, J.; Zhang, B. High Definition Map for Automated Driving: Overview and Analysis. J. Navigation
**2020**, 73, 324–341. [Google Scholar] [CrossRef] - Levinson, J.; Montemerlo, M.; Thrun, S. (Eds.) Map-Based Precision Vehicle Localization in Urban Environments. In Robotics: Science & Systems III, June; Georgia Institute of Technology: Atlanta, GA, USA, 2007. [Google Scholar]
- Chi, G.; Wenfei, G.; Guangyi, C.; Hongbo, D. A lane-level LBS system for vehicle network with high-precision BDS/GPS positioning. Comput. Intell. Neurosci.
**2015**, 2015, 7. [Google Scholar] - Heiko, G.; Seif, H.X. The key challenge of the self-driving car industry in smart cities-high-definition maps. Engineering
**2016**, 2, 27–35. [Google Scholar] - Sutarwala, B.Z.J.D.; Gradworks, T. GIS for Mapping of Lane-Level Data and Re-Creation in Real Time for Navigation. 2011. Available online: https://escholarship.org/content/qt56m28858/qt56m28858.pdf (accessed on 7 September 2021).
- Schreiber, M.; Knoppel, C.; Franke, U. (Eds.) LaneLoc: Lane marking based localization using highly accurate maps. In Proceedings of the Intelligent Vehicles Symposium (IV), Gold Coast, QLD, Australia, 23–26 June 2013. [Google Scholar]
- Hou, Q.; Li, B.; Cai, Y. High-precision lane-level map elements extracting based on high-resolution remote sensing image. Bull. Surv. Mapp.
**2021**, 3, 38–43. [Google Scholar] - Cai, Y.; Zhang, W.; Yan, Q.; Wang, X.; Bai, J.; Ma, X. Position accuracy and its test method for navigation digital maps. J. Navig. Position.
**2021**, 9, 10–14. [Google Scholar] - Levinson, J.; Thrun, S. (Eds.) Robust Vehicle Localization in Urban Environments Using Probabilistic Maps. In Proceedings of the IEEE International Conference on Robotics & Automation, Anchorage, AK, USA, 3–7 May 2010. [Google Scholar]
- Fairfield, N.; Urmson, C. (Eds.) Traffic light mapping and detection. In Proceedings of the IEEE International Conference on Robotics & Automation, Shanghai, China, 9–13 May 2011. [Google Scholar]
- Kim, K.; Cho, S.; Chung, W. Hd map update for autonomous driving with crowdsourced data. IEEE Robot. Autom. Lett.
**2021**, 99, 1895–1901. [Google Scholar] [CrossRef] - Wuhan University School of Surveying and Mapping Survey Leveling Discipline Group. Error Theory and Fundation of Surveying Adjustment-Second Edition; Wuhan University Press: Wuhan, China, 2009; ISBN 978-9-307-06896-4. [Google Scholar]
- Specifications for Quality Inspection and Acceptance of Surveying and Mapping Products, GB/T 24356-2009. 2009. Available online: http://www.cssn.net.cn/cssn/productDetail/f9f6f180a88e0246cf80283a1b52ff48 (accessed on 30 September 2009).
- Basic Requirements for Products of Digital Topographic Map, GB/T 17278-2009. 2009. Available online: http://www.cssn.net.cn/cssn/productDetail/5b693051528fa9474332a95d083acd3c (accessed on 6 May 2009).
- ISO 19157: 2013 Geographic Information—Data Quality. ISO 14041:2013(E), International Standards Organization. Available online: https://www.iso.org/standard/32575.html (accessed on 7 September 2021).
- Reid, T.; Houts, S.E.; Cammarata, R.; Mills, G.; Agarwal, S.; Vora, A.; Pandey, G. Localization Requirements for Autonomous Vehicles. Preprint
**2019**, 2, 173–190. [Google Scholar] [CrossRef] [Green Version] - Liu, J.; Wu, H.; Guo, C.; Zhang, H.; Zuo, W.; Yang, C. Progress and Consideration of High Precision Road Navigation Map. Strateg. Study CAE
**2018**, 20, 99–105. [Google Scholar] [CrossRef] - Besl, P.J.; Mckay, N.D.; Intelligence, M. A method for registration of 3-D shapes. IEEE Trans. Pattern Anal. Mach. Intell.
**1992**, 14, 239–256. [Google Scholar] [CrossRef]

**Figure 1.**Illustration of absolute and relative precisions of a geometric shape: (

**a**) description of what is contained in the first panel; (

**b**) description of what is contained in the second panel.

Level | Name | Vehicle Lateral and Longitudinal Movement Control | Target and Incident Detection and Response | Dynamic Driving Task Takeover | Design Operating Conditions | Map Data Content |
---|---|---|---|---|---|---|

0 | Emergency assistance | Drivers | Drivers and systems | Drivers | Restrictions apply | Traditional maps |

1 | Partial driving assistance | Drivers and systems | Drivers and systems | Drivers | Restrictions apply | Traditional maps |

2 | Combined driving assistance | Systems | Drivers and systems | Drivers | Restrictions apply | Traditional maps + ADAS data |

3 | Conditional autopilot | Systems | Systems | Dynamic driving tasks take over the user | Restrictions apply | Static HDMs |

4 | Highly automated driving | Systems | Systems | Systems | Restrictions apply | Dynamic HDMs |

5 | Fully automated driving | Systems | Systems | Systems | No restrictions | Smart HDMs |

Lane Number | Median Deviation Error | Real Length of the Curve | Median Error per 100 m | Limit Error of the Section |
---|---|---|---|---|

Section 1 | 0.132 | 166.5 m | 0.080 | 0.159 |

Section 2 | 0.134 | 203.3 m | 0.066 | 0.132 |

Section 3 | 0.143 | 155.7 m | 0.092 | 0.185 |

Section 4 | 0.163 | 179.3 m | 0.091 | 0.181 |

Lane Number | Median Error | Limit Error |
---|---|---|

Section 1 | 0.132 | 0.159 |

Section 2 | 0.134 | 0.132 |

Section 3 | 0.143 | 0.185 |

Section 4 | 0.163 | 0.181 |

Lane Number | Relative Accuracy of the Vertical | Relative Accuracy of the Horizontal |
---|---|---|

Section 1 | 0.091 | 0.0025 |

Section 2 | 0.127 | 0.0031 |

Section 3 | 0.151 | 0.0032 |

Section 4 | 0.116 | 0.0028 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Yu, T.; Huang, H.; Jiang, N.; Acharya, T.D.
Study on Relative Accuracy and Verification Method of High-Definition Maps for Autonomous Driving. *ISPRS Int. J. Geo-Inf.* **2021**, *10*, 761.
https://doi.org/10.3390/ijgi10110761

**AMA Style**

Yu T, Huang H, Jiang N, Acharya TD.
Study on Relative Accuracy and Verification Method of High-Definition Maps for Autonomous Driving. *ISPRS International Journal of Geo-Information*. 2021; 10(11):761.
https://doi.org/10.3390/ijgi10110761

**Chicago/Turabian Style**

Yu, Tengfei, He Huang, Nana Jiang, and Tri Dev Acharya.
2021. "Study on Relative Accuracy and Verification Method of High-Definition Maps for Autonomous Driving" *ISPRS International Journal of Geo-Information* 10, no. 11: 761.
https://doi.org/10.3390/ijgi10110761