近幾年來，高精地圖（High-Definition Maps, HD Maps）的快速發展已逐漸成為現代自動駕駛技術中的一項重要輔助信息，其被應用為自駕車的類感測器以實現道路安全的願景。相對於傳統的電子導航地圖，高精地圖對精準度的需求遠超出其它地圖，並且包含更多的道路環境信息和道路元素。由於車載移動雷射掃描（Mobile Laser Scanning, MLS）系統擁有快速收集高精度環境信息的能力，該系統已經成為現今廣為使用的資料採集方式。然而，後續的地圖數化等測繪任務仍依賴人為操作，這樣的過程需要消耗大量的人力和時間成本。
因此本研究致力於開發自動化演算法，著重於探討如何基於三維點雲的半自動化產製高精地圖架構及其在自駕車模擬器中的仿真模擬測試應用。我們自製的高精地圖包含道路的各種細節，如道路連接性、路面標記、交通號誌等重要資訊。首先，透過點雲處理技術萃取出各種道路元素。並使用被萃取出的車道線點雲進行道路參考線的建立，包括直線和曲線道路段的擬合，使模擬的準確度更高。其次，我們採用了數學模型來建立交叉路口的幾何特性跟連接性。自動化連接交叉路口車道，並輸出成OpenDRIVE格式的地圖。車道之三維均方根誤差於整個實驗場景下小於20公分，整體而言，本研究不僅能萃取特定路面標記，同時將車道進行建模以產製能供自駕車應用之高精度且可靠的高精地圖。
此外，研究中也將實際的交通燈號資訊匯入CARLA模擬器，使得模擬的情境更為真實。最後，我們在CARLA模擬器中進行測試，得出自製地圖在模擬自駕車行駛時的可用性和道路連接性。並在實時獲取自駕車當下位置所對應的道路曲率。這不僅驗證了自製高精地圖的可用性，也突顯了將高精地圖作為自駕車類感測器融入模擬中的潛力。
綜上所述，本研究成功開發出一套基於三維點雲之半自動化產製高精地圖的架構，並成功應用於自駕車模擬器的仿真模擬測試中，為未來自駕車的研究與應用提供了重要的基礎。

In recent years, the rapid evolution of High-Definition Maps (HD Maps) has progressively become a critical supplement to modern autonomous driving technology, serving as a pseudo-sensor for autonomous driving vehicles with the aim of promoting road safety. Compared to traditional electronic navigation maps, HD Maps demand an exceedingly high level of accuracy and cover a more extensive array of road environment information and road elements. Given that the Mobile Laser Scanning (MLS) system, when mounted on a vehicle, is capable of rapidly gathering high-precision environmental data, it has become a widely adopted method of data collection today. However, subsequent tasks such as digitizing the maps still rely heavily on manual operation, a process that needs a huge expense of human and time cost.
Consequently, this research is devoted to the development of automated algorithms, with a particular focus on how to semi-automate the production of HD Maps based on 3D point cloud data and their application in simulation testing in autonomous driving vehicle simulators. Custom HD Maps provided from this research cover various road details, including road connectivity, road markings, traffic signals, and other related information. Initially, this research extract various road elements through point cloud processing techniques. Then, the extracted lane line point clouds are utilized to establish road reference lines, including fitting straight and curved road segments, which enhances the accuracy of the simulation. Subsequently, this research employs mathematical models to construct the geometric features and connectivity of intersections, automate the connection of intersection lanes, and output the map in OpenDRIVE format. The Root Mean Square Error (RMSE) of the 3D lanes is less than 20 cm throughout the entire experimental scene. On the whole, this research not only extracts specific road markings but also models the lanes to create accurate and reliable HD Maps for autonomous driving applications.
Additionally, this research has imported actual traffic light information into the CARLA simulator to make the simulation scenario more realistic. Finally, conducting tests in the CARLA simulator and determined the usability and road connectivity of custom map during simulated autonomous driving. This research also acquired the road curvature corresponding to the current location of the autonomous driving vehicle in real time. This not only validates the usability of custom HD Maps but also underlines the potential of incorporating HD Maps as pseudo-sensors for self-driving vehicles in simulations.
In conclusion, this research has successfully developed a framework for semi-automatically creating HD Maps based on 3D point cloud data and successfully applied it in simulation testing in autonomous driving vehicle simulators. This work lays a significant foundation for future research and applications in autonomous driving.

Bai, M., Mattyus, G., Homayounfar, N., Wang, S., Lakshmikanth, S. K., & Urtasun, R. Deep multi-sensor lane detection. In Proceedings of 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 3102-3109, (2018).
Barsi, M., & Barsi, A. Building opendrive model from mobile mapping data. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, 43(B4-2021), 9–14, (2021).
Bieker-Walz, L., Behrisch, M., & Junghans, M. Analysis of the traffic behavior of emergency vehicles in a microscopic traffic simulation. Engineering EPiC Series in Engineering Analysis of the traffic behavior of emergency vehicles in a microscopic traffic simulation (Vol. 2), (2018).
Bock, T. Unsettled Issues on HD Mapping Technology for Autonomous Driving and ADAS, (2021).
Campello, R. J. G. B., Moulavi, D., Zimek, A., & Sander, J. Hierarchical density estimates for data clustering, visualization, and outlier detection. ACM Transactions on Knowledge Discovery from Data, 10(1), (2015).
CARTURE. 政府一小步，台灣智慧駕駛科技一大步, (2019). Retrieved from https://www.carture.com.tw/topic/article/5663
Chang, Y. F., Chiang, K. W., Tsai, M. L., Lee, P. L., Zeng, J. C., El-Sheimy, N., & Darweesh, H. The Implementation of semi-automated road surface markings extraction schemes utilizing mobile laser scanned point clouds for HD Maps production. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLVIII-1/W1-2023, 2023, pp.93-100, (2023).
Cheng, M., Zhang, H., Wang, C., & Li, J. Extraction and classification of road markings using mobile laser scanning point clouds. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 10(3), 1182-1196, (2016).
Chiang, K. W., Zeng, J. C., Tsai, M. L., Darweesh, H., Chen, P. X., & Wang, C. K. Bending the Curve of HD Maps Production for Autonomous Vehicle Applications in Taiwan. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 15, 8346–8359, (2022).
Chicco, D., & Jurman, G. Machine learning can predict survival of patients with heart failure from serum creatinine and ejection fraction alone. BMC Medical Informatics and Decision Making, 20(1), (2020).
Diaz-Diaz, A., Ocana, M., Llamazares, A., Gomez-Huelamo, C., Revenga, P., & Bergasa, L. M. HD maps: Exploiting OpenDRIVE potential for Path Planning and Map Monitoring. IEEE Intelligent Vehicles Symposium, Proceedings, 2022-June, 1211–1217, (2022).
Dimitrov, D., Knauer, C., Kriegel, K., Rote, G. On the Bounding Boxes Obtained by Principal Component Analysis, (2006).
Dosovitskiy, A., Ros, G., Codevilla, F., Lopez, A., & Koltun, V. CARLA: An Open Urban Driving Simulator, (2017).
Elhousni, M., Lyu, Y., Zhang, Z., Huang, X. Automatic Building and Labeling of HD Maps with Deep Learning, (2020).
Ellum, C. & El-Sheimy, N. Land-based mobile mapping systems. Photogrammetric engineering and remote sensing, 68(1), (2002).
Ester, M., Kriegel, H.-P., Sander, J., & Xu, X. A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise, (1996).
Fagnant, D. J., & Kockelman, K. Preparing a nation for autonomous vehicles: Opportunities, barriers and policy recommendations. Transportation Research Part A: Policy and Practice, 77, 167–181, (2015).
Gómez-Huélamo, C., Egido, J. D., Bergasa, L. M., Barea, R., López-Guillén, E., Arango, F., Araluce, F. & López, J. Train Here, Drive There: Simulating Real-World Use Cases with Fully-Autonomous Driving Architecture in CARLA Simulator. Part of the Advances in Intelligent Systems and Computing book series (AISC,volume 1285), (2020).
Gottschalk, S., Manocha, D., Lin, M. C., Brooks, F. P. Collision Queries using Oriented Bounding Boxes, (2000).
Gwon, G. P., Hur, W. S., Kim, S. W., Seo, S. W. Generation of a Precise and Efficient Lane-Level Road Map for Intelligent Vehicle Systems. IEEE Trans Veh Technol 66, 4517–4533, (2017).
Jermakian, J. S. Crash avoidance potential of four passenger vehicle technologies, (2011).
Jiang, H. Semi-automated Generation of Road Transition Lines Using Mobile Laser Scanning Data, (2017).
Kukko, A. Mobile Laser Scanning–System development, performance and applications. Finnish Geodetic Institute, Kirkkonummi, Finland, (2013).
Li, D., & Okhrin, O. Modified DDPG car-following model with a real-world human driving experience with CARLA simulator. Transportation Research Part C: Emerging Technologies, 147, (2023).
Li, F., Bonnifait, P. & Ibanez-Guzman, J. Estimating localization uncertainty using multi-hypothesis map-matching on high-definition road maps, IEEE 20th International Conference on Intelligent Transportation Systems (ITSC), Yokohama, Japan, 2017, pp. 1-6, (2017).
Liu, R., Wang, J., & Zhang, B. High Definition Map for Automated Driving: Overview and Analysis. In Journal of Navigation (Vol. 73, Issue 2, pp. 324–341). Cambridge University Press, (2020).
Křehlík, Š, Vanžura, M., & Skokan, A. Minimum required accuracy for HD maps. The Journal of Navigation, 1-17, (2023).
Smolnicki, P. M. & Sołtys, J. Car-sharing: The Impact on Metropolitan Spatial Structures, (2015).
Malik, S., Khan, M. A., Aadam, El-Sayed, H., Iqbal, F., Khan, J., & Ullah, O. CARLA+: An Evolution of the CARLA Simulator for Complex Environment Using a Probabilistic Graphical Model. Drones. Drones 7, no. 2: 111, (2023).
Matthias, V., Arndt, J. A., Aulinger, A., Bieser, J., Denier van der Gon, H., Kranenburg, R., Kuenen, J., Neumann, D., Pouliot, G., & Quante, M. Modeling emissions for three-dimensional atmospheric chemistry transport models. In Journal of the Air and Waste Management Association (Vol. 68, Issue 8, pp. 763–800), (2018).
Ma, L. Generation of Horizontally Curved Driving Lines for Autonomous Vehicles Using Mobile Laser Scanning Data. Master's thesis, University of Waterloo, Ontario, Canada, (2017).
Ma, L., Wu, T., Li, Y., Li, J., Chen, Y., & Chapman, M. Automated Extraction of Driving Lines from Mobile Laser Scanning Point Clouds. In Proceedings of 29th International Cartographic Conference (ICC 2019), 1, (2019).
Milakis, D., Van Arem, B., & Van Wee, B. Policy and society related implications of automated driving: A review of literature and directions for future research. Journal of Intelligent Transportation Systems: Technology, Planning, and Operations, 21(4), 324–348, (2017).
Otsu, N. A Threshold Selection Method from Gray-Level Histograms. IEEE Transactions on Systems, Man, and Cybernetics, vol. 9, no. 1, pp. 62-66, (1979).
Pai, H. Y. Automated Road-Networks Generation Based on OpenDRIVE Format and Converting to Lanelet2 for High-Definition Maps Utilizing Mobile Mapping System, (2022).
Poggenhans, F., Pauls, J., H., Janosovits, J., Orf, S., Naumann, M., Kuhnt, F. & Mayr, M. Lanelet2: A high-definition map framework for the future of automated driving. 21st International Conference on Intelligent Transportation Systems (ITSC), Maui, HI, USA, 2018, pp. 1672-1679, (2018).
Poggenhans, F., Pauls, Salscheider, N. O., Stiller, C. Precise Localization in High-Definition Road Maps for Urban Regions. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Madrid, Spain, pp. 2167-2174, (2018).
Powers, D. M. W. Evaluation: From precision, recall and F-measure to ROC, International Journal of Machine Learning Technology 2:1 (2011), pp.37-63, (2011).
Rusu, R. B. & Cousins, S. 3D is here: Point Cloud Library (PCL). IEEE International Conference on Robotics and Automation, (2011).
Rummelhard, L., Negre, A., & Laugier, C. Conditional Monte Carlo Dense Occupancy Tracker. IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC, 2015-October, 2485–2490, (2015).
Schoettle, B., & Sivak, M. Potential impact of self-driving vehicles on household vehicle demand and usage, (2015).
Schubert, E., Sander, J., Ester, M., Kriegel, H. P., & Xu, X. DBSCAN revisited, revisited: Why and how you should (still) use DBSCAN. ACM Transactions on Database Systems, 42(3), (2017).
Schwarting, W., Alonso-Mora, J., & Rus, D. Planning and Decision-Making for Autonomous Vehicles. Annu. Rev. Control Robot. Auton. Syst. 2018, 1, 187–210, (2018).
Seif, H. G., & Hu, X. Autonomous Driving in the iCity—HD Maps as a Key Challenge of the Automotive Industry. Engineering, 2(2), 159–162, (2016).
Sester, M., Arsanjani, J. J., Klammer, R., Burghardt, D., & Haunert, J. H. Integrating and generalising volunteered geographic information. In Advances in Cartography and GIScience (pp. 119-137), (2017).
Shimada, H., Yamaguchi, A., Takada, H., & Sato, K. Implementation and Evaluation of Local Dynamic Map in Safety Driving Systems. Journal of Transportation Technologies, 05(02), 102–112, (2015).
TAICS TR-0016. Verification and validation guidelines for HD Maps v0.4, (2020). Retrieved from https://gps.moi.gov.tw/sscenter/introduce/BulletinPage_Info.aspx?ID=100
Wang, Y., Chao, W. L., Garg, D., Hariharan, B., Campbell, M. & Weinberger, K. Q. Pseudo-LiDAR From Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Driving. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA, 2019 pp. 8437-8445, (2019).
Wen, C., Sun, X., Li, J., Wang, C., Guo, Y., & Habib, A. A deep learning framework for road marking extraction, classification and completion from mobile laser scanning point clouds. ISPRS journal of photogrammetry and remote sensing, 147, 178-192, (2019).
Yang, B., Fang, L., Li, Q., & Li, J. Automated extraction of road markings from mobile LiDAR point clouds. Photogrammetric Engineering & Remote Sensing, 78(4), 331-338, (2012).
Yu, Y., Li, J., Guan, H., Jia, F., & Wang, C. Learning hierarchical features for automated extraction of road markings from 3-D mobile LiDAR point clouds. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 8(2), 709-726, (2014).
Zeng, J., C. Automated Road-Elements Modelling and Centerline Generation for High-Definition Maps Utilizing 3D Point Cloud, (2020).
Zhang, W., Qi, J., Wan, P., Wang, H., Xie, D., Wang, X., Yan, G., 2016. An easy-to-use airborne LiDAR data filtering method based on cloth simulation. Remote Sensing 8, no. 6: 501, (2016).
Zhou, Z., Rother, C., & Chen, J. Event-Triggered Model Predictive Control for Autonomous Vehicle Path Tracking: Validation Using CARLA Simulator. IEEE Transactions on Intelligent Vehicles, (2023).