Á¤º¸°úÇÐȸ ÄÄÇ»ÆÃÀÇ ½ÇÁ¦ ³í¹®Áö (KIISE Transactions on Computing Practices)
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
¶óÀÌ´õ ±â¹Ý 3Â÷¿ø Á¡±º Áöµµ¸¦ ÀÌ¿ëÇÑ µµ·Î Áöµµ ÀÚµ¿ ÃßÃâ ¹æ¹ý |
¿µ¹®Á¦¸ñ(English Title) |
An Automatic Extraction Method for Road Maps using LiDAR-based 3D Point Maps |
ÀúÀÚ(Author) |
ÃÖ±ÔÁø
±è°Èñ
Kyujin Choi
Kanghee Kim
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 27 NO. 05 PP. 0234 ~ 0240 (2021. 05) |
Çѱ۳»¿ë (Korean Abstract) |
µµ½É ÀÚÀ² ÁÖÇàÀ» À§Çؼ´Â ÀÚÂ÷ ÃøÀ§¿ë °íÁ¤¹Ð Á¡±º Áöµµ¿Í °æ·Î °èȹ¿ë º¤ÅÍ ÇüÅÂÀÇ µµ·ÎÁöµµ°¡ ¿ä±¸µÈ´Ù. º¤ÅÍ Áöµµ´Â Â÷·® ¶óÀÌ´õ ¼¾¼¿Í Ä«¸Þ¶ó ¼¾¼¸¦ Ķ¸®ºê·¹À̼ÇÇÑ »óÅ¿¡¼ Ä«¸Þ¶ó À̹ÌÁö¸¦ ÀÌ¿ëÇÏ¿© Á¦ÀÛÇÏ´Â ¹æ¹ý°ú ÀÌ¿Í º°°³·Î °íºñ¿ë Ç×°ø»çÁøÀ» ÀÌ¿ëÇÏ¿© Á¦ÀÛÇÏ´Â ¹æ¹ýÀÌ ÀÖ´Ù. º» ³í¹®Àº Ä«¸Þ¶ó¸¦ »ç¿ëÇÏÁö ¾Ê°í, ¶óÀÌ´õ¸¸À» ÀÌ¿ëÇÏ¿© Á¦ÀÛµÈ Á¡±º Áöµµ·ÎºÎÅÍ Á÷Á¢ º¤ÅÍ Áöµµ¸¦ ÀÚµ¿ ÃßÃâÇÏ´Â ¹æ¹ýÀ» Á¦¾ÈÇÑ´Ù. Á¦¾ÈÇÏ´Â ¹æ¹ýÀ¸·Î ÃßÃâµÈ º¤ÅÍ Áöµµ´Â Á¡±º Áöµµ¿Í ÁÂÇ¥°è °üÁ¡¿¡¼ Àß µ¿±âȵǸç, ½ÇÁ¦ ÀÚÀ² ÁÖÇà Â÷·®ÀÌ Â÷·Î Áß¾Ó À¯Áö(lane centering) ±â´ÉÀ» ¼öÇàÇϴµ¥ ÃæºÐÇÑ Á¤¹Ðµµ¸¦ °¡ÁüÀ» ½ÇÇèÀ» ÅëÇØ È®ÀÎÇÏ¿´´Ù.
|
¿µ¹®³»¿ë (English Abstract) |
For autonomous driving in urban areas, an HD (High Definition) point map and a vector-type road map are required for ego-vehicle localization and route planning, respectively. The vector map can be extracted from camera images assuming that the LiDAR and camera sensors on the vehicle are well-calibrated, or can be extracted from highly expensive aerial photographs. This paper proposes an automatic method to extract the vector map directly from the point map that was composed using only LiDARs without cameras. Through the experiments, it is shown that the resulting vector map from the proposed method is well synchronized with the original point map in the point map coordinate system and, possesses sufficient precision for a real autonomous vehicle to perform lane centering.
|
Å°¿öµå(Keyword) |
°íÁ¤¹Ð Áöµµ
Á¡±º Áöµµ
º¤ÅÍ Áöµµ
¶óÀÌ´õ ¼¾¼
ÀÚµ¿ ÃßÃâ
HD maps
point maps
vector maps
lidar sensors
automatic extraction
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|