• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ÇÐȸÁö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ÇÐȸÁö > µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

µ¥ÀÌÅͺ£À̽º ¿¬±¸È¸Áö(SIGDB)

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) TPR Æ®¸®ÀÇ ¿¹Ãø ±ËÀûÀ» »ç¿ëÇÑ ÀÚÀ¯ À̵¿ °´Ã¼ Ãæµ¹ÀÇ ¿¹Ãø ŽÁö ¾Ë°í¸®Áò
¿µ¹®Á¦¸ñ(English Title) A Detection Algorithm of Collisions of Free Moving Objects Using Predicted Trajectories of TPR-tree
ÀúÀÚ(Author) ½ºÆ¼¾Æ¿Ï Çϸ®   È«ºÀÈñ   Hari Setiawan   Bonghee Hong  
¿ø¹®¼ö·Ïó(Citation) VOL 38 NO. 02 PP. 0055 ~ 0070 (2022. 08)
Çѱ۳»¿ë
(Korean Abstract)
À̵¿ °´Ã¼µéÀÇ ¹Ì·¡ Ãæµ¹ ŽÁö¸¦ ¿¹ÃøÇϱâ À§ÇÑ ½Ã°ø°£ ÀûÀÎ °ãħÀ» °è»êÇÏ´Â ¹æ¹ýÀº °è»ê¾çÀÌ Å©±â ¶§¹®¿¡ Çö½ÇÀûÀÌÁö ¸øÇÏ´Ù. ÀÌ ³í¹®Àº À̵¿ °´Ã¼µéÀÇ ¼ýÀÚ°¡ Áõ°¡ÇÒ ¶§ ±ÙÁ¢ÇÏ´Â °´Ã¼µéÀÇ ½ÖÀ» ¿©°úÇÑ ÈÄ¿¡ ¹Ì·¡ °ãħÀ» ½Ã°ø°£ »öÀÎÀ» »ç¿ëÇÏ´Â ¹æ¹ýÀ» Á¦¾ÈÇÑ´Ù. À̵¿ °´Ã¼µéÀÇ ±ÙÁ¢ À§Ä¡¿¡ ´ëÇÑ ¿©°ú 󸮸¦ À§ÇÏ¿© ½Ã°ø°£ »öÀÎÀ» ±¸¼ºÇÏÁö ¾Ê´Â´Ù. ¿©°úµÈ ±ÙÁ¢ Èĺ¸ ´ë»ó À̵¿ °´Ã¼µé¿¡ ´ëÇÏ¿© TPR-Æ®¸®¸¦ ±¸¼ºÇÏ¿© °ãħ ÁúÀÇ Ã³¸®¸¦ ÅëÇØ ¹Ì·¡ Ãæµ¹ ŽÁö¸¦ ¿¹ÃøÇÑ´Ù. ¶ÇÇÑ ±ËÀû ÇÊÅ͸µÀ» »ç¿ëÇÏ¿© À妽ÌÇØ¾ß ÇÏ´Â À̵¿ °´Ã¼ ´ë»ó Èĺ¸ ¼ö¸¦ ÁÙÀδÙ. ÀÌ ³í¹®ÀÇ ±â¿©´Â ¼Óµµ¿Í À§Ä¡°¡ ´Ù¸¥ ¼ö ¸¹Àº À̵¿ °´Ã¼µé¿¡ ´ëÇÏ¿© ½Ç½Ã°£À¸·Î ¹Ì·¡ÀÇ Ãæµ¹À» ¿¹ÃøÇÏ´Â ¾Ë°í¸®ÁòÀÇ ¼º´É °³¼±ÀÌ´Ù. ÀÌ ³í¹®ÀÇ ¹æ¹ýÀÇ ½ÇÇè °á°ú´Â ±âÁ¸ÀÇ SAT ±â¹Ý 1Â÷¿ø ÇÁ·ÎÁ§¼Ç °ãħ °è»ê ¹æ¹ý¿¡ ºñÇØ ¾à 60% Çâ»óµÇ´Â °ÍÀ» º¸¿©ÁÖ°í ÀÖ´Ù.
¿µ¹®³»¿ë
(English Abstract)
The existing methods of calculating the spatiotemporal overlapping for predicting future collision detection of moving objects are not realistic because of the large amount of computation. When the number of moving objects increases, this paper proposes a method of using a spatiotemporal index for detecting future overlapping after filtering pairs of proximity moving objects. A spatiotemporal index is not constructed for filtering processing on the proximity of moving objects. By constructing a TPR-tree for the filtered proximity candidate moving objects, future collision detection is predicted through overlapping query processing. It also uses trajectory filtering to reduce the number of vessel target candidates that need to be indexed. The contribution of this paper is to improve the performance of algorithms for predicting future collisions in real time for numerous moving objects of different speeds and positions. The experimental results of the proposed method show an improvement of about 60% compared to the existing SAT-based one-dimensional projection method.
Å°¿öµå(Keyword) Ãæµ¹ ȸÇÇ   ±ËÀû   À妽º ÁúÀÇ   ±ÙÁ¢ÁúÀÇ   ¿¹Ãø ÁúÀÇ   collision avoidance   trajectory   index   close proximity   prediction query  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå