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

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹Á¤º¸°úÇÐȸ ³í¹®Áö > Á¤º¸°úÇÐȸ ³í¹®Áö C : ÄÄÇ»ÆÃÀÇ ½ÇÁ¦

Á¤º¸°úÇÐȸ ³í¹®Áö C : ÄÄÇ»ÆÃÀÇ ½ÇÁ¦

Current Result Document : 1 / 2

ÇѱÛÁ¦¸ñ(Korean Title) Æ®·¡ÇÈ ÆÐÅÏ ÇнÀ ±â¹ÝÀÇ ´ÙÁß ±³Â÷·Î ±³Åë½ÅÈ£ Á¦¾î
¿µ¹®Á¦¸ñ(English Title) Multiple-Intersection Traffic Signal Control based on Traffic Pattern Learning
ÀúÀÚ(Author) ÃÖÁø¼·   Á¶¿µÅ   Á¤Àιü   Jinseop Choi   Youngtae Jo   Inbum Jung  
¿ø¹®¼ö·Ïó(Citation) VOL 20 NO. 03 PP. 0171 ~ 0179 (2014. 03)
Çѱ۳»¿ë
(Korean Abstract)
±³Â÷·ÎÀÇ ½ÅÈ£Á¦¾î´Â Áö±Ý±îÁö ¼³Ä¡ ¹× »ç¿ë °£´ÜÇÏ´Ù´Â ÀÌÀ¯·Î °íÁ¤½Ä Á¦¾î ¹æ½ÄÀ» »ç¿ëÇØ¿Ô´Ù. °íÁ¤½Ä Á¦¾î´Â ±³Â÷·ÎÀÇ ±³Åë·®À» ½Ç½Ã°£À¸·Î ¹Ý¿µÇÏÁö ¸øÇϱ⠶§¹®¿¡ È¿À²¼ºÀÌ ¶³¾îÁø´Ù. ÀÌ·¯ÇÑ ¹®Á¦¸¦ ÇØ°áÇϱâ À§ÇØ °¨ÀÀ½Ä Á¦¾î°¡ »ç¿ëµÇ°í ÀÖÁö¸¸ ´ÜÀÏ ±³Â÷·ÎÀÇ ±³Åë·®¸¸À» ÀÌ¿ëÇϱ⠶§¹®¿¡ º¹ÀâÇÑ µµ½ÉÁö ±³Åë ȯ°æ¿¡ Àû¿ëÇϱ⿡´Â ÇÑ°è°¡ ÀÖ´Ù. ÇÏÁö¸¸ µµ½ÉÁöÀÇ ±³ÅëÀº ¸ÅÀÏ ¹Ýº¹ÀûÀÎ ÆÐÅÏÀ» º¸Àδٴ Ư¡ÀÌ ÀÖ´Ù. º» ³í¹®¿¡¼­´Â ¹Ýº¹ÀûÀÎ ±³Åë ÆÐÅÏÀ» ÇнÀÇÏ¿© µµ½ÉÁö ±³Åë¸ÁÀ» È¿À²ÀûÀ¸·Î Á¦¾îÇÒ ¼ö ÀÖ´Â °­È­ÇнÀÀÌ·Ð ±â¹ÝÀÇ »õ·Î¿î ½ÅÈ£Á¦¾î ¾Ë°í¸®ÁòÀ» Á¦¾ÈÇÑ´Ù. Á¦¾ÈÇÏ´Â ¾Ë°í¸®ÁòÀº ±³ÅëÆÐÅÏ ÇнÀ, °æ·ÎŽ»ö, ±³Åë È¥Àⱸ°£ Ž»ö, ´ÙÁß ±³Â÷·Î Á¦¾îÀÇ 4°¡Áö ´Ü°è·Î ÀÌ·ç¾îÁö¸ç, À̵éÀÇ ¿¬µ¿Àº ±³Åë È¥ÀâÀ» ºü¸£°Ô ÇؼÒÇÒ ¼ö ÀÖ°Ô ÇÑ´Ù. ƯÈ÷, ½Ã°£ÀÌ Áö³¯¼ö·Ï ÇнÀ°á°ú´©ÀûÀ¸·Î ÇØ´ç ±³Åë ȯ°æ¿¡ ÃÖÀûÈ­µÈ ½ÅÈ£Á¦¾î°¡ °¡´ÉÇÏ´Ù. Á¦¾ÈÇÏ´Â ¾Ë°í¸®ÁòÀÇ Æò°¡¸¦ À§ÇÏ¿© ¸¶ÀÌÅ©·Î½ºÄÚÇÈ ½Ã¹Ä·¹ÀÌÅ͸¦ »ç¿ëÇÏ¿© ±âÁ¸ÀÇ ½ÅÈ£Á¦¾î ±â¹ý°ú ¼º´ÉÀ» ºñ±³ÇÑ´Ù.
¿µ¹®³»¿ë
(English Abstract)
A pre-timed signal control has been used due to the simplicity of installation and implementation. Since the pre-timed signal control does not reflect real-time traffic information, the efficiency of intersection control is limited. To address the inefficiency of pre-timed signal control, actuated signal control is introduced. However, the actuated signal control has limited applicability in urban environment with complex traffic flow because only a single intersection is considered. However, urban traffic flow has a characteristic of repetitive flow pattern. In this paper, a new traffic signal control based on reinforcement learning is proposed for the effective multiple intersections control of urban traffic environment. The proposed method is composed of traffic pattern learning, path searching, congested area searching, and multiple signals controlling. The interlocking of these components alleviate the traffic congestion problem as soon as possible. In particular, as the learning of traffic pattern is accumulated, the optimal signal control in various traffic conditions can be produced. To prove the efficiency of the proposed signal control method, other previous signal control methods are compared on microscopic simulator.
Å°¿öµå(Keyword) ±³Åë½ÅÈ£ Á¦¾î   ´ÙÁß ±³Â÷·Î Á¦¾î   °­È­ ÇнÀ À̷Р  ±³Åë ÆÐÅÏ   traffic signal control   multiple-intersection control   reinforcement-learning   traffic pattern  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå