°ÇÃ൵½Ã°ø°£¿¬±¸¼Ò

Architecture & Urban Research Institute

pdf¿ø¹®º¸±â ¿¡·¯ ÇØ°á¹æ¹ý ¹Ù·Î°¡±â



¹®ÇåȨ > ¿¬±¸³í¹® > »ó¼¼

[¿ø¹®º¸±â½Ã ¼ÒºñµÇ´Â Æ÷ÀÎÆ® : 100 Æ÷ÀÎÆ®] ¹Ì¸®º¸±â Àοë

Çѱ¹°Ç¼³°ü¸®ÇÐȸ|³í¹®Áý 2025³â 1¿ù

³í¹®¸í ±ÙÀüµµ ½ÅÈ£ ±â¹Ý Àΰ£-·Îº¿ ÇÚµå¿À¹ö »óÅ ÀνÄÀ» À§ÇÑ ÀΰøÁö´É ¸ðµ¨ ¼º´É Æò°¡ / Analysis of AI Model Performance for EMG-Based Human-Robot Handover State Recognition
ÀúÀÚ¸í ±èÅÂÀº(Kim, Taeeun) ; ¾ç°­Çõ(Yang, Kanghyeok)
¹ßÇà»ç Çѱ¹°Ç¼³°ü¸®ÇÐȸ
¼ö·Ï»çÇ× Çѱ¹°Ç¼³°ü¸®ÇÐȸ ³í¹®Áý, Vol.26 No.1 (2025-01)
ÆäÀÌÁö ½ÃÀÛÆäÀÌÁö(67) ÃÑÆäÀÌÁö(7)
ISSN 2005-6095
ÁÖÁ¦ºÐ·ù ½Ã°ø(Àû»ê)
ÁÖÁ¦¾î Àΰ£-·Îº¿ Çùµ¿; ÇÚµå¿À¹ö; ±ÙÀüµµ ½ÅÈ£; ·Îº¿ÆÈ; ½ÉÃþÇнÀ ; Human-Robot Collaboration; Handover; Electromyography Signal; Robotic Arm; Deep Learning
¿ä¾à1 º» ¿¬±¸´Â ·Îº¿°ú Çùµ¿ÇÏ¿© °Ç¼³ÀÛ¾÷À» ¼öÇàÇϱâ À§ÇØ ÇÊ¿äÇÑ °Ç¼³ÀÚÀç ÇÚµå¿À¹öÀÇ ÁøÇà»óȲÀ» ±Ù·ÎÀÚÀÇ ±ÙÀüµµ ½ÅÈ£¸¦ ÅëÇØ ÀνÄÇÏ´Â ±â¼úÀ» °í¾ÈÇÏ°í, ÀΰøÁö´É ¾Ë°í¸®Áò Á¾·ù¿¡ µû¸¥ Àνļº´ÉÀ» ºÐ¼®ÇÏ¿´´Ù. º» ¿¬±¸¿¡¼­´Â ÇÚµå¿À¹ö ÀÛ¾÷ Áß ¹°Ã¼ÀÇ ÆÄÁö Á¤µµ¿¡ µû¶ó Pre-grasp, Half grasp, Full grasp ¼¼ ´Ü°è·Î ³ª´©°í ½ÇÇè½Ç ȯ°æ¿¡¼­ °¢ »óÅ¿¡ ´ëÇÑ ±ÙÀüµµ ½ÅÈ£¸¦ ¼öÁýÇÏ¿´´Ù. ȹµæÇÑ µ¥ÀÌÅ͸¦ ½Ã°£ ¿µ¿ª°ú ÁÖÆļö ¿µ¿ª¿¡¼­ ½Ã°¢È­ÇÏ¿´À¸¸ç °¢ µ¥ÀÌÅÍ ¿µ¿ª¿¡ ´ëÇØ ÇÕ¼º°ö ½Å°æ¸Á(CNN)°ú Àå´Ü±â ±â¾ï ¸Þ¸ð¸®(LSTM)¸¦ È°¿ëÇÏ¿© ÀÎ½Ä ¼º´ÉÀ» Æò°¡ÇÏ°í ÃÖÀûÀÇ ÀΰøÁö´É ¸ðµ¨À» µµÃâÇÏ¿´´Ù. µ¥ÀÌÅÍ ºÐ¼® °á°ú ½Ã°£ ¿µ¿ª µ¥ÀÌÅÍ´Â CNN ±â¹Ý ¸ðµ¨ÀÌ Á¤È®µµ 0.99·Î ¼º´ÉÀÌ ¿ì¼öÇÏ¿´À¸¸ç, ÁÖÆļö ¿µ¿ª µ¥ÀÌÅÍ´Â LSTM ±â¹Ý ¸ðµ¨ÀÌ Á¤È®µµ 0.98·Î ´õ ¿ì¼öÇÑ °ÍÀ¸·Î ³ªÅ¸³µ´Ù. ¶ÇÇÑ ÇнÀÇÏÁö ¾ÊÀº »õ·Î¿î ½ÇÇèÀÚ µ¥ÀÌÅÍ·Î Æò°¡ÇÏ´Â Leave-one-subject-out cross-validationÀ» ºÐ¼®ÇÑ °á°ú LSTM ¸ðµ¨ÀÌ 0.69·Î °¡Àå ³ôÀº ¼º´ÉÀ» ³ªÅ¸³»´Â °ÍÀ» º¸¿´´Ù. ÇØ´ç ¿¬±¸°á°ú´Â Àΰ£°ú ·Îº¿ÀÌ Çùµ¿À¸·Î °Ç¼³ÀÛ¾÷À» ¼öÇàÇÏ´Â ±â¼ú °³¹ß¿¡ ÇÊ¿äÇÑ ±âÃÊ ¿¬±¸·Î¼­ ¿¬±¸ÀÇ ±â¿©°¡ ÀÖÀ¸¸ç, Çùµ¿·Îº¿À» ÅëÇÑ ¾ÈÀü¼º ¹× »ý»ê¼º Çâ»ó¿¡ ±â¿©ÇÒ ¼ö ÀÖÀ» °ÍÀ¸·Î ÆǴܵȴÙ.
¿ä¾à2 The study developed an approach to recognize handover tasks required for collaborative construction work with robots using a worker¡¯s electromyography (EMG) signals. The study investigated the recognition performance based on different artificial intelligence algorithms. The handover task was divided into three stages (Pre-grasp, Half grasp, and Full grasp) depending on the degree of object grasp. The EMG signals of each grasp state were collected in a laboratory environment. The collected data were visualized in time and frequency domains, and recognition performance was evaluated using convolutional neural networks (CNN) and long short-term memory (LSTM) networks for each data domain to derive the optimal AI model. The analysis results showed that the CNNbased model exhibited superior performance with an accuracy of 0.99 for time domain data, while the LSTM-based model achieved better performance with an accuracy of 0.98 for frequency domain data. Furthermore, the leave-onesubject-out cross-validation approach demonstrated that the LSTM model achieved a notably higher performance with an accuracy of 0.69 compared to the CNN model. The results of the study serve as foundational research for developing technologies for human-robot collaboration in construciton are expected to contribute to improvement of the safety and the productivity through collaborative construction robots.
¼ÒÀåó Çѱ¹°Ç¼³°ü¸®ÇÐȸ
¾ð¾î Çѱ¹¾î
DOI https://dx.doi.org/10.6106/KJCEM.2025.26.1.067