http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
MLC–MRAM 기반 디지털 PIM의 이점과 Sensing Margin 분석
안성민(Seongmin Ahn),나태희(Taehui Na) 대한전자공학회 2023 대한전자공학회 학술대회 Vol.2023 No.11
If using multi-level cell (MLC)-MRAM, it is possible to achieve the same inference accuracy with only half the area of single-level cell (SLC)-MRAM, or improve inference accuracy while maintaining the same area. However, to obtain a read-access pass yield of more than 3σ that enables to approach software baseline accuracy of processing-in-memory (PIM), the tunnel magnetoresistance (TMR) of magnetic tunnel junction (MTJ) should be very high and the offset voltage of sense amplifier (SA) should be very small. Therefore, research for better MTJ and SA should be conducted.
양요셉(Yo Seph Yang),안성민(Seongmin Ahn),김성현(Seong Hyeon Kim),최동일(Dongil Choi) 대한기계학회 2022 대한기계학회 춘추학술대회 Vol.2022 No.11
Recently, as the service robot market has grown, robots have emerged in various fields such as industry, service, and sports. In the field of sports, robots that can play with humans such as Forpheus, Robomintoner, and Ldric have been developed. These robots can act as coaches by providing human race data as well as athletic events. We developed a vision system that detects balls and predicts trajectories to develop tennis sports robots. In this paper, we introduce ball detection artificial neural networks and ball trajectory prediction using stereo vision. As a result, the accuracy of the neural network for ball detection in actual tennis images reaches 81%. The ball trajectory prediction error in Gaz ebo simulation is 29.6 cm in the x-axis, 7.2 cm in the y-axis, and 11.7 cm in the z-axis on average.
양요셉(Yo Seph Yang),안성민(Seongmin Ahn),김성현(Seong Hyeon Kim),최동일(Dongil Choi) 대한기계학회 2022 대한기계학회 춘추학술대회 Vol.2022 No.11
Recently, as the service robot market has grown, robots have emerged in various fields such as industry, service, and sports. In the field of sports, robots that can play with humans such as Forpheus, Robomintoner, and Ldric have been developed. These robots can act as coaches by providing human race data as well as athletic events. We developed a vision system that detects balls and predicts trajectories to develop tennis sports robots. In this paper, we introduce ball detection artificial neural networks and ball trajectory prediction using stereo vision. As a result, the accuracy of the neural network for ball detection in actual tennis images reaches 81%. The ball trajectory prediction error in Gaz ebo simulation is 29.6 cm in the x-axis, 7.2 cm in the y-axis, and 11.7 cm in the z-axis on average.