http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
GPS 기반 가상차선 구축을 통한 고신뢰성 ADB 실차 시험기법 개발
이도훈(Dohoon Lee),이동현(Donghyeon Lee),김건용(Geonyoung Kim),최문주(Moonjoo Choi),김태형(Tae-hyoung Kim),이태희(Taehee Lee) 한국자동차공학회 2023 한국자동차공학회 부문종합 학술대회 Vol.2023 No.5
Adaptive Driving Beam is an automobile’s lighting system which could coordinate strongness and direction depending on driving conditions and it was announced in 2022 at NHTSA to evaluate ADB capability quantitatively as a test protocol. To proceed the test as the order of ADB test protocol curve test road marked with lane of test scenario is needed. However there was a problem of low credibility and repeatability because a driver should drive after checking the lane with his(her) naked eyes. This problem was found through the data analysis of IIHS test which is similar with ADB. To solve this problem this paper proposes a test method which could increase credibility and repeatability about the result of ADB test through an accurate course traveling based on imaginary lane by using a GPS sensor and a steering robot. As a result of applying the proposed test method, the fact that the curve driving is possible without constructing any lane of test site was found and the test credibility and repeatability could be checked through the course error and standard deviation.
Lee Dohoon,Kim Sun 대한소아청소년과학회 2022 Clinical and Experimental Pediatrics (CEP) Vol.65 No.5
Cells survive and proliferate through complex interactions among diverse molecules across multiomics layers. Conventional experimental approaches for identifying these interactions have built a firm foundation for molecular biology, but their scalability is gradually becoming inadequate compared to the rapid accumulation of multiomics data measured by high-throughput technologies. Therefore, the need for data-driven computational modeling of interactions within cells has been highlighted in recent years. The complexity of multiomics interactions is primarily due to their nonlinearity. That is, their accurate modeling requires intricate conditional dependencies, synergies, or antagonisms between considered genes or proteins, which retard experimental validations. Artificial intelligence (AI) technologies, including deep learning models, are optimal choices for handling complex nonlinear relationships between features that are scalable and produce large amounts of data. Thus, they have great potential for modeling multiomics interactions. Although there exist many AI-driven models for computational biology applications, relatively few explicitly incorporate the prior knowledge within model architectures or training procedures. Such guidance of models by domain knowledge will greatly reduce the amount of data needed to train models and constrain their vast expressive powers to focus on the biologically relevant space. Therefore, it can enhance a model’s interpretability, reduce spurious interactions, and prove its validity and utility. Thus, to facilitate further development of knowledge-guided AI technologies for the modeling of multiomics interactions, here we review representative bioinformatics applications of deep learning models for multiomics interactions developed to date by categorizing them by guidance mode.
Lee, Myunggu,Lee, Junhak,Lee, Dohoon,Cho, Jaehoon,Kim, Sangyong,Park, Chulhwan Elsevier 2011 Enzyme and microbial technology Vol.49 No.4
<P><B>Abstract</B></P><P>A silica gel-based substrate feeding system was developed to prevent methanol inhibiting the catalyst during enzymatic biodiesel synthesis. In the system, silica gel swelled upon methanol addition and subsequently released it in a controlled manner to prevent excess methanol affecting the enzyme. Biodiesel was synthesized by the enzymatic transesterification of canola oil with methanol. For this reaction, enzyme loading, methanol/oil molar ratio, silica gel dosage, glycerol content, and methanol feeding method were tested using commercial immobilized enzymes (Novozym 435 and Lipozyme RM IM from Novozymes). The results showed that conversion was highest with controlled substrate feeding rather than direct methanol addition, suggesting that the method developed here can easily prevent enzyme inhibition by limiting methanol concentration to an acceptable level.</P>