http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Jigang Han,Jian Li,Xiaoqing Li,Qunlu Liu,Yonghong Hu 한국원예학회 2021 Horticulture, Environment, and Biotechnology Vol.62 No.2
Tree peonies distributed in Baokang County (Hubei Province, China) (BKTPs) displayed extensive morphological variation,especially in fl ower color patterns and fruit setting rates. The genetic diversity of BKTPs and their relationships to nine wildspecies and 14 representative traditional cultivars were assessed using simple sequence repeat (SSR) markers in this study. All 19 analyzed loci were polymorphic, and 85 alleles were found. The number of alleles per locus varied from 3 to 7, withan average of 4.474. Cluster analysis with UPGMA method based on SSR markers partitioned 46 accessions of BKTPs intotwo major clusters and six subgroups, closing to P. ostii , P. qiui , P. rockii , P. jishanensis and diff erent cultivars, respectively. Analysis of fl avonoids with UPLC-Q-TOF-MS/MS revealed that Pn3G5G and Cy3G5G were the main anthocyanins ofthe pink and reddish-purple fl owers of BKTPs, which were undetectable in white fl owers. The fruit and seed setting ratesof BKTPs ranged from 4.00 to 26.00 per plant, and 130.00 to 858.00 per plant, respectively. The content of unsaturatedfatty acids was over 90% of the total fatty acids in the seed oils of BKTPs, with α-linolenic acid as the dominant compound(42.51–54.93%). Both phenotypic and molecular data suggested that the accessions of BKTPs were a hybrid complex resultedfrom introgression between the sympatrically distributed wild species and cultivars in Baokang, which provided novel geneticresources for tree peony breeding and germplasm innovation.
Distributed Multi-agent Target Search and Tracking With Gaussian Process and Reinforcement Learning
Jigang Kim,Dohyun Jang,H. Jin Kim 제어·로봇·시스템학회 2023 International Journal of Control, Automation, and Vol.21 No.9
Deploying multiple robots for target search and tracking has many practical applications, yet the challenge of planning over unknown or partially known targets remains difficult to address. With recent advances in deep learning, intelligent control techniques such as reinforcement learning have enabled agents to learn autonomously from environment interactions with little to no prior knowledge. Such methods can address the explorationexploitation tradeoff of planning over unknown targets in a data-driven manner, streamlining the decision-making pipeline with end-to-end training. In this paper, we propose a multi-agent reinforcement learning technique (MARL) with target map building based on distributed Gaussian process (GP). We leverage the distributed GP to encode belief over the target locations in a scalable manner and incorporate it into centralized training with decentralized execution MARL framework to efficiently plan over unknown targets. We evaluate the performance and transferability of the trained policy in simulation and demonstrate the method on a swarm of micro unmanned aerial vehicles with hardware experiments.
Ab initio calculations of electronic structures of SrMoO4 crystals containing F and F+ color centers
Jigang Yin,Qiren Zhang,Tingyu Liu,Xiaofeng guo,Min Song,Xien Wang,Haiyan Zhang 한국물리학회 2009 Current Applied Physics Vol.9 No.6
The electronic structures of SrMoO4 crystals containing F and F+ color centers with the lattice structure optimized are studied within the framework of the fully relativistic self-consistent Dirac–Slater theory, using a numerically discrete variational (DV-Xα) method. From the calculation, it is concluded that F and F+ color centers have donor energy level in the forbidden band. The electronic transition energies from the donor level to the bottom of the conduction band are 1.855 eV and 2.161 eV, respectively, which correspond to the 670 nm and 575 nm absorption bands. It is predicted that the 670 nm and 575 nm absorption bands originate from the F and F+ centers in SrMoO4 crystals. The electronic structures of SrMoO4 crystals containing F and F+ color centers with the lattice structure optimized are studied within the framework of the fully relativistic self-consistent Dirac–Slater theory, using a numerically discrete variational (DV-Xα) method. From the calculation, it is concluded that F and F+ color centers have donor energy level in the forbidden band. The electronic transition energies from the donor level to the bottom of the conduction band are 1.855 eV and 2.161 eV, respectively, which correspond to the 670 nm and 575 nm absorption bands. It is predicted that the 670 nm and 575 nm absorption bands originate from the F and F+ centers in SrMoO4 crystals.
Design of Memory Space Extension for Small-Scale Embedded System
Jigang Du,Hyunwoo Kim,Munsik Bae,Yong Ho Song 대한전자공학회 2015 대한전자공학회 학술대회 Vol.2015 No.11
Recently many embedded systems use 32-bit processor core. The maximum memory space that a 32-bit processor could use is often limited to 4GB. However, for an application that requires large memory space beyond 4GB, the extension of memory space is provided. In this paper, we design a memory extension scheme that could extend the address space for embedded systems. We implemented the proposed scheme on a customized FPGA board. Evaluation results show that the scheme has successfully extended the memory space up to 16GB and that 31.53% slice LUTs and 12.52% slice registers of the Xilinx Zynq FPGA resource are used in this system design.
왕지강 ( Jigang Wang ),성연식 ( Yunsick Sung ) 한국정보처리학회 2022 한국정보처리학회 학술대회논문집 Vol.29 No.1
차량에 부착된 블랙박스의 교통사고 동영상은 사고 발생시 사고를 분석하기 위한 핵심 자료로 다양하게 활용되고 있다. 교통사고 동영상을 자동으로 분류할 수 있다면, 해당 동영상의 활용도를 더욱 높일 것으로 판단된다. 본 논문에서는 텍스트 기반 교통사고 특징 추출 및 분류 방법을 제안한다. 교통사고 동영상을 변환한 JSON 파일에서 불변 특징, 정적 특징 그리고 동적 특징을 추출하고 결합하여 합성 특징을 생성한다. 마지막으로 합성 특징을 사용하여 교통사고 동영상을 분류한다.
Yahong Xue,Jigang Chen,Xin Shi,Jichuan Yao,Junting Luo 한국섬유공학회 2018 Fibers and polymers Vol.19 No.3
A new parameterized finite element model was established for self-lubricating fabric liners in different weaving patterns. This model precisely simulated the spatial configuration of woven yarns with consideration of the cross-section deformation as well as the anisotropic material characteristics of yarns by converting the principal material direction along the yarn-path. Moreover, a set of simple and universal periodic boundary equations was proposed to solve the problem of the overabundance restriction in the boundary surfaces. To verify the validation of the finite element method proposed in this paper, an experimental prediction on elastic constants of self-lubricating fabric liners was carried out. The results indicate that the finite element model can successfully predict the macro mechanical properties of self-lubricating fabric liners with periodical structures. Based on the finite element model, the distributions of stress and strain, as well as the effects of weaving type and weaving density on the elastic modulus of fabric liners were investigated in details.
Zero-Shot Transfer Learning of a Throwing Task via Domain Randomization
Sungyong Park,Jigang Kim,H. Jin Kim 제어로봇시스템학회 2020 제어로봇시스템학회 국제학술대회 논문집 Vol.2020 No.10
Deep reinforcement learning (DRL) on continuous robot control has received a wide range of interests over the last decade. Collecting data directly from real robots results in high sample complexities and can cause safety accidents, so simulators are widely used as efficient alternatives for real robots. Unfortunately, policies trained in the simulation cannot be directly transferred to real-world robots due to a mismatch between the simulation and the reality, which is referred to as ‘reality gap’. To close this gap, domain randomization (DR) is commonly used. DR guarantees better transferability in the zero-shot setting, i.e. training agents in the source domain and testing them on the previously unseen target domain without fine-tuning. In this work, we identify the positive influence of DR on zero-shot transfer between the sim-to-sim setting with an object throwing task.
Yingying Zhang,Jigang Xue,Xiaoguang Song,Qilin Zhang 한국강구조학회 2018 International Journal of Steel Structures Vol.18 No.5
This paper presents the numerical parametric analysis on the loading capacity of Channel purlins with screw-fastened sheeting, in which the eff ects of anti-sag bar and corrugated steel sheet on the ultimate capacity are studied. Results show that the setup of anti-sag bars can reduce the deformations and improve the ultimate capacity of C purlins. The traditional method of setting the anti-sag bars in the middle of the web is favorable. The changing of sheeting type, sheeting thickness and rib spacing has signifi cant eff ects on the ultimate capacity of C purlins without anti-sag bars, compared with those with anti-sag bars. The proposed design formulas are relatively consistent with the calculations of EN 1993-1-3:2006, which is diff erent from those of GB 50018-2002.