http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
An Adaptive Approach based on Multi-State Constraint Kalman Filter for UAVs
Hoang Viet Do,Yong Hun Kim,Yeong Seo Kwon,San Hee Kang,Hak Ju Kim,JinWoo Song 제어로봇시스템학회 2021 제어로봇시스템학회 국제학술대회 논문집 Vol.2021 No.10
High accurate and robust motion estimation plays a fundamental role in the autonomous navigation system, especially for Visual Inertial Navigation System (VINS). The consistent of extended Kalman Filter (EKF)-based estimator highly relies on the prior information of the process noise (Q) and the measurement noise (R) covariance matrices. In this paper, we apply an adaptation rule to the well-known Multi-State Constraint Kalman Filter (MSCKF) using innovation and residual information for a monocular VINS that is mounted in a drone. The shortcoming of MSCKF is that the state vector grows as the number of detected features increasing. Since the vision process can give different estimated results due to various error sources such as the ambiguity or the quality of the camera, each epoch should have different weighting factors. Therefore, Q and R covariance matrices should be adapted. It is shown through the simulation that the proposed algorithm has more robustness and accuracy against the conventional approach which has fixed values of Q and R covariance matrices.
An Improvement of 3D DR/INS/GNSS Integrated System using Inequality Constrained EKF
Hoang Viet Do,Yeong Seo Kwon,Hak Ju Kim,Jin Woo Song 제어로봇시스템학회 2022 제어로봇시스템학회 국제학술대회 논문집 Vol.2022 No.11
It is well known that INS/GNSS integrated system is either unavailable or unreliable in high-rise buildings environment. This study proposes a novel framework to fuse Odometer, INS, and GNSS to provide robust pose estimation for the mentioned challenge. Motivated by the disadvantage of the recent development of 3D DR/GNSS, we relax its assumption where velocities and accelerometer biases are estimated without sensor pre-calibration. In particular, the traditional INS/GNSS and DR/GNSS are augmented into a single system without conflict to perform EKF. Moreover, inequalities-constrained EKF is derived based on the characteristic of the presented system to increase the robustness. This constraint exploits an empirical observable where the position estimation of the odometer is considered more accurate than INS since it only requires one-step integration. The proposed approach is validated through an author-designed Unreal Engine challenging map with the AirSim plugin of an autonomous ground vehicle. The results show a significant accuracy improvement in which the position and velocity error have been reduced respectively 68% and 39% on average over a 0.81km driving.
Do, Thi Huyen,Dao, Trong Khoa,Nguyen, Khanh Hoang Viet,Le, Ngoc Giang,Nguyen, Thi Mai Phuong,Le, Tung Lam,Phung, Thu Nguyet,Straalen, Nico M. van,Roelofs, Dick,Truong, Nam Hai Asian Australasian Association of Animal Productio 2018 Animal Bioscience Vol.31 No.5
Objective: In a previous study, analysis of Illumina sequenced metagenomic DNA data of bacteria in Vietnamese goats' rumen showed a high diversity of putative lignocellulolytic genes. In this study, taxonomy speculation of microbial community and lignocellulolytic bacteria population in the rumen was conducted to elucidate a role of bacterial structure for effective degradation of plant materials. Methods: The metagenomic data had been subjected into Basic Local Alignment Search Tool (BLASTX) algorithm and the National Center for Biotechnology Information non-redundant sequence database. Here the BLASTX hits were further processed by the Metagenome Analyzer program to statistically analyze the abundance of taxa. Results: Microbial community in the rumen is defined by dominance of Bacteroidetes compared to Firmicutes. The ratio of Firmicutes versus Bacteroidetes was 0.36:1. An abundance of Synergistetes was uniquely identified in the goat microbiome may be formed by host genotype. With regard to bacterial lignocellulose degraders, the ratio of lignocellulolytic genes affiliated with Firmicutes compared to the genes linked to Bacteroidetes was 0.11:1, in which the genes encoding putative hemicellulases, carbohydrate esterases, polysaccharide lyases originated from Bacteroidetes were 14 to 20 times higher than from Firmicutes. Firmicutes seem to possess more cellulose hydrolysis capacity showing a Firmicutes/Bacteroidetes ratio of 0.35:1. Analysis of lignocellulolytic potential degraders shows that four species belonged to Bacteroidetes phylum, while two species belonged to Firmicutes phylum harbouring at least 12 different catalytic domains for all lignocellulose pretreatment, cellulose, as well as hemicellulose saccharification. Conclusion: Based on these findings, we speculate that increasing the members of Bacteroidetes to keep a low ratio of Firmicutes versus Bacteroidetes in goat rumen has resulted most likely in an increased lignocellulose digestion.
Hoang Anh Tran,Hoang Viet Do,Jin Woo Song 제어로봇시스템학회 2020 제어로봇시스템학회 국제학술대회 논문집 Vol.2020 No.10
The linear motor has been widely applied in industry to provide directly straight motion. The Polysolenoid Linear Motor (PLM), as its name, is one type of the synchronous linear machine. Toward the control problem of the PLM, state observation plays a crucial role due to the lack of measurement. In this paper, an Adaptive Unscented Kalman Filter (AUKF), which can provide reliable information of system state including applied current as well as position and velocity, is proposed. Furthermore, our observer can deal with the uncertainty load force and unknown unbias measurement noises adaptively, which contributes to robust and effective control of PLM with uncertain load condition. A scenario will be made to test the robutness of the algorithm under the value variation of measurement noise covariance. The performance of the system is verified by simulation in an illustrative example.
Hoang Viet Hai,Do Tu Anh,Tran Anh Tuan,Lanos Christophe,Mélinge Yannick 한국유변학회 2023 Korea-Australia rheology journal Vol.35 No.3
This paper presents an experimental investigation of the effect of electro-osmosis on lubricating the interface between the cement paste-based material and metal wall. A new experimental apparatus was developed and set up in this study. Two scales of cement paste-based materials were used and tested: cement paste and mortar. Tests performed were as follows: (i) range of potential difference varies from 5 to 30 V; (ii) range of metallic plate slope varies from 7° to 15°. The pre-movement time was reduced and the sample velocity was increased by increasing the potential difference and the slope. The rheological properties of two mixtures were determined to identify the characteristics of the fluid film at the interface that plays an important role in lubricating the sample. The permeability coefficient for managing the contact lubrication was also determined in this study.
Compounds from the aerial parts of Piper bavinum and their anti-cholinesterase activity
Hoang Viet Dung,TODAO CUONG,Nguyen Minh Chinh,Do Quyen,김정아,변정수,우미희,Jae Sui Choi,민병선 대한약학회 2015 Archives of Pharmacal Research Vol.38 No.5
A new alkenylphenol, bavinol A (1), togetherwith six known compounds (2–7) were isolated from theaerial parts of Piper bavinum (Piperaceae). The chemicalstructures of these compounds were determined by spectroscopicanalyses including 2D NMR spectroscopy. Theanti-Alzheimer effects of compounds 1–7 were evaluatedfrom acetylcholinesterase (AChE) and butyrylcholinesterase(BChE) inhibitory activity assays. Bavinol A (1),ampelopsin (3), and violanthin (4) exhibited AChE inhibitoryactivities with IC50 values of 29.80, 59.47 and79.80 lM. Compound 1 also showed the most potentBChE inhibitory activity with an IC50 value of 19.25 lM.
FGW-FER: Lightweight Facial Expression Recognition with Attention
Huy-Hoang Dinh,Hong-Quan Do,Trung-Tung Doan,Cuong Le,Ngo Xuan Bach,Tu Minh Phuong,Viet-Vu Vu 한국인터넷정보학회 2023 KSII Transactions on Internet and Information Syst Vol.17 No.9
The field of facial expression recognition (FER) has been actively researched to improve human-computer interaction. In recent years, deep learning techniques have gained popularity for addressing FER, with numerous studies proposing end-to-end frameworks that stack or widen significant convolutional neural network layers. While this has led to improved performance, it has also resulted in larger model sizes and longer inference times. To overcome this challenge, our work introduces a novel lightweight model architecture. The architecture incorporates three key factors: Depth-wise Separable Convolution, Residual Block, and Attention Modules. By doing so, we aim to strike a balance between model size, inference speed, and accuracy in FER tasks. Through extensive experimentation on popular benchmark FER datasets, our proposed method has demonstrated promising results. Notably, it stands out due to its substantial reduction in parameter count and faster inference time, while maintaining accuracy levels comparable to other lightweight models discussed in the existing literature.