• 한국지능시스템학회논문지
• /
• 제20권4호
• /
• pp.586-591
• /
• 2010

본 논문에서는 SURF와 Particle filter를 이용한 이동로봇의 위치 추정 방법을 제안한다. 제안한 방법은 다음과 같다: 먼저, 이동 로봇의 위치를 찾기 위해 SURF 알고리증을 이용하여 카메라로부터 획득한 영상을 분석한다. 두 번째, 획득한 영상으로부터 이동로봇의 상대적인 위치를 알기 위해 이동로봇에 설치되어 있는 초음파 센서를 이용하여 주변 환경과의 거리를 측정한다. 그리고 측정된 센서 값들을 기반으로 하여 이동 로봇의 위치를 추정하는데 있어서 오차를 줄이고자 위치 추정에 많이 사용되는 Particle filter를 이용하여 이동 로봇의 위치를 추정한다. 마지막으로, 본 논문에서 제안한 방법은 실험을 통해 그 응용 가능성을 증명한다.

• 한국환경보건학회지
• /
• 제27권4호
• /
• pp.15-24
• /
• 2001

The objective of this study it to evaluate the microbial safety in rapid sand filters adapted in most drinking water treatment plants. The potential pathogens to cause water quality problems Are presumed to be Giardia and Gryptosporidium. They look like particles in view of their size. It has been reported that if the number of particles (larger than 2 ${\mu}{\textrm}{m}$ in water) is less than 100 per mL and its turbidity is below 0.1 NTU, it is considered as a safe water in terms of pathogens. In order to achieve such a good water quality. filter-aids (chemicals) were added to the inlet-channel of filter and their effectiveness was evaluated on the basis of water quality factors such as turbidity and particle counting. This study was conducted in she three steps of experiment: jar test. pilot plant test and real water treatment plant test (P plant in seoul). The experiment reult of the P water treatment showed that cationic polyamine was the most effective in the removal of particles and turbidity at the does of 0.25 mg/L. The turbidity without filer-aids showed in the range of 0.12 ~0.17 NTU during filtration and 0.14 NTU on the average. However. with addition of polyamine, the turbidity represented below(or less than) 0.1 NTU after 20 min in the start of filtration and kept 0.08 NTU on the average. On the other hand, as for number of particles, while no filter-aids led to the range of 111 ~270 per mL and 190 on the average, addition of polydmine led to 113 per mL on talc average, and kept below100 per mL after 20 min in the start of filtration.

• 한국자동차공학회논문집
• /
• 제4권5호
• /
• pp.179-186
• /
• 1996

The magnetic fibrous polymeric oil filter for automotive was obtained by pneumoextrusion processing prepared from thermoplastic polymer(polyamide) containing magnetic particulate filer (Ba ferrite), and treated subsequently in a magnetic fields. Using an oil filter tester and particle quantifier, the dependence of filtrational characteristics of the magnetic oil filter on the parameters of mean pore sizes and magnetic properties were analyzed. Also experimental engine-bed test of oil filters was investigated. The magnetic fibrous polymeric filter was shown to possess a highly filtration efficiency in filtering the fine metal particles with increasing the magnetic force of filter element. In this study, we knew that efficiency of magnetic fibrous polymeric filter was dependent on the magnetic strength of the filter elements.

• 환경위생공학
• /
• 제24권3호
• /
• pp.16-27
• /
• 2009

In this paper, I examined the level of fine dust in medical institutions, educational institutions and multi-purposed facilities to grasp the exact state of the present, and decided the level of air-borne particulate(KSM ISO Standard and ISO Standard 14644-1). We compared new proposed cleaner equipped with HEPA Filter with general cleaner and analyzed the rate of removal according to height, air volume and the equipment with the compulsive air intake. Through this comparison, I reached the conclusion as follows: 1. According to the examination, the fine dust of medical institutions, educational institutions and multi-purposed facilities in Kwang Ju is class 9. 2. The filter used in general cleaner on the market is that of HEPA-type, and its removal efficiency for fine particles($0.3{\sim}0.5{\mu}m$) is very low. 3. In the removal efficiency of new proposed cleaner equipped with HEPA Filter, the higher it is, the better, especially more than 180cm in height. 4. In case it is operated for 5 minutes under the condition of the space of $9.4m^{3}$ and the maximum air volume equipped with two induction pipes, we can keep the air cleanness level of 5 ~ 6. 5. To maintain the air cleanness for a long time, if we first operate for 5 minutes at maximum air volume and then operate at medium maximum air volume, we can keep the air cleanness with low energy.

• 한국분무공학회지
• /
• 제12권2호
• /
• pp.86-93
• /
• 2007

There is a growing interest to develop an eco-friendly air cleaner with high performance through a remanufacturing process. Two kinds of polyurethane filter media, a coarse (Filter-A) and a fine filter media (Filer-B), are used in this study to protect a vehicle engine from airborne particles. In order to improve the collection performance of the filters (Filter-A, Filter-B), an oil coating technology on the filter surface was introduced. As a result, inertial force is a dominant collection mechanism for a dry filter media, so that collection efficiency increases with increasing filtration velocity. However, intra-structure change of an oil-coated filter media influences on the collection mechanism, which shows a non-linear collection efficiency curve in terms of filtration velocity. The result shows that the developed filter media are eco-friendly and effective to protect a vehicle engine from airborne particles especially at low filtration velocity.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권10호
• /
• pp.4835-4855
• /
• 2018

In this paper, we study the sparse signal recovery that uses information of both support and amplitude of the sparse signal. A convergent iterative algorithm for sparse signal recovery is developed using Majorization-Minimization-based Non-convex Optimization (MM-NcO). Furthermore, it is shown that, typically, the sparse signals that are recovered using the proposed iterative algorithm are not globally optimal and the performance of the iterative algorithm depends on the initial point. Therefore, a modified MM-NcO-based iterative algorithm is developed that uses prior information of both support and amplitude of the sparse signal to enhance recovery performance. Finally, the modified MM-NcO-based iterative algorithm is used to estimate the time-varying sparse wireless channels with temporal correlation. The numerical results show that the new algorithm performs better than related algorithms.