KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 7 of 7

Some of the metrics are blocked by your
consent settings
2D and 3D LiDAR with CNN Models for Detecting Sediment Accumulation Underground after Disasters
(2025-01-01)
Krungseanmuang, Woranidtha
;
Morita, Fuka
;
Chaowalittawin, Vasutorn
;
Sathaporn, Posathip
;
Kanamori, Chisato
Global climate change impacts all regions and leads to natural disasters such as typhoons, which cause destruction, debris, and flooding. Postdisaster restoration is a very important activity that is mostly done manually and can be time-consuming and challenging, especially in subterranean environments owing to accumulated objects such as pipes, pillars, and mud distributed in confined underground areas. Therefore, in this study, we aim to utilize emerging AI technologies by comparing deep learning algorithms and evaluating four models for 2D object detection and four for 3D point cloud segmentation for detecting sediment accumulation and navigating around obstacles in underground areas after a disaster. Additionally, a custom dataset was developed to simulate underground disaster scenarios. As a result, the You Only Look Once version 11 (YOLOv11) model achieved the highest mean average precision 50 (mAP50: 91.1%) for general detection within the pillar-pipe dataset, whereas the YOLOv12 model performed the best in detecting pipes (mAP50: 87.7%). In the mud dataset, the YOLOv8 segmentation (YOLOv8-seg) model demonstrated superior performance with mAP50 scores of 93.0% (detection) and 86.4% (segmentation). For 3D point cloud segmentation, PointNet achieved the highest accuracy (98.61%), whereas RandLA-Net was optimal for pipe segmentation, achieving an intersection over union score of 37.1%. These findings highlight AI’s potential to accelerate disaster recovery, reduce manual labor, and ensure faster cleanup. Integrating deep learning models into post-typhoon restoration efforts can enable communities to recover more quickly and efficiently after climate change impacts or disaster events.
Some of the metrics are blocked by your
consent settings
High performance periodical paths tracking of two-dimensional handheld micromanipulator using explicit model predictive control
(2017-02-18)
Boksuwan, Sungwan
;
Cheypoca, Thepjit
;
Chesof, Abdulhafiz
;
Kanamori, Chisato
;
Panaudomsup, Sumit
The paper proposes an explicit model predictive control framework for a two-dimensional handheld micromanipulator to track periodical reference paths within microscopic working range and for all holding angles of users' hand. The control objective is to achieve a minimum tracking error and a wide bandwidth. The micromanipulator's design is oriented to such bio cell treatments. Its structure consists of two flexible decoupling links, each of which is driven by the combination of electric coils and permanent magnets as an actuator. In the experiments, the proposed framework is compared with two types of PID controllers, parameters of which are designed by Skogestand's IMC and IMC method based on the underdamped transfer function, respectively. The experimental results exhibit the effectiveness of the two-dimensional handheld micromanipulator controlled by the model predictive control framework over the PID controller ones.
Some of the metrics are blocked by your
consent settings
Approximation of linear gain slope equalizer using Bernstein-Stancu polynomials
(2017-02-08)
Chutchavong, Vanvisa
;
Janchitrapongvej, Kanok
;
Nomura, Hideyuki
;
Kanamori, Chisato
;
Aoyama, Hisayuki
This paper presents a design of linear gain slope equalizer for correcting the linear gain distortion. It is based on the Bernstein-Stancu polynomials, where the well-known and readily available approximated of the desired transfer function are applied. As it is known that the Bernstein-Stancu polynomials have the several advantages. For example, there is a flexible parameter as a that can be used to adjust the magnitude response for the best result. If the parameter a equals to zero, it becomes to the classical Bernstein polynomial. In addition, the phase response is linear. As the results, the proposed method is capable of designing linear gain slope equalizer which is also shown to be efficient performance without degrading its phase characteristics. The stability of the approximated of the transfer function can guarantee with Mihailov's criterion.
Some of the metrics are blocked by your
consent settings
New crossover network based on q-bernstein polynomial
(2016-01-01)
Sookcharoenphol, Dolchai
;
Nomura, Hideyuki
;
Kanamori, Chisato
;
Aoyama, Hisayuki
;
Janjitrapongvej, Kanok
This paper presents a new audio crossover network based on q-Bernstein polynomial which can be adjustable attenuation in stop-band by varying parameters of the transfer function. The crossover network gives an in-phase response and a flat summed of magnitude response of each filter through audio frequency band. The simulation results shown that the crossover network given high attenuation in stop-band which better than conventional crossover network at second-order and fourth-order. Moreover, the crossover network structure can be reduced complexity by based on subtractive circuits. Therefore, the proposed crossover network provided a high quality crossover network and can useful for professional reproduction sound systems.
Some of the metrics are blocked by your
consent settings
Portable ECG real time signal monitoring device
(2016-01-01)
Purahong, Boonchana
;
Tongkrairat, Somsin
;
Nomura, Hideyuki
;
Kanamori, Chisato
;
Aoyama, Hisayuki
This paper introduces about the application of Embedded system for analyzing heart rate signal in ECG (Electrocardiography) format using Microcontroller. Microcontroller must process signal in DSP method in this case the heart rate signal must be processed in real- Time. Our system can analyze heart rate to find the frequency of heart beat and show ECG signal and heart rate on display of Microcontroller board. To obtaining heart rate signal we use electrode placed on body in lead 2 orientations. We decide to make our system portable and can run on battery all over a day for recording signal to analyze during a day.
Some of the metrics are blocked by your
consent settings
Tremor suppression for handheld micromanipulator using robust hybrid control
(2014-12-16)
Boksuwan, Sungwan
;
Benjanarasuth, Taworn
;
Kanamori, Chisato
;
Aoyama, Hisayuki
This paper proposes a suppression of tremor effects, fluctuations caused by a user's hand, on a handheld micromanipulator by means of a robust hybrid control scheme. The proposed control scheme is a combination between an explicit model predictive control and a PID controller. The advantage of the PID controller is a robustness. It is suitable for handling the holding angle changes. The explicit model predictive control provides the excellent tracking performance. In addition, the robust hybrid control can handle tremor effects successfully. The control objective is to achieve robust tracking performance, to dampen the vibration of the mechanism and to suppress a disturbance, a tremor. The experimental results are used to investigate the effectiveness of the proposed method.
Some of the metrics are blocked by your
consent settings
Robust hybrid control for two-dimensional handheld micromanipulator
(2014-01-01)
Boksuwan, Sungwan
;
Benjanarasuth, Taworn
;
Kanamori, Chisato
;
Aoyama, Hisayuki
This paper proposes a two-dimensional handheld mi-cromanipulator oriented toward bio cell handling. The micromanipulator consists of two flexible links, each of which is constructed with a parallel beam structure. Electric coils and permanent magnets are used to produce an actuator to form double drivers. An explicit model predictive control combined with a PID controller called a robust hybrid control is proposed not only to achieve robust tracking performance, but also to dampen the vibration of the mechanism. The experimental results are compared to results from a standard PID controller to investigate the effectiveness of the micromanipulator.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results