|
Comparison and Analysis of Convex Optimization Methods for Potential Field-based Path Planning |
|
Yoonho Han, Heoncheol Lee(Kumoh National Institute of Technology, Korea) |
| In this paper, the comparison and analysis of two
convex optimization methods that gradient method and newton's method applied to the potential field-based path planning. The simulation results show that the convex optimization methods were conducted successfully for potential field-based path planning. Also, we found that the Newton’s method was conducted faster than the Gradient descent method in potential field. In future works, the convex optimization methods will be tested with more various types potential field environments. Besides, other convex optimization methods such as interior point method will be applied to potential field-based path planning |
|
|
Fault Detection and Diagnosis of Sensors and Actuators for Unmanned Surface Vehicles |
|
Nak Yong Ko, Gyeongsub Song(Chosun University, Korea), Hyun Taek Choi(Korea Research Institute of Ships and Ocean Engineering, Korea), Joono Sur(Agency for Defense Development, Korea) |
| This paper describes an algorithm for fault detection and diagnosis (FDD) of unmanned surface vehicles (USV). The algorithm estimates the faults in sensor measurements and actuation force on which autonomous navigation of a USV highly relies: faults in measurements of acceleration and angular rate of an inertial measurement unit (IMU) and thrust force from thrusters. The models of fault dynamics and measurements for FDD are used for the algorithm which is based on Kalman filter, specifically iterated optimal two stage extended Kalman filter. The FDD for IMU measurements estimates the fault immediately when fault takes place, while the FDD for thrust force exhibits long delay. |
|
|
Three-Dimensional Mapping of Indoor and Outdoor Environment using LIO-SAM |
|
Nak Yong Ko, Henok Tegegn Warku, Hong Gi Yeom(Chosun University, Korea), Woong Choi(Gunma College, Japan) |
| This paper describes the construction of three-dimensional mapping of the indoor and outdoor environment by using Lidar Inertia Odometry via Smoothing and Mapping (LIO-SAM) algorithm. We have tested the algorithm experimentally using Velodyne Puck (VLP-16) 3D lidar sensor with 16 channels and Xsens MTI-G-700 to get inertial measurement unit (IMU) data in order to build and visualize the three-dimensional map of the environment on Robot Operating System (ROS). From the experimental result, we have observed that the system can create a compatible 3D map with detailed information of objects inside the environment. |
|
|
Calculation of Mahjong Score using AI |
|
Takuma Kano(National Institute of Technology, Gunma College, Japan), Liang Li(Ritsumeikan University, Japan), Nak Yong Ko(Chosun University, Korea), WOONG CHOI(National Institute of Technology, Gunma College, Japan) |
| Image recognition and machine learning can be used effectively for many things, but they are not used very often. We have developed a progress support system using image recognition for mahjong, which has complicated rules. Since the mahjong tiles to be recognized are too small to be recognized, we used various methods to improve the recognition accuracy. We have also developed a system that calculates the score from the recognized mahjong tiles, so that beginners can know the score instantly. |
|
|
Posture Recognition System using Depth Sensor |
|
Shogo Sekiguchi(National Institute of Technology, Gunma College, Japan), Liang Li(Ritsumeikan University, Japan), Nak Yong Ko(Chosun University, Korea), WOONG CHOI(National Institute of Technology, Gunma College, Japan) |
| Stiff shoulders and low back pain have become more severe as national diseases in Japan. Musculoskeletal symptoms are considered to occur when daily factors. In addition, it is considered that about 20% of chronic patients had alleviated their symptoms by temporary medical treatment. Stiff shoulders and low back pain are easy to get and hard to cure. This research aimed to implement a system that proposes a posture that is able to prevent musculoskeletal disorders by deriving the degree of fatigue based on parameters using Azure Kinect DK, which is a three-dimensional depth sensor. |
|
|
Multi Functional Brain Computer Interface |
|
Woosung Choi, Hong Gi Yeom, Nak Yong Ko(Chosun University, Korea) |
| Brain-computer interface (BCI) is a technology that controls computers or machines using brain signals. To use the BCI system in daily life, the BCI should be able to predict various intentions. However, the previous BCI methods have a limitation that they can predict only one type of intention. In this paper, we propose a multi-functional BCI method that can predict various intentions simultaneously. Noise was removed from electroencephalography (EEG) data through pre-processing. After that, features that reflect user's multiple intentions are extracted using power spectrum analysis and normalization process. Finally, artificial neural networks predict multiple intentions. |
|