Flooding is a major hazard in rural and urban areas worldwide and accounts for a large portion of life and economic losses caused by natural disasters. Non-structural measures to reduce flood damage are very economical compared to structural measures.

Accurate and rapid prediction of flood inundation extent is an important task to reduce flood risk. Non-structural measures make it possible to predict inundation areas by combining accurate flood forecast and real-time observations.

Seeded Region Growing (SRG) is widely used for flooded area extraction. The seeded region growing method requires a large amount of memory space because the entire DEM dataset is initially assigned to memory, which cannot be implemented with the existing PC’s memory space. Furthermore, the limited use of computer memory space prevents the implementation of computational continuity during flood inundation analysis. When there is a large area of inundation extent and a large amount of DEM data, many recursive operations required by the seeds region growing algorithm reduce computational efficiency and can cause stack overflow due to the large data. For this reason, the use of seeded region growing algorithm applied to flood inundation prediction is limited.

Ko Song Hwan, a researcher at the Institute of Information Technology, proposed an isosurface region growing method to reduce floodplain extraction time using high resolution DEM, and a locally detailed grid SRG method to solve the problem of different inundation area by different DEM resolutions.

On a contour map with small contour intervals, if all the intersected points on the two adjacent contour lines are submerged, it can be seen that the isosurface between those contours is completely submerged. The isosurface region growing method can reduce the time for flood inundation extent extraction more than the seeded region growing method as it extends inundation areas by isosurface along the contour tree. The locally detailed grid SRG method divides the cells intersecting with top contours of the terrain barrier into small size cells. Then, a seed point is selected on the isosurface which is not completely submerged and the seeded region growing method is applied. This method can solve the problem of distorted diffusion of inundation areas around the terrain barrier.

Comparing the inundation area extraction time by the seed area growth method with that by the contour area growth method and the detailed grid SRG method, he found that the proposed method consumes less computation time.

His paper “Flood inundation extent extraction using isosurface region growing and locally detailed grid SRG” was presented at “2025 10th International Conference on Intelligent Information Technology (ICIIT 2025)”.

Moving packed bed drying can reduce the costs of transportation and storage of products. Grain dryers can be categorized into fixed bed dryers, fluidized bed dryers, rotary kilns and moving bed dryers.

A packed moving bed dryer using the air of low temperature (100℃) was developed in order to get large amounts of grain with required moisture in a given time.

Jon Chol Jin, a section head at the Faculty of Thermal Engineering, has proposed a steady moisture equation for analysing a grain drying process at a packed moving bed, and carried out a numerical simulation based on the governing equations of the grain energy equation.

The simulation results showed that the grain moisture grows lower when the moving velocity of the packed bed becomes lower.

The present model can be applied to predicting the positional distribution of physical quantities such as temperatures of air and grains as well as the relative humidity of air and the moisture content of grains in an arbitrarily-shaped packed moving grain dryer.

For further details, please refer to his paper “Steady State Analysis for Heat and Mass Transfer of Moving Grain in Cylindrical Dryer” in “Proceedings of KUTIC-2025”.

Tracking control is a very important and practical problem in the motion control of mobile robots. Tracking control can be divided into trajectory tracking control and path tracking control. In trajectory tracking control, the motion path of a mobile robot is given as a function of time, and in path tracking control, it is given by easily realizable geometrical parameters such as arcs. Trajectory tracking control is necessary when a robot should be in a certain position at a certain time. Path tracking control is appropriate when a robot has to move along the path given by the geometrical parameters at a constant speed.

Several control methods have been proposed to realize trajectory tracking control of mobile robots. In the previous literature, there has been a lot of research on the methods for achieving trajectory tracking control of robots by improving the speed control performance of individual actuators of mobile robots, but there is no report on improving the trajectory tracking performance of differential-driven mobile robots by compensating the errors of actuators.

Kim Yong Il, a section head at the Faculty of Mechanical Science and Technology, has proposed a method of designing a cross-coupled controller to improve the trajectory tracking performance by compensating the errors of actuators.

He implemented the PID control of individual motors and the cross-coupling control for error compensation to reduce trajectory tracking errors.

To verify the effectiveness of the proposed method, he carried out a simulation by MATLAB/Simulink. The experimental results showed that the trajectory tracking error could be significantly reduced by the proposed method.

You can find the details in his paper “Trajectory Tracking Control of a Differential-Driven Mobile Robot using a Cross-Coupled Controller” in “Proceedings of KUTIC-2025”.