Local obstacle avoidance is important for mobile robot’s movement.

Several methodologies have been suggested to solve path planning problems. They include mathematical programming, cell decomposition, roadmap approach and potential fields. These techniques prove to be inefficient due to their high computational cost and inaccuracy caused by the trapping in the local minimum.

A vehicle model is a highly nonlinear and time-varying system as far as obstacle avoidance algorithm is concerned. The algorithm must be able to adapt to a variety of changing environments. However, model-based control methods face limitations due to changing environment. Different heuristic techniques such as fuzzy logic systems and cluster optimization algorithms can overcome the weakness of the above algorithm.

Kim Thae Yong, a researcher at the Faculty of Automatics, proposed an obstacle avoidance algorithm based on Deep Q-Learning Network (DQN) that combined reinforcement learning and deep learning.

The inputs of DQN are the coordinates of a mobile robot and obstacles, the moving velocity, and the distance between them. The output is the steering angle of a mobile robot.

The trained DQN model includes both path planning and path tracking techniques and it is computationally fast and applicable to complex environments because the dynamic model of a robot is not used.

Using MATLAB, he built an environment model and a model of mobile robot and performed simulations comparing the proposed algorithm with other obstacle avoidance algorithms in various situations. The simulation results showed that the mobile robot successfully avoided obstacles after 25 000 iterations.

You can find details in his paper “Obstacle Avoidance Control of Autonomous Mobile Robot Based on Deep Q-Learning Network” in “Proceedings of KUTIC-2025”.

Aquatic organisms such as Daphnia magna are very sensitive to water toxicity, and their swimming pathway features have some differences in normal and toxic environments. Therefore, aquatic organisms such as D. magna are generally used to scientifically analyze water toxicity, environmental pollution, etc.

Analyzing the swimming characteristics of aquatic organisms in various kinds of water to identify toxicity is important for ensuring human safety and preventing environmental pollution. In particular, tracking the real-time swimming pathway monitoring of D. magna with cameras and finding the anomalous behaviour is a practical and suitable method for quickly and directly identifying the real effects of water on living organisms without recourse to any expensive analytical instruments and samples and for saving much time and effort.

For analysing swimming trajectory of aquatic organisms such as D. magna, 3D trajectory tracking by real-time object monitoring is a priority, and a toxicity detection classifier based on trajectory analysis is needed to detect abnormal behaviour moment.

Jong Myong Guk, a section head at the Faculty of Information Science and Technology, proposed a classifier scheme to identify abnormal behavior start moment of D. magna in order to determine water toxicity.

The experiments on five toxic substances have shown that the moments at which the abnormal behavior of D. magna occurs are different depending on the concentration of toxicity, and that the proposed method could rapidly and accurately estimate even very low toxicity.

You can find more information about it in his paper “A New Method of Measuring Daphnia Abnormal Behavior Start Moment for Water Toxicity Assessment by using Three-Dimensional Pathway Features and SVM” in “Proceedings of KUTIC-2025”.

In flotation processes, flotation reagents are effective means of controlling the flotation behavior of minerals by using their surface properties. The recovery and flotation of pulp surfaces from which valuable minerals float by altering the hydrophobicity and hydrophilicity of mineral surfaces with flotation reagents is called positive line, whereas flotation of gangue minerals is called negative line.

Proper addition of reagents guarantees the optimum concentration of reagents in the pulp. Depending on the ore characteristics, reagent properties and process requirements, appropriate reagents and different reagent addition modes should be selected. By doing so, the optimum concentration of reagents in the pulp should be maintained to maximize the concentrate extraction.

Kim Thae Ung, a researcher at the Faculty of Automatics, conducted a theoretical study on multivariate statistical process control (MSPC) to improve the concentrate grade and reduce production cost in the control system design of flotation process. Based on it, he realized multivariate statistical process monitoring and built an economic technological index prediction model by applying the MSPC method to a flotation process, a typical continuous process.

The simulation results and industrial tests showed that the prediction model has high accuracy and the reagent addition method developed based on this model ensures stable flotation operation, high flotation recovery and high practical value.

For more details, you can refer to his paper “Application of Multivariate Statistical Process Control in Flotation Process” in “Proceedings of KUTIC-2025”.