Vibration Modeling/Analysis

As fundamental research, vibration principles and finite element-based numerical and experimental vibration modeling and analysis are studied with emphasis on the fluid-structure coupled system and uncertainty and inverse problems (structural system identification, model updating, and fault detection). By combining with data-driven AI algorithms, this vibration analysis works as a strong building block for machine diagnosis and human/machine health monitoring.  

Acoustic Event Recognition

 

Targeting robust and adaptable acoustic event recognition against harsh listening conditions, e.g., reverberation and background noises, we focus on sound classification, localization, and separation incorporated with AI. All methodologies strongly rely on the acoustic domaizn knowledge of the involved acoustic phenomena, which can differentiate the outcomes from the other state-of-the-art acoustic recognition neural networks. As a basis, an in-depth functional analysis of the human audiology/vocal system is carried out in terms of frequency-time characteristics. 

Biometric Recognition by Human Body Vibration

Non-invasive health sensors measure blood pressure, glucose level, skin, muscle, and bone conditions by means of human body vibration measurement, modeling, and analysis. Intensive time-frequency signal processing, machine learning scheme, and coupled-field finite element analysis are utilized interdisciplinarily. Wearables, continuous monitoring, biometric authentication, and cosmetology applications are the expected outcomes of this research. 

3D Shape and Motion Measurement

Non-contact 3D shape and motion sensors and core technologies are studied as a building block for the understanding of human/object conditions. This group's novel optical system and signal processing provide the compact, low-cost Lidar system to autonomous vehicles and robots. Vibration amplifier system is developed for the non-contact, sensor-free measurement of vibrations. Color-depth fusion is utilized for robust 3D measurement focusing on the optical system knowledge and AI. 

Prognostics and health management (PHM)

Prognostics and health management (PHM) is a multifaceted discipline that protects the integrity of components, products, and systems of systems by avoiding unanticipated problems that can lead to performance deficiencies and adverse effects on safety. Prognostics is the process of predicting a system’s remaining useful life (RUL). By estimating the progression of a fault given the current degree of degradation, the load history, and the anticipated future operational and environmental conditions, PHM can predict when a product or system will no longer perform its intended function within the desired specifications. Human laboratory pursues to develop the methodology that diagnoses health and predicts the remaining useful life (RUL) of engineered systems in real-time. This research can lead to maximizing facility availability and reducing maintenance costs. It can make stable facility operations by minimizing the occurrence of failures.