DR-NTU (Digital Repository of NTU)
Not a member yet
116018 research outputs found
Sort by
EEGaitPredict: predicting gait from high density pre-frontal EEG electrodes
No full textBackground: Stroke-induced gait impairments are one of the leading causes of reduced mobility and functional independence among survivors, with many patients experiencing abnormal walking patterns such as hemiparetic gait and foot drop. While conventional rehabilitation approaches can partially restore locomotion, they often plateau after the initial recovery phase and are resource-intensive. This has prompted growing interest in EEG-based brain-computer interface (BCI) systems to decode motor intention directly from neural activity to assist gait rehabilitation. However, most existing BCI studies rely on full-head (FH) EEG, with limited investigation into more practical configurations such as prefrontal cortex (PFC)-only EEG despite evidence on its core involvement in gait planning
Objective: The objective of this project is to investigate whether EEG signals collected exclusively from the prefrontal cortex can predict continuous lower-limb joint angles with accuracy comparable to full-head EEG. The study also explores how different walking conditions affect decoding performance,
Method: A multimodal experimental framework was developed to capture synchronized EEG and joint kinematics from 16 healthy participants. EEG signals were recorded using both a full-head 64-channel cap and an 8-channel prefrontal EEG system, while continuous joint angles were obtained using wireless goniometers. Subjects performed a 10-meter walking task under four different conditions: Free, Mindful, Slow, and Fast walking. The collected EEG data were preprocessed and used to train a deep learning model (GaitNet) to predict joint angles. Performance was evaluated using Pearson’s correlation coefficient (r-value)
Results: preliminary result points that the highest decoding performance is achieved during Free and Mindful walking conditions, with mean r-values of 0.59 and 0.591, respectively. Slow walking produced the lowest decoding performance (mean r = 0.417), while Fast walking exhibited higher variance but occasional high r-values (up to 0.802). These findings suggest that moderate-paced walking yields more consistent and decodable EEG patterns. Across all conditions, some trials exhibited poor r-values, highlighting variability in trial-level signal quality and decoding robustness.
Conclusion: Preliminary findings indicate that decoding lower-limb kinematics from EEG is most effective during moderate, self-paced walking scenarios. However, these conclusions are based solely on full-head EEG data and intra-block validation. Further analysis—including inter-condition generalization and decoding using only PFC EEG—is still underway. These upcoming evaluations are necessary to determine whether PFC EEG can effectively substitute full-head EEGBachelor's degre
Neuroscience-inspired approaches for visual‑audio multimodal visual search
No full textThis project aimed to investigate the effect of auditory stimuli on visual search in
humans through human eye-tracking experiments and to explore various visual-audio
search models to examine how a multi-sensory approach would influence visual search
efficiency and accuracy. The human experiment involved exposing the subject to
240 different trials of images and audio stimuli under seven different conditions and
20 different target categories. The eye fixations are then collected and processed to
determine the effects of auditory stimuli on visual search. The results found that,
while audio semantics has a slight effect in aiding visual search, the difference was
not significant. However, human participants heavily relied on visual cues instead of
audio cues. The model experiment aimed to explore various sound localisation models
and integrate them with the human-inspired visual search model, IVSN. The results
showed that the integration of the IVSN significantly boosted the weak performance of
the sound localisation model in our dataset, but concluded that future work needs to
reduce the hybrid model’s reliance on the IVSN representation.Bachelor's degre
Thermoelectric energy harvesting from metakaolin-based geopolymers
No full textMetakaolin-based geopolymers are being discussed as potential candidates as sustainable
construction material by acting as a supercapacitor for energy harvesting. Various research
studies suggest that pore structure, as well as circuit arrangement can lead to changes in Seebeck
coefficient of metakaolin-based geopolymers, which in turn affects their energy harvesting
performance, but have yet to be explored. This Final Year Project (FYP) aims to investigate how
these variables affect the thermoelectric properties of metakaolin-based geopolymers.
The results of the project showed a decrease in Seebeck coefficient with increasing curing
temperatures, with excessively high curing temperatures potentially showing negative effects to
the thermoelectric and mechanical properties. Furthermore, the project showcases various
possible circuit, height and cross-sectional area configurations to determine the most optimal
design for efficient energy harvestingBachelor's degre
Enhancing Zn anode stability with bioderived electrolyte additive for aqueous Zn-ion batteries
No full textTo overcome the limitations of short cycle life and poor reversibility of Zn-ion batteries, trace amounts (20 mM) of bioderived furfuryl alcohol (FAL) is introduced to the zinc sulfate (ZS) electrolyte. Density functional theory calculations, together with experimental characterization, indicate a preferential adsorption of FAL over H2O molecules on Zn electrode surfaces, which suppresses parasitic reactions and generation of irreversible by-products e.g. Zn4SO4(OH)6·xH2O. FAL also increases the zinc deposition polarization, leading to heightened nucleation overpotential and slower deposition rates associated with planar growth of Zn film, in contrast to severe dendrite formation on FAL-free Zn surfaces. Consequently, the transference number of Zn2+ increases from 0.59 to 0.88 with the introduction of FAL. Zn||Zn symmetric cells with FAL additive exhibits an extreme cycle life of 4049 h at 1 mA cm−2 and areal capacity of 1 mAh cm−2. Similarly, ZnǀǀNaV3O8∙1.5H2O (NVO) (N/P = 139) full cells with FAL exhibits a capacity retention 10 % higher, after 700 cycles at a rate of 1 A g−1. In addition, Zn||V2O5 cell (N/P = 11) fails after 400 cycles but FAL additive enabled the cell to maintain a stable discharge capacity of 163 mAh g−1 up to 1000 cycles.Nanyang Technological UniversitySubmitted/Accepted versionThis work was supported by C.Q.L’s Nanyang Assistant Professorship funding (#022081-00001)
Event-triggered adaptive control for a class of uncertain nonlinear systems
No full textThis technical note simulates and researches on the event-triggered adaptive controller for a class of nonlinear systems, based on a previously developed control framework. The study focuses on remodeling the adaptive strategies in Simulink to replicate and further understand the proposed approach.
Through a combination of adopting fixed threshold strategy, relative threshold strategy and switching threshold strategy, the event-triggered adaptive controller included measures to obtain certain unknown parameters introduced by a plant system, such that it no longer requires any input-to-state stability(ISS) assumption. The proposed control scheme could ensure convergence of tracking/stabilization error and all closed loop signals are globally bounded, and the nonlinearity in a plant system is not required to be globally Lipschitz.
The replication study shows minor deviation of results in Simulink as compared to the original article, but overall is able to produce a tracking error to reduce the deviation from the sample reference signal. However, the Relative Threshold Strategy study in this article does not show a confident replication result and would require further modelling research.Bachelor's degre
Develop a methodology to assess the impact of fiber orientation on the mechanical properties of thermoplastic composites derived from the trim waste of pre-preg
No full textThe increased application of fiber-reinforced thermoplastic composites (FRTCs) in aerospace and automotive industries has generated substantial manufacturing waste, primarily in the form of trim and prepreg scrap. This study presents a methodology to evaluate the influence of fiber orientation on the mechanical properties of recycled thermoplastic composites derived from this waste. Three fiber configurations—random (LKF–RAND), ordered discontinuous (LFC–ORD), and geometrically varied random (RND)—were fabricated using compression moulding of chopped woven carbon fiber-reinforced polyphenylene sulphide (PPS) prepreg. Specimens were subjected to mechanical testing, differential scanning calorimetry (DSC), and microscopic failure analysis. The results demonstrate that fiber alignment significantly affects tensile performance. LFC–ORD showed superior mechanical strength, while RND exhibited the highest ductility. DSC analysis confirmed material stability with consistent melting
transitions near 285 °C and moderate crystallinity. Failure modes were correlated to internal flaws, fiber breakage, and matrix delamination. This study reinforces the potential of fiber orientation control in enabling the reuse of composite waste for secondary structural applications while promoting sustainability in composite manufacturing.Bachelor's degre
Mobile app for virtual fashion assistant
No full textStudents often experience high levels of stress in university due to heavy workloads, academic
deadlines, and social pressures. They cope with this stress in various ways, including self-care
routines such as dressing well and expressing themselves through fashion. Style can be a
powerful tool in boosting self-confidence and improving mood. Supporting students in making
quick, thoughtful fashion choices can contribute positively to their mental well-being and
reduce the stress of daily decision-making.
The aim of this project is to develop a virtual fashion mobile assistant that helps NTU students
and staff streamline their outfit planning and express their personal style with ease. The app
centralises wardrobe management, outfit recommendations, and shopping suggestions by AI
chatbot. It also encourages sustainable and mindful fashion choices.Bachelor's degre
Exploring acoustic sensing as a motor feedback alternative for implantable drug delivery systems
No full textTraditional approaches typically rely on high-precision embedded sensors, which,
although effective, often require invasive modifications that complicate mechanical
design and increase system complexity. In this dissertation, a non-invasive acoustic
sensing method is proposed for real-time monitoring of drug delivery devices without
altering their structural integrity.
The proposed framework employs a high-sensitivity MEMS microphone to capture
acoustic signals generated during device operation. Through comprehensive signal
processing techniques, the method effectively extracts informative features associated
with motor and rotor activity. Experimental validation was conducted using multiple
gear motors across a range of operating voltages, and ground-truth comparisons were
established through synchronized slow-motion video recording. The results
demonstrate that the dominant acoustic frequencies strongly correlate with device
actuation conditions, allowing accurate inference of operational states.
Furthermore, simulation experiments involving silicone embedding, which
approximates the mechanical properties of human tissue, confirm that the system
retains high reliability and accuracy under near-realistic subcutaneous conditions,
despite some signal attenuation. Overall, this study highlights the significant
advantages of acoustic monitoring — particularly its non-intrusiveness, low cost, and
robustness — and positions it as a promising alternative to conventional sensor-based
approaches for future biomedical device supervision.Master's degre
The Inn Between: Crafting a narrative-driven game from a game design and user-experience design perspective
No full textThe Inn Between is a three-dimensional (3D) story-based game about cleaning up and moving on. This report aims to cover the development journey of The Inn Between, with a particular focus on crafting a narrative-driven game from a game design and user-experience design perspective. This includes the project’s core game design, detailing iterations of the core game loop and game mechanics, user experience and user interface design, including work within the game’s engine.Bachelor's degre
Enhanced electrokinetic energy harvesting by selective ion sweeping in microfluidic channels
No full textElectrokinetic energy harvesting (EKEH) within microfluidic channels represents a novel
method for transforming ambient mechanical energy into electrical power, presenting
significant potential for sustainable energy systems. This research examines the
optimization of microfluidic systems to improve the efficiency of electrokinetic energy
harvesting by analyzing fluid dynamics, electrode designs, and ion transport. The research
integrates experimental techniques and COMSOL simulations to analyze the impact of
critical factors, including ion concentration, electric field intensity, and electrode
positioning, on energy conversion efficiency. Results demonstrate that optimizing these
parameters can enhance energy extraction, especially via the selective ion sweeping
method. The research emphasizes the problems encountered throughout the fabrication
process, including substrate variability and alignment accuracy, and suggests viable ways
to address these issues. The paper also addresses future endeavors, encompassing
improvements in electrode materials, scalability, and automated fabrication methods, with
the objective of connecting laboratory-scale demonstrations to practical applications. This
research advances the development of low-power devices and provides fresh insights into
sustainable energy solutions in the realm of microfluidic systems.Bachelor's degre