A5000828711- EEG and Brain-Computer Interfaces
- Virtual Reality Applications and Impacts
- Robotic Path Planning Algorithms
- Functional Brain Connectivity Studies
- Neural and Behavioral Psychology Studies
- Smart Grid Energy Management
- Advanced Technologies in Various Fields
- Distributed Control Multi-Agent Systems
- Action Observation and Synchronization
- Free Will and Agency
- Retinal Imaging and Analysis
- Urban Stormwater Management Solutions
- Infrastructure Maintenance and Monitoring
- Control and Dynamics of Mobile Robots
- Adaptive Dynamic Programming Control
- Reinforcement Learning in Robotics
- Asphalt Pavement Performance Evaluation
- Guidance and Control Systems
Xi’an Jiaotong-Liverpool University
2022-2025
University of Liverpool
2023
Guangzhou University
2019-2021
Chang'an University
2021
The imbalanced development between deep learning-based model design and motor imagery (MI) data acquisition raises concerns about the potential overfitting issue— models can identify training well but fail to generalize test data. In this study, a Spatial Variation Generation (SVG) algorithm for MI augmentation is proposed alleviate issue. essence, SVG generates using variations of electrode placement brain spatial pattern, ultimately elevating density raw sample vicinity. prevents from...
The increasing number of dispersed EEG dataset publications and the advancement large-scale Electroencephalogram (EEG) models have increased demand for practical tools to manage diverse datasets. However, inherent complexity data, characterized by variability in content metadata, data formats, poses challenges integrating multiple datasets conducting model research. To tackle challenges, this paper introduces EEGUnity, an open-source tool that incorporates modules "EEG Parser", "Correction",...
Abstract The path planning problem of mobile robot in unknown dynamic environment (UDE) is discussed this article by building a continuous simulation environment. To achieve collision‐free UDE, the reinforcement learning theory with deep Q‐network (DQN) applied for to learn optimal decisions. A reward function designed weight balance obstacle avoidance and approach goal. Moreover, it found that relative motion between moving obstacles robots may cause abnormal rewards further lead collision...
Virtual movement augmentation, which refers to the visual amplification of remapped movement, shows potential be applied in motion-related virtual reality programs. Sense agency (SoA), measures user's feeling control their action, has not been fully investigated for augmented movement. This study effect at three different levels (baseline, medium, and high) on users' SoA using both subjective responses electroencephalography (EEG). Results show that can boosted slightly medium augmentation...
Currently, there does not exist any established test method to evaluate the long-term performance of a porous asphalt pavement against clogging, with or without maintenance cleaning. This study proposes laboratory bridge this gap. The consisted three main procedures: (i) A procedure introduce clogging materials into specimen, (ii) measure permeability coefficient and (iii) de-clogging cleaning simulating high-pressure washing. Three mixtures were studied. results showed that higher initial...
Because of "the non-repeatability the experiment settings and conditions" variability brain patterns among subjects", data distributions across sessions electrodes are different in cross-subject motor imagery (MI) studies, eventually reducing performance classification model. Systematically summarised based on existing a novel temporal-electrode distribution problem is investigated under both intra-subject inter-subject scenarios this paper. Based presented issue, bridging domain adaptation...
Users' motion representation in virtual reality (VR) can be modulated visually by introducing a mismatch with their real motion, which bring benefits to exercise and rehabilitation has great potential for exergame applications VR. experience of control is critical consideration user human–computer interaction should paid special attention when movement modulation implemented However, how affects users' motor performance not been fully investigated detail. This research included 49...
The increasing number of dispersed EEG dataset publications and the advancement large-scale Electroencephalogram (EEG) models have increased demand for practical tools to manage diverse datasets. However, inherent complexity data, characterized by variability in content metadata, data formats, poses challenges integrating multiple datasets conducting model research. To tackle challenges, this paper introduces EEGUnity, an open-source tool that incorporates modules 'EEG Parser', 'Correction',...
This paper discussed the guidance problem of autonomous mobile robots with multiple constraints: target tracking, synchronization, and obstacle avoidance. The strategy is proposed according to structure adaptive dynamic programming enable online learning optimization. An action neural network a critic are designed estimate cost function, respectively. Then, an optimal intelligent law obtained weight updating rules networks. Finally, validity scheme demonstrated simulation five tracking in...