A5019600243- Evacuation and Crowd Dynamics
- Real-time simulation and control systems
- Video Surveillance and Tracking Methods
- Industrial Technology and Control Systems
- AI and Big Data Applications
- Face recognition and analysis
- Fuel Cells and Related Materials
- Belt Conveyor Systems Engineering
- Scientific Research Methodologies and Applications
- Indoor and Outdoor Localization Technologies
- Machine Fault Diagnosis Techniques
- Iterative Learning Control Systems
- Advanced Battery Technologies Research
- Electric and Hybrid Vehicle Technologies
- Energy Efficient Wireless Sensor Networks
- Face and Expression Recognition
- Economic and Technological Systems Analysis
- Advanced Decision-Making Techniques
- Hydraulic and Pneumatic Systems
- Advanced Control Systems Optimization
- Underwater Vehicles and Communication Systems
- Fault Detection and Control Systems
- Elevator Systems and Control
- E-commerce and Technology Innovations
- Impact of AI and Big Data on Business and Society
North China University of Technology
2024
Wenzhou University
2021-2022
Jilin University
2021-2022
Fujian Polytechnic of Information Technology
2021
The learning-based model predictive control (LB-MPC) is an effective and critical method to solve the path tracking problem in mobile platforms under uncertain disturbances. It well known that machine learning (ML) methods use historical real-time measurement data build data-driven prediction models. (MPC) provides integrated solution for systems with interactive variables, complex dynamics, various constraints. LB-MPC combines advantages of ML MPC. In this work, technique summarized,...
Automation of bucket-filling is crucial significance to the fully automated systems for wheel loaders. Most previous works are based on a physical model, which cannot adapt changeable and complicated working environment. Thus, in this paper, data-driven reinforcement-learning (RL)-based approach proposed achieve automatic bucket-filling. An algorithm Q-learning developed enhance adaptability autonomous scooping system. A nonlinear, non-parametric statistical model also built approximate real...
Accurate prediction of the throttle value and state for wheel loaders can help to achieve autonomous operation, thereby reducing cost accident rate. However, existing methods based on a physical model cannot accurately reflect operator’s driving habits interaction between environment. In this paper, deep-learning-based is developed predict by learning from data. Multiple long–short-term memory (LSTM) networks are used extract temporal features different stages during operation loader. Two...
Predicting evacuation travel time in staircases is crucial to improve the emergency efficiency super high-rise buildings. Given that simulation and simplified theoretical methods used previous studies of building have defects, this study proposes a novel prediction method by using artificial neural networks. Firstly, data 71-storey office are analyzed for feature engineering. Secondly, three types networks, including convolutional RNN-based model (the traditional recurrent long-short term...
Abstract Aiming at the problems of low identification accuracy and long time in traditional proton exchange membrane fuel cell model parameter method, a method based on improved Harris hawks particle swarm optimization algorithm is proposed. The main technical parameters are determined through capacity, internal resistance, discharge depth power cell, simulation established under Matlab / Simulink, introduced to improve algorithm, used identify cell. results show that proposed has high short time.