A5008201587- Neural Networks and Applications
- Fault Detection and Control Systems
- Adaptive Control of Nonlinear Systems
- Fuzzy Logic and Control Systems
- Advanced Control Systems Optimization
- Control Systems and Identification
- Vibration Control and Rheological Fluids
- Advanced Algorithms and Applications
- Iterative Learning Control Systems
- Mineral Processing and Grinding
- Anomaly Detection Techniques and Applications
- Machine Learning and ELM
- Structural Health Monitoring Techniques
- Robot Manipulation and Learning
- Prosthetics and Rehabilitation Robotics
- Adaptive Dynamic Programming Control
- Reinforcement Learning in Robotics
- Hydraulic and Pneumatic Systems
- Seismic Performance and Analysis
- Neural Networks Stability and Synchronization
- Muscle activation and electromyography studies
- Fuzzy Systems and Optimization
- Face and Expression Recognition
- Water Quality Monitoring and Analysis
- Dynamics and Control of Mechanical Systems
Center for Research and Advanced Studies of the National Polytechnic Institute
2016-2025
Agricultural Information Institute
2016-2025
Chinese Academy of Agricultural Sciences
2016-2025
Institute of Information Engineering
2017-2025
Chinese Academy of Sciences
2014-2025
Xuzhou Medical College
2024-2025
Northeast Forestry University
2025
Fanjingshan National Nature Reserve
2025
Instituto Politécnico Nacional
2015-2024
The University of Sydney
2023-2024
Online vehicle routing is an important task of the modern transportation service provider. Contributed by ever-increasing real-time demand on system, especially small-parcel last-mile delivery requests, route generation becoming more computationally complex than before. The existing algorithms are mostly based mathematical programming, which requires huge computation time in city-size networks. To develop routes with minimal time, this paper, we propose a novel deep reinforcement...
A panoptic driving perception system is an essential part of autonomous driving. high-precision and real-time can assist the vehicle in making reasonable decision while We present a network (YOLOP) to perform traffic object detection, drivable area segmentation lane detection simultaneously. It composed one encoder for feature extraction three decoders handle specific tasks. Our model performs extremely well on challenging BDD100K dataset, achieving state-of-the-art all tasks terms accuracy...
It is of great significance to apply deep learning for the early diagnosis Alzheimer's disease (AD). In this work, a novel tensorizing GAN with high-order pooling proposed assess mild cognitive impairment (MCI) and AD. By three-player cooperative game-based framework, model can benefit from structural information brain. incorporating scheme into classifier, make full use second-order statistics holistic magnetic resonance imaging (MRI). To best our knowledge, Tensor-train, High-order...
This paper studies the relationship between some characteristics of corporate board and firm’s capital structure in Chinese listed firms. The findings provide preliminary empirical evidence seem to suggest that managers tend pursue lower financial leverage when they face stronger governance from board. However, results relationships are statistically significant only case composition CEO tenure. insignificant size fixed compensation. may general that, up time period our investigation,...
In general, fuzzy neural networks cannot match nonlinear systems exactly. Unmodeled dynamic leads parameters drift and even instability problem. According to system identification theory, robust modification terms must be included in order guarantee Lyapunov stability. This paper suggests new learning laws for Mamdani Takagi-Sugeno-Kang type based on input-to-state stability approach. The schemes employ a time-varying rate that is determined from input-output data model structure. Stable...
In order to minimize steady-state error with respect uncertainties in robot control, proportional-integral-derivative (PID) control needs a big integral gain, or neural compensator is added the classical proportional-derivative (PD) large derivative gain. Both of them deteriorate transient performances control. this paper, we extend popular PD into PID This novel natural combination industrial linear and compensation. The main contributions paper are semiglobal asymptotic stability local...
Long-Short Term Memory (LSTM) is a type of Recurrent Neural Networks (RNN). It takes sequences information and uses recurrent mechanisms gate techniques. LSTM has many advantages over other feedforward NNs in modeling time series, such as audio video. However, non-linear system normal does not work well(Wang, 2017). In this paper, we combine with NN, use the advantages. The novel neural model consists hierarchical networks one multilayer perceptron. We design special learning algorithm which...
Abstract Background The high cost and insufficient supply of human papillomavirus (HPV) vaccines have slowed the pace controlling cervical cancer. A phase III clinical trial was conducted to evaluate efficacy, safety, immunogenicity a novel Escherichia coli-produced bivalent HPV-16/18 vaccine. Methods multicenter, randomized, double-blind started on November 22, 2012 in China. In total, 7372 eligible women aged 18–45 years were age-stratified randomly assigned receive three doses test or...
The identification problem incorporated in feedback control of uncertain nonlinear systems exhibiting complex behavior has been solved different ways. Some these solutions have used artificial intelligence methods like fuzzy logic and neural networks. However, their individual implementation suffers from certain drawbacks, such as the black-box nature network finding suitable membership functions for systems. These weaknesses can be avoided by implementing a hybrid structure combining two...
The emotion analysis of hotel online reviews is discussed by using the neural network model BERT, which proves that this method can not only help platforms fully understand customer needs but also customers find suitable hotels according to their and affordability recommendations be more intelligent. Therefore, pretraining BERT model, a number analytical experiments were carried out through fine-tuning, with high classification accuracy was obtained frequently adjusting parameters during...
Introduction How does environmental education affect quality? There is no consensus among theorists. This paper devoted to exploring the influence mechanism of and quality under background a low-carbon economy from theoretical model empirical analysis. Methods The research method this includes two aspects. First, consideration central planner, draws on improves Ramsey Model explore interaction education, green growth. Second, uses provincial panel data China 2011 2017 for analysis, which...
In this paper the adaptive nonlinear identification and trajectory tracking are discussed via dynamic neural networks. By means of a Lyapunov-like analysis we determine stability conditions for error. Then analyze error by local optimal controller. An algebraic Riccati equation differential one used analysis. As our main original contributions, establish two theorems: first gives bound error, second establishes We illustrate effectiveness these results examples: second-order relay system...
Nonlinear system online identification via dynamic neural networks is studied in this paper. The main contribution of the paper that passivity approach applied to access several new stable properties neuro identification. conditions for passivity, stability, asymptotic and input-to-state stability are established certain senses. We conclude gradient descent algorithm weight adjustment an L(infinity) sense robust any bounded uncertainties.
In this paper we present a new simple controller for chaotic system: the Lorenz equation. The design is based on passive technique. final structure of has linear feedback form. Using method, prove stability closed loop system.