A5100755566- Piezoelectric Actuators and Control
- Advanced Vision and Imaging
- Image Processing Techniques and Applications
- Magnetic Properties and Applications
- Advanced Image Processing Techniques
- Shape Memory Alloy Transformations
- Optical measurement and interference techniques
- Image Enhancement Techniques
- Advanced Manufacturing and Logistics Optimization
- Advanced Battery Technologies Research
- Image Retrieval and Classification Techniques
- Industrial Vision Systems and Defect Detection
- Innovation and Knowledge Management
- Remote-Sensing Image Classification
- Distributed Control Multi-Agent Systems
- Advanced Image and Video Retrieval Techniques
- Dielectric materials and actuators
- Advancements in Battery Materials
- Smart Parking Systems Research
- 3D Shape Modeling and Analysis
- Cell Image Analysis Techniques
- Human Pose and Action Recognition
- Optimization and Packing Problems
- Transportation and Mobility Innovations
- Advanced Algebra and Logic
South China University of Technology
2014-2024
China Three Gorges University
2024
Shanghai International Studies University
2010
Battery management is the key technical link for electric vehicles. A good battery system can realize balanced charge and discharge of batteries, reducing capacity degradation loss health caused by overcharge discharge, which all depend on real-time accurate estimation battery’s state (SOC). However, SOC has highly complex nonlinear time-varying characteristics related to chemical physical dynamic environmental conditions, are difficult measure directly, this become a difficulty in design...
When the systems preceded by unknown hysteresis nonlinearities, existence of will degrade system performance. The main task is to develop a control approach mitigate effects caused these nonlinearities. In this paper, special case Duhem model, Coleman-Hodgdon employed describe By exploring characteristics sliding mode proposed as an illustration show method mitigating nonlinearities without constructing inverse. ensures asymptotically stability and error dynamics toward switching surface....
Abstract Lithium‐ion battery manufacturing is a multiprocess serial system and the consistency of each single cell will affect performance safety after grouping. Therefore, optimizing preparation process improving have become key technical issues in process. In this study, inconsistency finished batteries caused by analyzed from three processes: electrode preparation, assembly, liquid injection formation structure–activity relationship between data studied. Taking actual capacity,...
Recently, deep convolutional neural networks (CNNs) have emerged as powerful tools for the correspondence problem in stereo matching task. However, existence of multiscale objects and inevitable ill-conditioned regions, such textureless real-world scene images continue to challenge current CNN architectures. In this article, we present a robust scale-aware network, which aims predict disparity maps fuse them achieve more accurate map. To end, feature representations are extracted from...
Fast generalized fuzzy C-means (FGFCM) is one of the popular image segmentation algorithms. However, a single FGFCM algorithm easy to get stuck at local optimum and choices initial cluster centers affect consequences seriously. Furthermore, conventional ignores relationship between selections neighbors edges so that some details are not considered. In order solve these problems, this paper presents an improved approach using with edges-based neighbor selection strategy PSO. PSO used search...
Ionic polymer-metal composites(IPMCs) is a class of new smart materials, which can generate large bending motions under low driving voltage, and this property IPMCs be used as artificial muscle devices without the traditional components. However, one strong non-smooth nonlinearity, hysteresis, exists between its curvature applied input voltage. The hysteresis curve altered with change voltage rate, causes in rate-dependent characteristics. In paper, modified Preisach model proposed to...
Ultrasound imaging is widely used in clinical diagnosis due to its non-invasive nature and real-time capabilities. However, conventional ultrasound diagnostics face several limitations, including high dependence on physician expertise suboptimal image quality, which complicates interpretation increases the likelihood of diagnostic errors. Artificial intelligence (AI) has emerged as a promising solution enhance diagnosis, particularly detecting abnormalities across various biomedical...
With the widespread use of autonomous guided vehicles (AGVs), avoiding collisions has become a challenging problem. Addressing issue is not straightforward since production efficiency, collision avoidance, and energy consumption are conflicting factors. This paper proposes novel edge computing method based on vehicle intelligence to solve energy-efficient collision-free machine/AGV scheduling First, architecture was built, corresponding state transition diagrams for were developed. Second,...
For the systems preceded by smart actuators which exhibit hysteresis nonlinearities with saturation property, main task is to develop a control approach mitigate effects caused these saturated nonlinearities. As an illustration, adopting saturated-type Prandtl-Ishlinskii model combined play operator describe phenomena, adaptive reference proposed in this paper show access of mitigating hysteresis. The global boundedness closed-loop system and tracking desired trajectory proper precision are...
In Recent years, flexible printed circuit boards (FPC) have been widely used in consumer electronic products. The problem of automatic visual inspection defects including pinholes, scratches and particles FPC surfaces has considered this paper. A global 1-D Fourier based image reconstruction scheme that directly works on the line images instead traditional 2-D is proposed. It firstly eliminates frequency components represent periodic pattern a spectrum, then transforms domain to spatial...
Focusing on a class of pure-feedback nonlinear time-delay systems with unknown hysteresis input, an adaptive control method is discussed in this paper. For such systems, the nonaffine structure and non-smooth nonlinearities are major challenges for controller design. To deal problem mean-value theorem adopted to transform as affine form, neural network approximator utilized estimate system functions. preceded nonlinearities, Backlash-like model illustration show way dealing input. Following...
In this paper, the formation control problem of a multi-agent system is studied. The foraging behavior modeled as finite-horizon non-cooperative differential game under local information, and existence properties Nash equilibria are formations achieved in an intrinsic way sense that they only attributed to inter-agent interaction geometric network, where desired not designated beforehand. Through design individual costs network topology, regular polygons, antipodal Platonic solids while...
A rapid 3D reconstruction method for IC chips based on color fringe projection is proposed in this paper. In the defect inspection of chips, present methods can't meet requirement high speed and accuracy. paper, periodic patterns are designed projection. The relationship between pattern offset surface height object can be obtained from mathematical model system. new using duty ratio to calculate offset. Only a single image needed realize chips. Experimental results show that real time obtain...
The substantially similar texture features of sticks and shiitake mushrooms in the mushroom-growing environment make precisely labeled samples more expensive semantic segmentation challenging. In this paper, a search focus network(SFNet) for was proposed, which utilized group-reversal attention module(GRAM) to strengthen information understanding trained via transfer learning data augmentation strategies. experimental results on self-built mushroom dataset revealed that structural measure...
In recent years, the three-dimensional (3D) reconstruction of industrial products within digital twins predominantly relies on LiDAR and stereo vision technologies. However, both these methods come with stringent environmental prerequisites. To tackle challenges associated high standards, this research introduces a general end-to-end 3D system based deep learning, which directly generates models from videos products. Moreover, study refines system's performance concerning outcomes. mitigate...