A5103070213- Robot Manipulation and Learning
- Robotic Process Automation Applications
- Advanced Computing and Algorithms
- Optical Systems and Laser Technology
- Advanced Image Fusion Techniques
- Text and Document Classification Technologies
- Vehicle Dynamics and Control Systems
- Education and Work Dynamics
- Simulation and Modeling Applications
- Digital Transformation in Industry
- Policing Practices and Perceptions
- Advanced Measurement and Detection Methods
- Advanced Wireless Communication Techniques
- Advanced Computational Techniques and Applications
- Geomechanics and Mining Engineering
- Infrared Target Detection Methodologies
- Robotic Path Planning Algorithms
- Manufacturing Process and Optimization
- Wireless Communication Networks Research
- Coding theory and cryptography
- Industrial Vision Systems and Defect Detection
- Explainable Artificial Intelligence (XAI)
- Advanced Algorithms and Applications
- Advanced Sensor and Control Systems
- Higher Education and Teaching Methods
Ocean University of China
2024
Shandong University
2018-2019
Suzhou Research Institute
2018
Huawei Technologies (China)
2017
Zhejiang University
2013-2017
Guangzhou Design Institute
2010
Chongqing Technology and Business University
2008
This paper investigates an adaptive method of dim small target detection in infrared images with a complex background. We analyze depth the characteristics background, target, and noise gray intensity, space frequency domain images. The modified top-hat transformation using interrelated structuring elements is adopted to adaptively detect darker brighter targets greatly suppress cluttered Lateral pattern inhibition enhances local contrast ratio simultaneously identifies interest. automatic...
Deep reinforcement learning has achieved great success in many challenging domains. However, sample efficiency and safety issues still prevent from applying deep directly robotics. Sim-to-real transfer is one feasible solution to tackle these problems address the reality gap between simulation reality. In this letter, we propose combine meta-reinforcement progressive neural network (PNN) by meta-training a policy for multiple source tasks transferring it real-world robot via PNN (MetaPNN)....
In a traditional complementary coded CDMA (CC-CDMA) system, signals are spread using the direct sequence (DS) technique to suppress multipath interference (MI) and multiple access (MAI). However, DS spreading suffers from very poor bandwidth efficiency. This paper proposes scrambling-based CC-CDMA namely scrambling (CCSMA). A CCSMA system offers unified platform integrate CC-CDMA, multiple-input multiple-output (MIMO), seamlessly. Utilizing codes in modulation, is able MI MAI effectively....
This paper describes a real-time multi-vehicle motion planning (MVMP) algorithm for the emergency vehicle clearance task. To address inherent limitations of human drivers in perception, communication, and cooperation, we require that surrounding normal vehicles are connected automated (CAVs). The concerned MVMP task is to find cooperative trajectories such can efficiently pass through ahead. We use an optimal-control based formulation describe problem, which centralized, straightforward,...
In industrial robotic assembly process, the work surrounding environment is generally described by contact state when vision-based systems fail for occluded parts. To solve recognition problem, paper builds an process model based upon Support Vector Regression (SVR) and Particle Swarm Optimal (PSO), which maps relationship between executive action. The established SVR model, whose parameters optimized PSO, used to predict next motion of robot. effectiveness accuracy on PSO are further...
Monitoring the change of seawater biological number is very important for nuclear power plant. A large sea invasion will plug plant into water source cold device, resulting in unit unplanned shutdown. At present, facing to unknown invasion, only can do some additional interceptor network and take artificial salvage, they haven't find a better solution response situation serious. Therefore, order ensure safety circulating from plants, minimize or avoid huge economic losses caused by...
Aiming at the problem that marine organisms entering water intake threaten safety of cold source system, a fixed-point acoustic monitoring method for in open channel nuclear power plant is designed. Ek80 scientific fish detector used to monitor all kinds organisms, so as obtain density change juvenile and other front net interception network, efficiency collecting by network; The designed first mesh network seawater channel, beam direction vertical downward. results show under normal sea...