- Reinforcement Learning in Robotics
- Adaptive Control of Nonlinear Systems
- Power Systems and Renewable Energy
- Optimization and Search Problems
- Advanced Battery Technologies Research
- Robotic Path Planning Algorithms
- Distributed Control Multi-Agent Systems
- Power Systems and Technologies
- Reliability and Maintenance Optimization
- Software Reliability and Analysis Research
- Advanced Control Systems Optimization
- Distributed and Parallel Computing Systems
- Gene Regulatory Network Analysis
- Computational Geometry and Mesh Generation
- Aerospace Engineering and Control Systems
- Evacuation and Crowd Dynamics
- Markov Chains and Monte Carlo Methods
- Genetics, Aging, and Longevity in Model Organisms
- Microgrid Control and Optimization
- Auction Theory and Applications
- Advanced Bandit Algorithms Research
- Advanced Malware Detection Techniques
- Energy Efficient Wireless Sensor Networks
- Aerospace and Aviation Technology
- Video Coding and Compression Technologies
Harbin Institute of Technology
2013-2025
Fudan University
2025
Westlake University
2025
Institute of Basic Medical Sciences of the Chinese Academy of Medical Sciences
2025
Shenzhen Institute of Information Technology
2021
Xinjiang University
2015
Suzhou University of Science and Technology
2012
Hong Kong University of Science and Technology
2008
Walking on different terrains leads to biomechanics, which motivates the development of exoskeletons for assisting walking according type a terrain. The design lightweight soft exoskeleton that simultaneously assists multiple joints in lower limb is presented this paper. It used assist both hip and knee single system, assistance force directly applied joint flexion extension, while indirectly extension also. Based biological torque human at three slopes, novel strategy developed improve...
With the rapid development of mobile robot technology, robots are playing an increasingly important role in people's daily lives. As one key technologies basic functions robots, navigation also needs to deal with new challenges. How navigate efficiently and collision-free complex changeable human environments is problems that need be solved urgently. Currently, can achieve efficient static environments. However, unstructured fast-changing society, make more flexible strategies dynamic...
Fast and accurate estimation of battery state charge (SOC) is a key technology in the management system. Based on non-linear response characteristics lithium batteries, an adaptive Kalman filter algorithm put forward this paper. It known that model parameters vary with SOC, temperature aging. Moreover, relationship between open circuit voltage (OCV) SOC nonlinear. To solve these issues, piecewise linear approximation proposed based then nonlinear turned into one. On bases, can be implemented...
A heterogeneous wireless network can be modelled as a directed disk graph, in which sensor u receive signals from v if and only is within the transmission range of v. virtual backbone such strongly connected dominating absorbing set (SCDAS). This paper presents (2 + ε)-approximation algorithm for minimum problem (DAS), which, far we know, first one achieving constant approximation ratio DAS without requiring bounded maximum over range. Based on DAS, (4 3 ln(2 ε)opt SCDAS presented, where opt...
For the task assignment problem in a smart logistic center with many autonomous mobile robots, method is proposed based on market auction idea. Due to existence of asymmetrical shelf-shift tasks within and diverse self-finished cost each besides correlation between tasks, multi-robot different from routing problem. By introducing function function, model presented. Then, bidding strategy given by analyzing performance difference. On these bases, an approach developed for static dynamic...
The technologies of multimedia compression, copyright protection and content authentication are essential to the booming industry. Watermarking schemes for compressed video effective solutions In this paper, we first present a hybrid compression scheme that adopts both scalar quantisation transform domain vector quantisation, then multipurpose watermarking algorithm is proposed. digital fingerprint, robust fragile watermarks embedded in different parts stream accomplish multiple purposes...
As the battery energy storage station has function of improving fluctuations power output and regulating system's economic benefit caused by time-of-use pricing, optimization model is proposed to stabilize maximize profit micro-grid with known each distributed generator. In order avoid disadvantages dynamic programming algorithm genetic algorithm, we transform a convex problem obtain optimal scheduling policy. Take specified data in north area as example, comparison diagram on without use...
The quasi-static operations of MEMS mirror are very sensitive to undesired oscillations due its low damping. It has been shown that closed-loop control can be superior reduce those than open-loop in the literature. For control, conventional way implementing sliding mode (SMC) algorithm is forward Euler method, which results numerical chattering input and output. This paper proposes an implicit implementation scheme super twisting observer for a commercial actuated by electrostatic staggered...
Beijing–Tianjin–Hebei coordinated development is a major national strategy of China. This study first analyzes the evolution process energy production, consumption, and structure proportion in Beijing, Tianjin, Hebei, indicating that overall supply this region less than consumption; although rich renewable resources, its consumption mainly fossil energies efficiency requires further improvement. Since was proposed six years ago, ecological environment has continuously improved, but situation...
In this paper, we investigate a planar parallel mechanism called five-bar which is used in wire bonders of semiconductor packaging system where the required accuracy about 2 (um), acceleration reaches 10–15 (g). However, conventional modeling and analyzing methods that treat all components manipulator as rigid ignore possible deformations, definitely are insufficient to satisfy those rigorous demands. Thus, dynamic model with consideration both motion flexible displacement developed for...
On the basis of sensitivity-based optimization, we develop a unified optimization approach for semi-Markov decision processes (SMDPs) with infinite horizon discounted and average reward criteria. We show that sensitivity formula under criteria is limitation case performance formulas, provide formulation policy iteration algorithms
In this paper, we discuss the problem of sample-path-based (on-line) performance gradient estimation for Markov systems. The existing on-line algorithms generally require a standard importance sampling assumption. When assumption does not hold, these may lead to poor estimates gradients. We show that can be relaxed. propose few provide systems do satisfy Simulation examples are given illustrate accuracy estimates.