A5108616249- Reinforcement Learning in Robotics
- Distributed Control Multi-Agent Systems
- Supercapacitor Materials and Fabrication
- Advancements in Battery Materials
- Advanced Battery Materials and Technologies
- Opinion Dynamics and Social Influence
- Advanced Control Systems Optimization
- Mathematical and Theoretical Epidemiology and Ecology Models
- Fault Detection and Control Systems
- Modular Robots and Swarm Intelligence
- Robotic Path Planning Algorithms
- Evacuation and Crowd Dynamics
- Advanced battery technologies research
- 3D Shape Modeling and Analysis
- Autonomous Vehicle Technology and Safety
- Computational Geometry and Mesh Generation
- 3D Modeling in Geospatial Applications
Beihang University
2023-2024
Beijing Institute of Technology
2023
Xi'an Jiaotong University
2016-2018
To address the volume-change-induced pulverization problems of electrode materials, we propose a "silica reinforcement" concept, following which silica-reinforced carbon nanofibers with encapsulated Sb nanoparticles (denoted as SiO2/Sb@CNFs) are fabricated via an electrospinning method. In this composite structure, insulating silica fillers not only reinforce overall structure but also contribute to additional lithium storage capacity; encapsulation into carbon-silica matrices efficiently...
The design of tin-based anode materials (SnO2 or Sn) has become a major concern for lithium ion batteries (LIBs) owing to their different inherent characteristics. Herein, particulate SnO2 Sn crystals coupled with porous N-doped carbon nanofibers (denoted as SnO2/PCNFs and Sn/PCNFs, respectively) are fabricated via the electrospinning method. electrochemical behaviors both Sn/PCNFs systematically investigated anodes LIBs. When nanofibers, nanoparticles micro/nanoparticles display superior...
Swarm models hold significant importance as they provide the collective behavior of self-organized systems. Boids model is a fundamental framework for studying emergent in swarms It addresses problems related to simulating autonomous agents, such alignment, cohesion, and repulsion, imitate natural flocking movements. However, traditional often lack pinning adaptability quickly adapt dynamic environment. To address this limitation, we introduce reinforcement learning into solve problem...
In swarm robotics, confrontation including the pursuit-evasion game is a key scenario. High uncertainty caused by unknown opponents' strategies and dynamic obstacles complicates action space into hybrid decision process. Although deep reinforcement learning method significant for since it can handle various sizes, as an end-to-end implementation, cannot deal with Here, we propose novel hierarchical approach consisting of target allocation layer, path planning underlying interaction mechanism...
Reinforcement learning suffers from limitations in real practices primarily due to the number of required interactions with virtual environments. It results a challenging problem because we are implausible obtain local optimal strategy only few attempts for many methods. Hereby, design an improved reinforcement method based on model predictive control that models environment through data–driven approach. Based learned model, it performs multi–step prediction estimate value function and...
The bird-oid object (Boids) model proposes a control algorithm to make the positions between agents achieve cooperative stability. By changing parameters of cohesion and repulsion in algorithm, swarm can be made converge different positions, causing expansion contraction formation. But it is often more difficult select appropriate form ideal Therefore, this paper method improve cohesive repulsive Boids based on Q-learning network simulation scenario with continuous obstacle avoidance maximum...
Multi-agent reinforcement learning based methods are significant for online planning of feasible and safe paths agents in dynamic uncertain scenarios. Although some like fully centralized decentralized achieve a certain measure success, they also encounter problems such as dimension explosion poor convergence, respectively. In this paper, we propose novel training with execution method on multi-agent to solve the obstacle avoidance problem online. approach, each agent communicates only...
Reinforcement learning suffers from limitations in real practices primarily due to the number of required interactions with virtual environments. It results a challenging problem because we are implausible obtain local optimal strategy only few attempts for many methods. Hereby, design an improved reinforcement method based on model predictive control that models environment through data-driven approach. Based learned model, it performs multi-step prediction estimate value function and...