eess.SY(2026-10-02)
📊 共 6 篇论文
🎯 兴趣领域导航
支柱一:机器人控制 (Robot Control) (2)
支柱二:RL算法与架构 (RL & Architecture) (2)
支柱四:生成式动作 (Generative Motion) (1)
支柱八:物理动画 (Physics-based Animation) (1)
🔬 支柱一:机器人控制 (Robot Control) (2 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 1 | Hierarchical Control via MPC-RL for Multi-Timescale Battery Systems | 提出层次控制框架解决多时间尺度电池系统问题 | MPC model predictive control reinforcement learning | ||
| 2 | Mission-Centric Requirements Analysis of Model Predictive Control in Spacecraft Rendezvous and Proximity Operations | 提出经济模型预测控制以优化航天器接近操作 | MPC model predictive control |
🔬 支柱二:RL算法与架构 (RL & Architecture) (2 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 3 | TwinJEPA: Action-Preferred Predictive Representations for Goal-Conditioned Control | 提出TwinJEPA以解决动作偏好预测不足问题 | JEPA Joint-Embedding Predictive Architecture joint-embedding predictive architecture | ||
| 4 | FARM: Fundamental Agentic Reward Model For Multi-task Wireless Network Optimization | 提出FARM以解决多任务无线网络优化中的知识重用问题 | reinforcement learning SAC Mamba |
🔬 支柱四:生成式动作 (Generative Motion) (1 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 5 | Risk-Aware Input-Constrained Safe Intercept Guidance Against Multiple Moving Defenders | 提出风险感知的输入约束安全拦截指导法应对多移动防御者问题 | penetration |
🔬 支柱八:物理动画 (Physics-based Animation) (1 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 6 | MagLearn 2: High-Fidelity, Saturation-Aware, and History-Efficient Sequence-to-sequence Modeling of Transient B-H Behaviour Under PWM Excitation | 提出MagLearn 2以解决PWM激励下瞬态B-H建模问题 | PULSE |