| 1 |
Real-Time Adaptive Motion Planning via Point Cloud-Guided, Energy-Based Diffusion and Potential Fields |
提出一种基于点云引导和能量扩散的实时自适应运动规划方法,用于追逃场景。 |
motion planning classifier-free guidance |
|
|
| 2 |
Informed Hybrid Zonotope-based Motion Planning Algorithm |
提出HZ-MP算法,通过混合zonotope分解和启发式采样解决非凸空间最优路径规划问题 |
motion planning |
|
|
| 3 |
Towards Human-level Dexterity via Robot Learning |
提出基于结构化探索的强化学习框架,提升机器人灵巧操作能力 |
manipulation reinforcement learning imitation learning |
|
|
| 4 |
Learning to Move in Rhythm: Task-Conditioned Motion Policies with Orbital Stability Guarantees |
提出OSMP框架,结合学习的微分同胚编码器与Hopf分岔,实现轨道稳定且任务可控的周期运动控制。 |
locomotion diffusion policy |
|
|
| 5 |
Constrained Style Learning from Imperfect Demonstrations under Task Optimality |
提出基于约束马尔可夫决策过程的模仿学习方法,提升机器人任务性能与风格 |
ANYmal |
|
|
| 6 |
PRAG: Procedural Action Generator |
PRAG:用于机器人操作任务的程序化动作生成器 |
manipulation |
|
|