近期看到的好的论文以及开源项目20240720

Scaling Up Dynamic Human-Scene Interaction Modeling

项目地址：https://jnnan.github.io/trumans/

这个环境看起来像是基于Gazebo搭建的

Adaptive Horizon Actor-Critic for Policy Learning in Contact-Rich Differentiable Simulation

https://adaptive-horizon-actor-critic.github.io/

一看就是基于mujoco的 控制策略 强化学习 等等

Haisor: Human-Aware Indoor Scene Optimization via Deep Reinforcement Learning

http://www.geometrylearning.com/HAISOR/

URDFormer: A Pipeline for Constructing Articulated Simulation Environments from Real-World Images

https://urdformer.github.io/

这个有点吊啊 很方便

AutoMate: Specialist and Generalist Assembly Policies over Diverse Geometries

https://developer.nvidia.com/blog/training-sim-to-real-transferable-robotic-assembly-skills-over-diverse-geometries/

RVT-2: Learning Precise Manipulation from Few Examples

https://robotic-view-transformer-2.github.io/

Human-In-The-Loop Task and Motion Planning for Imitation Learning

英伟达机器人组

https://research.nvidia.com/labs/srl/

https://ddshan.github.io/#work
全是牛人 导师李飞飞的博士生
https://wenlong.page/

Deep Learning Institute

https://learn.nvidia.com/courses/course?course_id=course-v1:DLI+S-OV-03+V1&unit=block-v1:DLI+S-OV-03+V1+type@vertical+block@18c0ffefca0745a98cfea926dd669b8e

感兴趣的作者

https://research.nvidia.com/labs/srl/authors/kaichun-mo/

https://research.nvidia.com/labs/srl/authors/jie-xu/