policy

MDP-DRLF-Multimodal-Diffusion-Policy-with-Reinforcement-Fine-tuning

cxcAxxy · PyTorch

or hover any field below to flag it

Overview

Name
MDP-DRLF-Multimodal-Diffusion-Policy-with-Reinforcement-Fine-tuning
Author
cxcAxxy
Framework
PyTorch
License
MIT
Skill type
manipulation
Evidence level
untested
Task description
This project develops a unified framework for complex robotic manipulation by integrating multimodal perception with diffusion-based policy learning. Using RGB-D, point clouds, and tactile inputs, it models high-dimensional actions and leverages reinforcement learning for robust and generalizable pe

Spaces

Action space
other · 0-dim · 0Hz
Observation space
  • type: other

Links

HuggingFace repo
null
Paper (arXiv)
null

Compatible robots

20

Compatible environments

0

No environments list MDP-DRLF-Multimodal-Diffusion-Policy-with-Reinforcement-Fine-tuning yet.

Datasets that reference this policy

0

No datasets reference MDP-DRLF-Multimodal-Diffusion-Policy-with-Reinforcement-Fine-tuning yet.