policy

RTWholeBodyMPPI

jrapudg · PyTorch

or hover any field below to flag it

Overview

Name
RTWholeBodyMPPI
Author
jrapudg
Framework
PyTorch
License
MIT
Skill type
locomotion
Evidence level
community
Task description
We present the first successful deployment of a whole-body sampling-based MPC system on a real-world quadruped robot. Our approach enables real-time, contact-rich locomotion and manipulation using a single control policy derived from Model-Predictive Path Integral (MPPI) control.

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 RTWholeBodyMPPI yet.

Datasets that reference this policy

0

No datasets reference RTWholeBodyMPPI yet.