RoboHive

RoboHive is a collection of environments/tasks simulated with the MuJoCo physics engine exposed using the OpenAI-Gym API.

Getting Started

Getting started with RoboHive is as simple as -

# Install RoboHive and demo an environemnt
pip install robohive
python -m robohive.utils.examine_env -e FrankaReachRandom-v0

or, alternatively,

git clone --recursive https://github.com/vikashplus/robohive.git; cd robohive
pip install -e .
python -m robohive.utils.examine_env -e FrankaReachRandom-v0

See detailed installation instructions for options on mujoco-python-bindings and visual-encoders (R3M, RRL, VC), and frequently asked questions for more details.

Suites

RoboHive contains a variety of environement, which are organized as suites. Each suites is a collection of loosely related environements. Following suites are provided at the moment with plans to improve the diversity of the collection.

- Hand Manipulation Suite

This suite contains a collection of environement centered around dexterous manipulation. Standard ADROIT benchmarks introduced in Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrations, RSS2018.) are a part of this suite

- Arm Manipulation Suite

This suite contains a collection of environement centered around Arm+Gripper manipulation.

- Myo Suite

This suite contains a collection of environements related to biomechanics. Standard MyoSuite benchmarks are a part of this suite

- MultiTask Suite

This suite contains a collection of environement centered around multi-tassk. Standard RelayKitchen benchmarks are a part of this suite

- TCDM Suite (WIP)

This suite contains a collection of environement centered around dexterous manipulation. Standard TCDM benchmarks are a part of this suite

- ROBEL Suite (Coming soon)

This suite contains a collection of environement centered around real world locomotion and manipulation. Standard ROBEL benchmarks are a part of this suite

liuzc188/robohive