mihdalal/raps
[NeurIPS 2021] PyTorch Code for Accelerating Robotic Reinforcement Learning with Parameterized Action Primitives
This project helps roboticists and AI researchers accelerate the training of robot agents for complex tasks. By providing a framework that leverages pre-defined, high-level actions (primitives) instead of raw motor commands, it allows for quicker learning. Researchers input robot environment data and task definitions, and the output is a more efficiently trained robotic agent capable of performing intricate actions in environments like a kitchen or a manufacturing setup.
No commits in the last 6 months.
Use this if you are a roboticist or AI researcher developing reinforcement learning agents for robots and want to significantly speed up their learning process for complex, multi-step tasks.
Not ideal if you are working with very simple robot movements or if your robot's environment and tasks are highly unstructured and don't benefit from pre-defined action sequences.
Stars
79
Forks
11
Language
Python
License
MIT
Category
Last pushed
Jul 11, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/mihdalal/raps"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
LucasAlegre/sumo-rl
Reinforcement Learning environments for Traffic Signal Control with SUMO. Compatible with...
hilo-mpc/hilo-mpc
HILO-MPC is a Python toolbox for easy, flexible and fast development of...
reiniscimurs/DRL-robot-navigation
Deep Reinforcement Learning for mobile robot navigation in ROS Gazebo simulator. Using Twin...
kyegomez/RoboCAT
Implementation of Deepmind's RoboCat: "Self-Improving Foundation Agent for Robotic Manipulation"...
cbfinn/gps
Guided Policy Search