Going into Orbit: Massively Parallelizing Episodic Reinforcement Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Oberst, Jan, Bonneau, Johann
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910452856913920
author Oberst, Jan
Bonneau, Johann
author_facet Oberst, Jan
Bonneau, Johann
contents The possibilities of robot control have multiplied across various domains through the application of deep reinforcement learning. To overcome safety and sampling efficiency issues, deep reinforcement learning models can be trained in a simulation environment, allowing for faster iteration cycles. This can be enhanced further by parallelizing the training process using GPUs. NVIDIA's open-source robot learning framework Orbit leverages this potential by wrapping tensor-based reinforcement learning libraries for high parallelism and building upon Isaac Sim for its simulations. We contribute a detailed description of the implementation of a benchmark reinforcement learning task, namely box pushing, using Orbit. Additionally, we benchmark the performance of our implementation in comparison to a CPU-based implementation and report the performance metrics. Finally, we tune the hyper parameters of our implementation and show that we can generate significantly more samples in the same amount of time by using Orbit.
format Preprint
id arxiv_https___arxiv_org_abs_2405_11512
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Going into Orbit: Massively Parallelizing Episodic Reinforcement Learning
Oberst, Jan
Bonneau, Johann
Robotics
The possibilities of robot control have multiplied across various domains through the application of deep reinforcement learning. To overcome safety and sampling efficiency issues, deep reinforcement learning models can be trained in a simulation environment, allowing for faster iteration cycles. This can be enhanced further by parallelizing the training process using GPUs. NVIDIA's open-source robot learning framework Orbit leverages this potential by wrapping tensor-based reinforcement learning libraries for high parallelism and building upon Isaac Sim for its simulations. We contribute a detailed description of the implementation of a benchmark reinforcement learning task, namely box pushing, using Orbit. Additionally, we benchmark the performance of our implementation in comparison to a CPU-based implementation and report the performance metrics. Finally, we tune the hyper parameters of our implementation and show that we can generate significantly more samples in the same amount of time by using Orbit.
title Going into Orbit: Massively Parallelizing Episodic Reinforcement Learning
topic Robotics
url https://arxiv.org/abs/2405.11512