OmniDrones: An Efficient and Flexible Platform for Reinforcement Learning in Drone Control

Botian Xu, Feng Gao, Chao Yu, Ruize Zhang, Yi Wu +1 more
9/22/2023
cs.ROcs.AIcs.LG

Official code: btx0424/OmniDrones (583 stars, MIT, last commit 2026-01-20).

Abstract

In this work, we introduce OmniDrones, an efficient and flexible platform tailored for reinforcement learning in drone control, built on Nvidia's Omniverse Isaac Sim. It employs a bottom-up design approach that allows users to easily design and experiment with various application scenarios on top of GPU-parallelized simulations. It also offers a range of benchmark tasks, presenting challenges ranging from single-drone hovering to over-actuated system tracking. In summary, we propose an open-sourced drone simulation platform, equipped with an extensive suite of tools for drone learning. It includes 4 drone models, 5 sensor modalities, 4 control modes, over 10 benchmark tasks, and a selection of widely used RL baselines. To showcase the capabilities of OmniDrones and to support future research, we also provide preliminary results on these benchmark tasks. We hope this platform will encourage further studies on applying RL to practical drone systems.

View on arXivView PDF

Code Implementations(1)

btx0424/OmniDronesOfficial100%
MIT
58390Shell, PythonFeb 9, 20239 months agoMIT
reinforcement-learningroboticssimulation

Similar Papers

Open-World Drone Active Tracking with Goal-Centered Rewards

85% match1 repo282024

Towards Task-Oriented Flying: Framework, Infrastructure, and Principles

83% match2025

aerial-autonomy-stack -- a Faster-than-real-time, Autopilot-agnostic, ROS2 Framework to Simulate and Deploy Perception-based Drones

82% match2026

Reinforcement Learning for Decision-Level Interception Prioritization in Drone Swarm Defense

82% match1 repo222025

AirPilot: Interpretable PPO-based DRL Auto-Tuned Nonlinear PID Drone Controller for Robust Autonomous Flights

82% match1 repo42024

Motion Control in Multi-Rotor Aerial Robots Using Deep Reinforcement Learning

81% match2025

Cite this paper

@article{xu2023omnidrones,
  title  = {OmniDrones: An Efficient and Flexible Platform for Reinforcement Learning in Drone Control},
  author = {Botian Xu and Feng Gao and Chao Yu and Ruize Zhang and Yi Wu and Yu Wang},
  year   = {2023},
  eprint = {2309.12825},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2309.12825v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2309.12825.svg)](https://codeofpaper.com/paper/2309.12825)

Discussion