HyperDrone: Simulation-to-Reality Transfer at Millions of Frames per Second

Jonas Eschmann; Giuseppe Loianno

cs.ROOct 20, 2026

HyperDrone: Simulation-to-Reality Transfer at Millions of Frames per Second

Authors: Jonas Eschmann, Giuseppe Loianno

Organizations: University of California, Berkeley

Abstract

We present HyperDrone, an extremely fast quadrotor simulator focused on high-performance batch rendering that is more than 20x faster than existing state-of-the-art batch quadrotor simulators. HyperDrone combines highly optimized multirotor dynamics with HyperLux, a new ray-tracing engine that supports Nvidia OptiX, Vulkan, Apple Metal and WebGPU backends that takes full advantage of modern ray-tracing hardware, capable of running both components entirely on the GPU. HyperLux renders RGB, depth, semantic/instance segmentation, surface normals and optical flow observations at configurable levels of fidelity, including reflections, realistic lighting, motion blur, anti-aliasing, and Physically Based Rendering (PBR). Even in the highest-fidelity settings, HyperLux achieves millions of Frames Per Second (FPS) on consumer-grade GPUs while rendering realistic 3D scenes. We demonstrate the merits of HyperDrone by training a vision-based yaw-control policy purely in simulation and transferring it to a real small-scale quadrotor (Crazyflie) with a camera, running completely onboard.

Figures & tables

Figure 3: Control Architecture Overview.

Figure 6: Rendering Fidelity Comparison with Four Secondary Drones.

		20 Objects				ProcTHOR
		RGB		Depth		RGB		Depth
Task	#envs	Render	Physics & Render	Render	Physics & Render	Render	Physics & Render	Render	Physics & Render
Nvidia RTX 4090 16GB (Mobile, 80W, OptiX)
Aerial Gym	$256$	$4771$	$4347$	$126\,024$	$68\,651$	–	–	$133.32$	$128.32$
Aerial Gym	$4096$	$1375$	$1328$	$127\,397$	$107\,428$	–	–	$130.97$	OOM
OmniDrones	$8$	$359.33$	$295.97$	$371.90$	$299.13$	$276.53$	$233.78$	$297.27$	$246.07$

Table 2: Simulator Throughput Comparison. Resolution: 64

\times

64, Unit: Frames Per Second (FPS). OOM = out of memory, T/O = timeout

>15

min.

Figure 8: Real-World Target Frame Tracking: There is a chair and a ladder in the field-of-view.

Robotics Simulation

Robotics simulation builds virtual robots and environments to model, train, and evaluate autonomous systems before real-world deployment.

See all 322 papers

Droneulator: A Portable UAV Simulator for Agricultural Workflows with RotorPy and Godot 4
May 22, 2026Jacob Swindell, Michael Lowen, Marija Popovic +1Robotics Simulation Robot Navigation
Lincoln Centre for Autonomous Systems (L-CAS), University of Lincoln, Lincoln, UK. · Faculty of Aerospace Engineering, MAVLab, Delft University of Technology (TU Delft), Delft, Netherlands.
Simple but Stable, Fast and Safe: Achieve End-to-end Control by High-Fidelity Differentiable Simulation
Apr 12, 2026Fanxing Li, Shengyang Wang, Yuxiang Huang +5Quadrotor Control Vision-Based Robot Control
Shanghai Jiao Tong University · Kunshan Duke University
Crazyflow: An Accurate, GPU-Accelerated, Differentiable Drone Simulator in JAX
May 31, 2026Martin Schuck, Marcel P. Rath, Yufei Hua +3Aerial Robotics Robotics Simulation
Technical University of Munich · University of Toronto · Simon Fraser University

May 22, 2026J/K move · Enter open · S save

Droneulator: A Portable UAV Simulator for Agricultural Workflows with RotorPy and Godot 4

Jacob Swindell, Michael Lowen, Marija Popovic, Riccardo Polvara

Lincoln Centre for Autonomous Systems (L-CAS), University of Lincoln, Lincoln, UK. · Faculty of Aerospace Engineering, MAVLab, Delft University of Technology (TU Delft), Delft, Netherlands.

Robotics Simulation Robot Navigation Agricultural Robotics

Open paper AskPDF arXiv

Agricultural UAV research requires simulators that integrate realistic 3D scenes, high-fidelity vehicle dynamics, and robotics middleware, while remaining practical to deploy across heterogeneous development machines. We present Droneulator, a portable UAV simulator architecture that combines RotorPy for multirotor dynamics with Godot 4 for rendering and sensor generation. Droneulator exposes both PX4-based control and a lightweight WebSocket command path, and publishes synchronised visual and state streams through a Zenoh-based ROS~~2-compatible pipeline. This integration enables a single stack to support inspection-oriented data capture, ROS~~2/PX4 local planning, and reinforcement learning experiments without modifying the simulator infrastructure. We present quantified validation of the current system across three agricultural UAV workflows: tree-scale image collection for 3D reconstruction with COLMAP, local planning around canopy obstacles using EGO-Planner, and closed-loop reinforcement learning through a custom Gymnasium environment. In the reported setup, the results show that the simulator can sustain low-latency sensing, support reconstruction-oriented data collection under varying capture density, execute collision-free local planning around canopy obstacles, and support stable depth-sensing-based policy training for obstacle-aware navigation. Together, these results show the potential of Droneulator for agricultural UAV inspection, planning, and learning within one deployable stack.

Abstract

Robotics Simulation

Explore similar work

Droneulator: A Portable UAV Simulator for Agricultural Workflows with RotorPy and Godot 4

Simple but Stable, Fast and Safe: Achieve End-to-end Control by High-Fidelity Differentiable Simulation

Crazyflow: An Accurate, GPU-Accelerated, Differentiable Drone Simulator in JAX