Reproduce “Pegasus Simulator: An Isaac Sim Framework for Multiple Aerial Vehicles Simulation” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Marcelo Jacinto, João Pinto, Jay Patrikar, John Keller, Rita Cunha, Sebastian Scherer, António Pascoal

Pegasus Simulator extends NVIDIA Isaac Sim with an open-source aerial-robotics layer for realistic, multi-vehicle multirotor simulation. It combines vehicle dynamics, aviation-specific sensors, PX4/MAVLink and ROS2 interfaces, a graphical configuration workflow, and a Python API; a two-quadrotor aggressive relay experiment demonstrated real-time operation and bounded trajectory-tracking errors, though broader fidelity and scalability were not quantified.

Key insights

  • Combining Isaac Sim’s visual and physics capabilities with PX4-compatible interfaces connects high-fidelity perception simulation to the flight-control software used in real UAV systems.
  • The modular backend design supports different controller pathways and multi-aircraft experiments without requiring researchers to modify the simulator core.
  • The initial dynamics and sensor models prioritize usability and integration, but simplified rotor behavior, absent wind and turbulence, and limited hardware validation constrain claims about physical fidelity.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.