1. What is RflySim?

The RflySim Automated Design and Development Toolchain for Intelligent Unmanned Swarm Systems (RflySim Toolchain) is a professional, free, and open development toolchain created by Beijing Droneyee FeiSi Labs in collaboration with Xunhua Dai's team at Central South University and Quan Quan's team at Beihang University. It provides a full-process automation solution for unmanned systems spanning algorithm development, hardware-in-the-loop (HIL) simulation, and real-aircraft deployment. Built on a Model-Based Design (MBD) philosophy and leveraging PX4, MATLAB/Simulink, ROS, and related intelligent hardware, the toolchain supports (but is not limited to) simulation and hands-on practice in unmanned agent control, swarm coordination, and computer vision. Researchers can perform unmanned system modeling, controller design, Software-In-the-Loop (SIL) simulation, and HIL simulation. Through the automatic code generation capabilities of MATLAB/Simulink, controllers can be conveniently deployed to hardware for HIL simulation and real flight testing, achieving a seamless Simulation-to-Reality (Sim2Real) transition. 🔗See link for details🔗

RflySim originates from: R, Reliable; fly, Flight; Sim, Simulation. Pronounced R-fly-Sim, the full name stands for Reliable Flight Simulation.

2. Supporting Textbooks

3. Video Replay of RflySim Labs 2024 Summer Public Training

4. Version Tiers

The current free and publicly available version of the RflySim Toolchain already covers the vast majority of unmanned system development needs. For more complex or customized features (primarily distributed multi-vehicle simulation, UE5 engine support, latest PX4 firmware, etc.), RflySim Labs also offers paid feature packages and project customization services. For details, please contact: service@rflysim.com.

For more details on version differences, as well as customization and training services, see: 🔗Version Comparison🔗

5. Teaching Aids & Equipment

The official RflySim Labs Taobao store is now live! We regularly list software and hardware products and teaching aids for unmanned system development. Currently available products include the HIL Simulation Kit, the RflySim MiniQuad150 UAV Development Platform, unmanned system vehicle dynamic model packages, visual display asset packages, and more. Store link: https://rflysim.taobao.com/

6. Key Features

The Advanced and Customized editions of the RflySim Toolchain offer the following key features:

• Unified framework. The entire research framework extends to all unmanned control systems, forming a standardized development, testing, and evaluation ecosystem;

  

• High-fidelity UAV physics simulation. The development team consists entirely of UAV researchers with extensive domain experience;

• Ease of use. On the Windows platform, users enjoy one-click installation, one-click code generation, one-click firmware deployment, one-click SIL/HIL simulation, and rapid real flight testing. There is no need to understand flight controller source code, Linux programming, C/C++ programming, network communication, or aircraft assembly. With just basic Simulink (or Python) knowledge, users can quickly validate their algorithms through multiple verification stages and deploy them to real aircraft, allowing them to focus on algorithm development and testing.

• Fully distributed architecture. All application components can run multiple instances on one or more computers, communicating with each other via UDP. This distributed architecture is particularly well-suited for large-scale vision-enabled UAV swarm simulation and testing;

  

• Multi-vehicle type simulation. Supports ground vehicles, fixed-wing aircraft, vertical take-off and landing (VTOL) aircraft, and more. Users can build vehicle models in Simulink using standardized interfaces, then automatically generate DLL files for HIL simulation. The platform can be further extended to any unmanned system;

  

• Large-scale UAV swarm SIL/HIL simulation. Within the same LAN, developers can use CopterSim to connect multiple Pixhawk autopilots for hardware or software in-the-loop simulation. Simulink or C++ programs can be used to control the vehicles, with commands sent to Pixhawk via MAVLink protocol over serial (telemetry radio) or network (Wi-Fi);

  

• Highly realistic 3D visualization. Source code and tutorials are provided to help developers build photorealistic 3D scenes in Unreal Engine (UE) for indoor/outdoor environment simulation and vision-based algorithm development. Scenes support physics-based collision, global terrain and maps, OSGB + Cesium oblique photogrammetry map import, custom GPS coordinates, arbitrary multi-window view switching, and sensor data output including RGB, depth, grayscale, IMU, and LiDAR. Shared memory and UDP-based image streaming to specified IP addresses are supported, enabling onboard-computer hardware-in-the-loop SLAM simulation.

  

• Vision-based control. The UE-based 3D visualization platform also supports viewpoint switching, enabling convenient access to image data from multiple perspectives. It supports real-time image acquisition and processing in Simulink, Python, C/C++, and other platforms via shared memory. Processed visual data can be sent back to CopterSim or Simulink via UDP, forming a closed-loop vision-enabled hardware-in-the-loop simulation.

  

• Multiple fault injection. Supported fault types include model faults, communication faults, and environmental faults.