Develop Drones Without a Drone: ArduPilot Environment Setup and SITL Simulation (2026)

In our previous article, we built a QGroundControl (QGC) development environment and compiled the source code. Many readers have since asked:

What if I don’t have a flight controller?
Can I learn drone development without a drone?
How do I debug the MAVLink protocol?
How do I verify QGC interface changes?

The answer is: absolutely yes.

In fact, most drone developers spend more time on SITL (Software In The Loop) simulation than on real aircraft during the early stages of development.

With SITL, you run the complete flight controller program on your PC, enabling:

Flight controller logic validation
MAVLink communication debugging
Mission planning and testing
QGroundControl integration
Autonomous flight algorithm verification

You can complete most development work without a flight controller, a drone, or even a remote control. Today we will set up the ArduPilot development environment from scratch and launch your first SITL simulated aircraft.

At Aomway, we build FPV video transmission systems that connect flight controllers to pilots’ goggles — understanding the ArduPilot and MAVLink ecosystem is essential for our engineering team.

What is ArduPilot?

ArduPilot is one of the most popular open-source autopilot projects in the world. It supports:

Multirotor (Copter)
Fixed-wing (Plane)
Ground vehicles (Rover)
Boats
Submersibles (Sub)

Its core features:

Fully open source
Active community
MAVLink protocol support
Support for mainstream hardware such as PX4, Cube, and Pixhawk

The system architecture:

Ground Station (QGC/Mission Planner)
           ↑
        MAVLink
           ↑
    Flight Controller
           ↑
       ArduPilot

SITL allows the flight controller program to run directly on your PC.

What is SITL?

SITL stands for Software In The Loop.

In simple terms:

Real flight controller → runs on Pixhawk
becomes:
Flight controller code → runs on PC

This lets us:

Debug code
Inspect MAVLink messages
Test autonomous flight logic
Validate ground station features

All without any hardware. For developers, this is the lowest-cost way to learn.

Development Environment

This tutorial uses:

Windows 11
WSL2 Ubuntu 22.04
ArduPilot Copter 4.6.3

Recommended specs:

CPU: 4+ cores
RAM: 16GB+
Disk: 50GB+

Install Base Dependencies

Update the system:

sudo apt update
sudo apt upgrade

Install development tools:

sudo apt install git
sudo apt install python3
sudo apt install python3-pip
sudo apt install python3-dev
sudo apt install wget

Install ArduPilot dependencies:

sudo apt install software-properties-common

Download ArduPilot Source Code

Create a working directory:

mkdir ~/drone
cd ~/drone

Clone the source:

git clone https://github.com/ArduPilot/ardupilot.git

Enter the project:

cd ardupilot

Update submodules:

git submodule update --init --recursive

This step takes a while — please be patient.

Install the ArduPilot Environment Script

The official one-click installer script is provided:

Tools/environment_install/install-prereqs-ubuntu.sh -y

The script automatically installs:

gcc
g++
python
waf
sim_vehicle
MAVProxy
Build toolchain

After installation, run:

source ~/.profile

to activate the environment variables.

Compile ArduCopter

Enter the project root:

cd ~/drone/ardupilot

Configure the build environment:

./waf configure

Compile the multirotor flight controller:

./waf copter

The first build may take a few minutes. On success you will see:

Build successful

Launch Your First SITL

Run:

sim_vehicle.py -v ArduCopter -I 0 --sysid=1 --out 192.168.191.1:14550

After a moment you will see:

Starting ArduCopter

Then:

Ready to FLY

Your flight controller has successfully started — a complete virtual autopilot is now running on your PC.

Connect QGroundControl

For QGC users, connection is simpler than with Mission Planner. SITL sends MAVLink data to:

UDP 14550

QGC usually auto-discovers the aircraft on startup. If not:

Application Settings → Comm Links

Add a new link:

UDP, 14550

Test Automatic Takeoff

In the MAVProxy console, enter:

mode guided

Then arm the aircraft:

arm throttle

Execute takeoff:

takeoff 10

This commands a takeoff to 10 meters. In QGroundControl you will see:

Aircraft armed
Altitude rising
Attitude changes
MAVLink data refreshing

The entire process requires no real aircraft.

Testing Multi-Vehicle Simulation

The official multi-vehicle command:

sim_vehicle.py -v Copter --map --console --count 5 --auto-sysid --location CMAC --auto-offset-line 90,10 --mcast --out 192.168.111.1:14550

In practice this is hard to control. We recommend launching multiple single-vehicle terminals as described above — it runs much more smoothly.

What Can SITL Do?

Many beginners think SITL is just a demo tool. In reality, a huge amount of development work happens inside SITL. For example:

MAVLink Protocol Analysis

Inspect messages such as:

HEARTBEAT
ATTITUDE
GLOBAL_POSITION_INT
SYS_STATUS

QGroundControl Development

Debug interfaces like:

Flight View
Mission
Map
Parameter System

Autonomous Flight Algorithms

Validate features such as:

Waypoint flight
Auto return-to-launch
Auto landing
Obstacle avoidance logic

Common Issues

QGC Cannot Find the Aircraft

Check whether UDP port 14550 is blocked by your firewall.

SITL Fails to Start

Re-run:

git submodule update --init --recursive

and confirm all dependencies are complete.

Compilation Errors

Run:

./waf clean
./waf configure
./waf copter

and rebuild.

Summary

For drone developers, the first step to learning flight controller development is not buying an aircraft — it is mastering SITL.

With ArduPilot + SITL, you can complete on an ordinary PC:

Flight controller development
MAVLink debugging
Ground station integration
Autonomous flight validation

All with zero hardware cost. At Aomway, our engineers use the same workflow to validate FPV video links and telemetry integration before hardware is ready.

In the next article we will connect QGroundControl to ArduPilot SITL — from MAVLink communication to automatic takeoff.

If you have any questions about this topic, feel free to contact us at [email protected]

Have questions about this article? Feel free to contact us at [email protected] — we’re happy to help!