ardupilot_gazebo — installing the plugin and connecting SITL to Gazebo

3. Install Gazebo plugins

cd ~
git clone https://github.com/ArduPilot/ardupilot_gazebo.git

cd ~/ardupilot_gazebo
mkdir build && cd build
cmake .. -DCMAKE_BUILD_TYPE=RelWithDebInfo
make -j4

Install gstreamer:

sudo apt install libgstreamer1.0-dev libgstreamer-plugins-base1.0-dev

Plugin built successfully! Add the paths:

echo 'export GZ_SIM_SYSTEM_PLUGIN_PATH=$HOME/ardupilot_gazebo/build:${GZ_SIM_SYSTEM_PLUGIN_PATH}' >> ~/.bashrc
echo 'export GZ_SIM_RESOURCE_PATH=$HOME/ardupilot_gazebo/models:$HOME/ardupilot_gazebo/worlds:${GZ_SIM_RESOURCE_PATH}' >> ~/.bashrc
source ~/.bashrc

ls ~/ardupilot_gazebo/models

We see ready-made models — iris_with_ardupilot.

gz sim ~/ardupilot_gazebo/worlds/iris_runway.sdf -r

The -r flag runs the simulation — same as pushing play in the Gazebo layout.

Connecting SITL to Gazebo

Open a new terminal:

cd ~/ardupilot/ArduCopter
sim_vehicle.py -v ArduCopter -f gazebo-iris --model JSON --console

Breakdown:

-f gazebo-iris — use the iris model for Gazebo
--model JSON — SITL talks to Gazebo via the JSON protocol

If Gazebo doesn’t render (GeForce GPU)

sudo nvidia-settings --assign CurrentMetaMode="nvidia-auto-select +0+0 { ForceFullCompositionPipeline = On }"

Data flow

gz sim (body)
    ↕ UDP 9002/9003
sim_vehicle.py (brain)
    ↕ ros_gz_bridge (translator)
sensor_monitor.py (our code)