Linux build

Table of Contents

• Documentation
• Prerequisites
  • System Recommendation
• Quick Guide
• Unreal related installations
  • Unreal Engine
  • Set Unreal Engine environment variable
  • Clone & Build AGRARSENSE
• ROS related installations
  • ROS1 and ROS2
  • Install ROSBridge
  • Custom ROS Messages
• Using & testing AGRARSENSE
  • Visualizing sensor output with rviz
  • Creating packaged build
  • Modifying AGRARSENSE source

Documentation

Topic	Link
Getting Started	Getting Started
Using the Simulator	Using the Simulator
Using Vehicles	Using Vehicles
Using Walkers	Using Walkers
Sensors	Sensors
Creating own maps	Creating own maps
Building from source on Linux	Building on Linux
Building from source on Windows	Building on Windows
Installing on Linux	Installing on Linux
Installing on Windows	Installing on Windows

Prerequisites

In this section, you will find essential information regarding the system requirements, software installations, and other dependencies necessary before starting the development of AGRARSENSE.

The project directory structure is as follows:

├── AGRARSENSE [root]
    └── Data (Automatically created when simulator is launched for the first time)
        ├── All data saved from the Simulator
    └── Docs
        ├── All documentation files
    └── Examples
        └── ExampleJsonFiles
        └── ROS1 Python examples
        └── ROS2 Python examples
    └── Utils
        ├── Build scripts for Agrarsense unreal project
    └── Unreal
        └── Agrarsense [Unreal project files]
            └── Content
                ├── All blueprints, models and other assets
            └── Plugins
                ├── Plugins C++ source
            └── Source
                ├── Agrarsense simulator C++ source
            ├── LinuxCompile.sh         - build script for compiling the project for Linux
            ├── LinuxOpenProject.sh     - build script for opening the project for Linux
            ├── LinuxCreatePakcage.sh   - build script for creating packaged build for Linux

System Recommendation

• 9th gen Intel i5/i7/i9 or newer / 2nd gen AMD Ryzen 5/7/9 or newer
• 16+ GB RAM or more
• NVIDIA RTX 3070/3080 or newer / AMD RX 6700 XT or newer
• 150+ GB disk space for source version
• Ubuntu 20.04 / Windows 10 / Windows 11

Quick Guide

1. Install pre-compiled Unreal Engine 5.3 for Linux or follow Unreal Engine section to install AGRARSENSE fork of the Engine.
2. Clone AGRARSENSE repository

git clone https://AMKFrostBit@dev.azure.com/AMKFrostBit/AGRARSENSE/_git/Agrarsense

1. Compile and open the project:

# run LinuxOpenProject.sh

1. Optional: Install ROS, follow ROS related installations section

Unreal related installations

Unreal Engine

You can either install pre-compiled Unreal Engine 5.3 for Linux by following this guide, or you can install our fork of the Unreal Engine which is preferred.

We have made our own fork of the Unreal Engine where we have made few optimizations to the engine to improve sensors performance.

To install our fork of the engine, you need to have a GitHub account linked to Unreal Engine's account.

Note. Compiling Unreal Engine from source will take ~1-4 hours depending on your CPU!

To build our fork of the Unreal Engine:

1. Clone the repository

git clone -b AGRARSENSE https://github.com/frostbitunreal/unrealengine ~/UnrealEngine

2. Navigate to cloned folder:

cd ~/UnrealEngine

3. Run the configuration scripts and compile the engine:

If you are installing the Unreal Engine for the first time on your machine, these commands will to prompt register Unreal Engine types and ask for sudo password. Click yes and enter sudo password to be able to compile the engine.

./Setup.sh && ./GenerateProjectFiles.sh && make

If you modify the Unreal Engine source after first compilation, you must compile the Unreal Engine again with following command. Unreal Engine will only compile the parts you have modified, not the entire Unreal Engine.

make

Set Unreal Engine environment variable

In order to AGRARSENSE to find the correct installation of Unreal Engine, we need to set the environment variable.

To set the variable for this session only:

export UE5_ROOT=~/UnrealEngine

To set the variable so it persists across sessions:

1. Open ~/.bashrc or ./profile.

gedit ~/.bashrc
# or 
gedit ~/.profile

2. Add the following line to the bottom of the file:

export UE5_ROOT=~/UnrealEngine

3. Save the file and reset the terminal.

Clone & Build AGRARSENSE

1. Clone the repository

git clone https://AMKFrostBit@dev.azure.com/AMKFrostBit/AGRARSENSE/_git/Agrarsense

2. Navigate to AGRARSENSE Unreal project folder

cd Agrarsense/Unreal/Agrarsense/

3. Build AGRARSENSE

To compile and open the Unreal Editor with one command

./LinuxOpenProject.sh

ROS related installations

ROS1 and ROS2

ROS noetic is required for the project to be able to communicate between Unreal Engine and ROS. The simulator supports ROS2 via ROS1 to ROS2 bridge.

ROS noetic is only offically supported on Ubuntu 18.04 and 20.04. Ubuntu 22.04 doesn't offically support ROS noetic.

AGRARSENSE Unreal project will work without a ROS installation but you cannot get or send any data to or from the simulator.

Install ROS noetic:

sudo sh -c 'echo "deb http://packages.ros.org/ros/ubuntu $(lsb_release -sc) main" > /etc/apt/sources.list.d/ros-latest.list'
 
sudo apt install curl
 
curl -s https://raw.githubusercontent.com/ros/rosdistro/master/ros.asc | sudo apt-key add -
 
sudo apt update
 
sudo apt install ros-noetic-desktop-full -y
 
echo "source /opt/ros/noetic/setup.bash" >> ~/.bashrc
 
source ~/.bashrc

Install ROSBridge

https://github.com/code-iai/ROSIntegration states this: The recommended way to install rosbridge is from source (requires the rosauth package) – Even though you could install rosbridge using apt, there have been numersous issues where these apt packages do not reflect the code in the ros1 branch. Hence, it is best to install from source.

Install from apt packages:

sudo apt-get install ros-noetic-rosauth
sudo apt-get install ros-noetic-rosbridge-suite
echo "alias ROS1='source /opt/ros/noetic/setup.bash'" >> ~/.bashrc

When launching ROS server, you can launch it manually from Ubuntu Terminal:

roslaunch rosbridge_server rosbridge_tcp.launch bson_only_mode:=True

To run python scripts with ROS connection you need to install Python ROS packages:

pip install --extra-index-url https://rospypi.github.io/simple/ rospy
pip install --extra-index-url https://rospypi.github.io/simple/ tf2_ros

If connection doesn't go through from windows to Ubuntu WSL, you need to set **"ROS_IP"** in command prompt.

set ROS_IP=192.168.x.x

When ROS is installed follow this guide to install ROS2 (Foxy).

First, you need to install ROS2 without ros1_bridge.

First Launch new terminal and do following:

locale  # check for UTF-8
 
sudo apt update && sudo apt install locales
sudo locale-gen en_US en_US.UTF-8
sudo update-locale LC_ALL=en_US.UTF-8 LANG=en_US.UTF-8
export LANG=en_US.UTF-8
locale  # verify settings

sudo apt install software-properties-common
sudo add-apt-repository universe

sudo apt update && sudo apt install curl
sudo curl -sSL https://raw.githubusercontent.com/ros/rosdistro/master/ros.key -o /usr/share/keyrings/ros-archive-keyring.gpg

echo "deb [arch=$(dpkg --print-architecture) signed-by=/usr/share/keyrings/ros-archive-keyring.gpg] http://packages.ros.org/ros2/ubuntu $(. /etc/os-release && echo $UBUNTU_CODENAME) main" | sudo tee /etc/apt/sources.list.d/ros2.list > /dev/null

sudo apt update && sudo apt install -y \
  libbullet-dev \
  python3-pip \
  python3-pytest-cov \
  ros-dev-tools
 
python3 -m pip install -U \
  argcomplete \
  flake8-blind-except \
  flake8-builtins \
  flake8-class-newline \
  flake8-comprehensions \
  flake8-deprecated \
  flake8-docstrings \
  flake8-import-order \
  flake8-quotes \
  pytest-repeat \
  pytest-rerunfailures \
  pytest
  
sudo apt install --no-install-recommends -y \
  libasio-dev \
  libtinyxml2-dev
  
sudo apt install --no-install-recommends -y \
  libcunit1-dev

mkdir -p ~/ros2_foxy/src
cd ~/ros2_foxy
vcs import --input https://raw.githubusercontent.com/ros2/ros2/foxy/ros2.repos src

sudo apt upgrade

sudo rosdep init
rosdep update
rosdep install --from-paths src --ignore-src -y --skip-keys "fastcdr rti-connext-dds-5.3.1 urdfdom_headers"

cd ~/ros2_foxy/
colcon build --symlink-install --packages-skip ros1_bridge

After ROS2 is build from source (without ros1_bridge), You need to 'source' both ROS1 and ROS2.

source /opt/ros/noetic/setup.bash
source ~/ros2_foxy/install/setup.bash

then build only ros1_bridge

cd ~/ros2_foxy/
colcon build --symlink-install --packages-select ros1_bridge --cmake-force-configure

NOTE: To use ROS1 rosbridge_server, you shouldn't add both ROS1 and ROS2 to ~.bashrc, since it mixes the packages. Instead, use ROS1 as intended and source ROS2 ONLY when needed

echo "alias ROS2='source ~/ros2_foxy/install/setup.bash'" >> ~/.bashrc

By typing ROS2 in terminal, it sources the ROS2 bash file and removes ROS2 alias

Custom ROS Messages

1. Clone AGRARSENSE ROS messages

   cd ..
   git clone https://AMKFrostBit@dev.azure.com/AMKFrostBit/AGRARSENSE/_git/AGRARSENSE_ROS_Messages

2. Build ROS1 messages

Type "ROS1" to launch ros2 build bash. (Alias made earlier in .bashrc)

ROS1
cd ~/AGRARSENSE_ROS_Messages/ros1ws/
catkin_make

when the ROS1 build is finished close the wsl windows

1. Open New Terminal

Type "ROS2" to launch ros2 build bash. (Alias made earlier in .bashrc)

ROS2
cd ~/AGRARSENSE_ROS_Messages/ros2ws/
colcon build

when the ROS2 build is finished close the wsl windows

Rebuild ROS1_Bridge again

Launch new Terminal and Source both ROS1 and ROS2 and both workspaces:

ROS1 && ROS2
source ~/AGRARSENSE_ROS_Messages/ros1ws/devel/setup.bash
source ~/AGRARSENSE_ROS_Messages/ros2ws/install/setup.bash

Next navigate to ros2_foxy directory and rebuild bridge

cd ~/ros2_foxy/
colcon build --packages-select ros1_bridge --cmake-force-configure

After the bridge has successfully built

echo "'source ~/AGRARSENSE_ROS_Messages/ros1ws/devel/setup.bash'" >> ~/.bashrc

You should now be able to launch the ROS1 bridge that Simulator uses to communicate with ROS1.

sudo apt install python3-roslaunch
 
ROS1 && roslaunch rosbridge_server rosbridge_tcp.launch bson_only_mode:=True

Using & testing AGRARSENSE

To test everything works correctly you can the following:

1. Open new terminal and Launch Unreal Editor

cd ~/Agrarsense/Unreal/Agrarsense && ./LinuxOpenProject.sh

1. Start ROS1 with new terminal

roslaunch rosbridge_server rosbridge_tcp.launch bson_only_mode:=True

1. Start ROS1 to ROS2 Bridge with new terminal

ROS2 && ros2 run ros1_bridge dynamic_bridge --bridge-all-topics

1. In Unreal Editor navigate to Content/Agrarsense/Maps/TestMaps/SensorShowCase and open SensorShowCaseMap level
2. Press play
3. Spawn wanted sensors through the user interface
4. Open new terminal and type following command to see all current ROS topics

rostopic list

Visualizing sensor output with rviz

rviz is short for ROS Visualization. It's easy way visualize ROS messages, for example camera images and Lidar pointclouds.

1. First follow Testing AGRARSENSE
2. Open new terminal and launch rviz

rviz

1. Now that rviz is open click 'Add' from the bottom left corner. This opens new window. In the window click 'By topic' tab. This tab should show all topics that rviz is able to visualize. For example if you have Lidar sensor active in the simulation, you should see /Agrarsense/Sensors/Lidar0/PointCloud2. Select that and click ok.
2. Click 'Fixed Frame' and rename it to world and you should now see Lidar sensor pointcloud being visualized.

Creating packaged build

To create packaged shipping build for disturbution, you can either:

A. Create package without opening the Unreal Editor with following command. This script packages the project into ./Build folder

./LinuxCreatePackage.sh

B. or launch the Unreal Editor with following command and create the package in the Unreal Editor.

./LinuxOpenProject.sh

Maps to include in the packaged build is defined in DefaultGame.ini file.

Modifying AGRARSENSE source

When modifying Blueprints or Widget Blueprints in Unreal Editor, simply click compile and save in the Unreal Editor as in with any Unreal project.

When modifying AGRARSENSE Unreal project C++ source or plugins C++ source, you must close the Unreal Editor, compile C++ and then open the Unreal Editor again or else changes will not take effect. This can be done with same scripts as above.

Note. Unreal Engine does support compiling C++ in Unreal Editor through Live Coding but your experience may vary.

To compile and open the Unreal Editor with one command:

./LinuxOpenProject.sh

To just compile C++ source and plugins:

./LinuxCompile.sh