While there is some value in doing these courses in the Ignite system, I find I’m missing a TON of how I could actually do this on a production system, which is why I’m taking the Perception course.
So, I don’t have the ability to actually see what’s happening like in the upper right hand screen when I’m doing it on my own. I assume this is RVIZ, but where’s the config file. How would I go about doing all of this on my own?
Hi @rich,
Welcome to the community 
I recommend doing the ignite tutorials first. This gives you the knowledge on how things work. After that you can do it on your own system. I tried it the other way around and I wasted a lot of time. Knowing what different modules and programs are and how they work together is vital. If you finish the beginner path:
You should be knowledgeable enough to tackle your own system.
PS: in the academy, the top right is gazebo, not rviz
Awesome, thanks!
I did the Udacity 4 month course, so I have a pretty good idea of the basics. They didn’t want you using your own system either. It’s challenging, because if you want to do anything with what you learn, you have to be able to implement all the ROS paths, etc, and everything else that’s a pain.
I’ll definitely go through those tuts tonight to see how Ignite does it. I have to prepare a demo in 3 days, so I was REALLY hoping to have the Perception stuff ready.
If you really want to do it now I can give you some pointers. For that it is important to know which operating Ubuntu version you are running.
Ubuntu 18.04, ROS Melodic
I’m trying to show the people in Gazebo as in the demo passing simply camera info to them. Unit 7: People Tracking.
I went ahead and finally (I think) got all the modules and dependencies compiled. I just don’t know how to launch the gazebo world that is in the upper right to view.
Which robot/simulation are you trying to run?
And for proper ros installation with all required packages you can use these commands:
#install Ros Melodic
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-key adv --keyserver 'hkp://keyserver.ubuntu.com:80' --recv-key C1CF6E31E6BADE8868B172B4F42ED6FBAB17C654
sudo apt update
sudo apt install ros-melodic-desktop-full
sudo rosdep init
rosdep update
#install dependencies
sudo apt install python-rosinstall python-rosinstall-generator python-wstool build-essential
echo "source /opt/ros/melodic/setup.bash" >> ~/.bashrc
source ~/.bashrc
#installing controllers
sudo apt-get install ros-melodic-ros-controllers*
for visualization in general you run your robot in the gazebo physics simulation and read the camera output in rviz. Rviz is NOT a simulator, it is merely a tool to visualize topics (robot pose, laserscans, camera etc.)
Awesome, thank you!
I’m installed, trying to figure out what the commands are to load the gazebo world and simulations. Usually it’s something like gazebo.., but where’s the world located?
which robot / simulation are you trying to run?
Unit #7, so it looks like the Fetch robot and a person in the world.
I’m trying to find the Rviz setups as well to assign all the topics, which you can see from the video is custom: https://i.imgur.com/gVwq3EO.jpg
Alright, here is what you do:
1. Install fetch
source ~/.bashrc
cd ~
mkdir catkin_ws
cd catkin_ws
mkdir src
cd src
git clone https://github.com/fetchrobotics/fetch_gazebo.git
cd ~/catkin_ws/
rosdep install --from-paths src --ignore-src --rosdistro melodic -y
catkin_make
echo "source ~/catkin_ws/devel/setup.bash" >> ~/.bashrc
source ~/.bashrc
2. launch the simulation
roslaunch fetch_gazebo simulation.launch
rosrun rviz rviz
In rviz on the left you click on ‘ADD’ and add RobotModel and Camera.
Expand the camera and under Image Topic select one of the 3 available topics. You should now see the camera image.
I appreciate that!!
Everything worked for bringing up the Fetch robot! The catkin_make put everything under the folder ~/catkin_make/install
I had missed the rosdep on everything, I did it earlier on a per package basis.
Lastly, do you know how to get the person in the simulation? What commands?
I’m still having other issues on this, like how I would connect a live camera to this to publish to the topics, but I think I’ve asked enough for now on getting this live. Have to install a LOT of modules!
Thanks again!
Glad it worked
In gazebo on the top left you can just click on insert and manually add any model to the world. There is a sizeable example library gazebo comes with.
Fetch robot already comes with a camera, which you can read out in rviz