Initial support to the new Gazebo #3634

ahcorde · 2023-06-16T14:56:58Z

Basic Info

Info	Please fill out this column
Ticket(s) this addresses	#2997

Description of contribution in a few bullet points

Modified tb3_simulation_launch.py launch file to use the new Gazebo. It includes some new dependencies: gz_ros_bridge, gz_ros_sim and gz_ros2_control with diff_drive_controller and joint_state_broadcaster controllers

Test it

ros2 launch nav2_bringup tb3_simulation_launch.py use_gz:=True robot_sdf:=`ros2 pkg prefix nav2_bringup`/share/nav2_bringup/urdf/gz_turtlebot3_waffle.urdf world:=`ros2 pkg prefix nav2_bringup`/share/nav2_bringup/worlds/gz_world_only.model

ros2 launch nav2_bringup tb3_simulation_launch.py use_gz:=False headless:=False

mergify · 2023-06-16T14:58:35Z

@ahcorde, your PR has failed to build. Please check CI outputs and resolve issues.
You may need to rebase or pull in main due to API changes (or your contribution genuinely fails).

Signed-off-by: Alejandro Hernández Cordero <ahcorde@gmail.com>

nav2_bringup/launch/tb3_simulation_launch.py

nav2_bringup/urdf/turtlebot3_waffle.urdf

nav2_bringup/urdf/gz_turtlebot3_waffle.urdf

nav2_bringup/config/control.yaml

SteveMacenski · 2023-06-16T16:30:22Z

Thanks! The screen cap looks great ❤️

Signed-off-by: Alejandro Hernández Cordero <ahcorde@gmail.com>

mergify · 2023-06-16T23:23:43Z

@ahcorde, your PR has failed to build. Please check CI outputs and resolve issues.
You may need to rebase or pull in main due to API changes (or your contribution genuinely fails).

nav2_bringup/CMakeLists.txt

nav2_bringup/urdf/gz_turtlebot3_waffle.urdf

Signed-off-by: Alejandro Hernández Cordero <ahcorde@gmail.com>

SteveMacenski

Otherwise, beyond the other open item LGTM

nav2_bringup/launch/tb3_gz_simulation_launch.py

nav2_bringup/urdf/gz_turtlebot3_waffle.urdf

Signed-off-by: Alejandro Hernández Cordero <ahcorde@gmail.com>

SteveMacenski · 2023-06-20T18:41:56Z

Check CI for linting errors, I think when you adjusted the copyright you misformatted it a little

SteveMacenski

Otherwise LGTM.

nav2_bringup/urdf/gz_turtlebot3_waffle.urdf

SteveMacenski · 2023-06-20T18:47:58Z

I want to have someone else take a look for me as well on this one since there's alot of small subtle things that could cause problems + I know someone that's been working with new-gazebo alot recently. @bperseghetti could you do the honors?

Signed-off-by: Alejandro Hernández Cordero <ahcorde@gmail.com>

ahcorde · 2023-06-20T21:16:03Z

This one is fixed, odom frame is in the right place

SteveMacenski · 2023-06-20T21:46:07Z

This one is fixed, odom frame is in the right place

Huh?

ahcorde · 2023-06-20T21:47:47Z

This one is fixed, odom frame is in the right place

Huh?

it's weird, but this one is fixed, I'm trying to fix the other turtlebot3 model in the nav2_simple_commander package

SteveMacenski · 2023-06-20T21:53:58Z

Sorry, I read this one before the simple commander one 😄

Signed-off-by: Alejandro Hernández Cordero <ahcorde@gmail.com>

SteveMacenski · 2023-06-21T14:56:52Z

The min range cannot be zero? Why is that?

ahcorde · 2023-06-21T21:29:10Z

The min range cannot be zero? Why is that?

it's an OGRE(2) restriction:

what():  OGRE EXCEPTION(2:InvalidParametersException): Near clip distance must be greater than zero. in Frustum::setNearClipDistance at ./OgreMain/src/OgreFrustum.cpp (line 127)

azeey · 2024-04-25T02:15:22Z

Gazebo should now be available on rolling from the ros2-testing repo. I tried to test this PR on noble, but I'm getting build failures. I see #4267 will fix one of the failures, but I'm also getting the following warning, which is being treated as an error

In file included from /opt/ros/rolling/include/behaviortree_cpp/behavior_tree.h:23,
                 from /opt/ros/rolling/include/behaviortree_cpp/bt_factory.h:28,
                 from ~/ws/nav2_ws/src/navigation2/nav2_behavior_tree/plugins/control/recovery_node.cpp:128:
/opt/ros/rolling/include/behaviortree_cpp/controls/switch_node.h: In lambda function:
/opt/ros/rolling/include/behaviortree_cpp/controls/switch_node.h:91:15: error: declaration of ‘ports’ shadows a previous local [-Werror=shadow]
   91 |     PortsList ports;
      |               ^~~~~
/opt/ros/rolling/include/behaviortree_cpp/controls/switch_node.h:90:20: note: shadowed declaration is here
   90 |   static PortsList ports = []() {
      |                    ^~~~~
cc1plus: all warnings being treated as errors

Is this a known issue?

SteveMacenski · 2024-04-25T21:13:14Z

Known and fixed! BehaviorTree/BehaviorTree.CPP#811 I inquired about when we can expect a release but its in BT.CPP source. You can either disable the aggressive build flags we have in https://github.com/ros-planning/navigation2/blob/main/nav2_common/cmake/nav2_package.cmake#L40 or build BT.CPP from source to get past that

Signed-off-by: Steve Macenski <stevenmacenski@gmail.com>

SteveMacenski · 2024-05-28T22:21:37Z

CI is now on Rolling/24.04 if you rebase. We can now actually test with Harmonic and this PR in simulation in CI

nav2_bringup/worlds/gz_world_only.sdf.xacro

Co-authored-by: Addisu Z. Taddese <addisuzt@intrinsic.ai> Signed-off-by: Steve Macenski <stevenmacenski@gmail.com>

SteveMacenski · 2024-05-28T23:04:04Z

@azeey any thoughts on the get_package_share_directory('turtlebot3_gazebo')s? There's no turtlebot3-anything in 24.04 (except msgs, which I'm not ensurely sure what those are for) so we can't get `turtlebot3_common, gazebo, the sandbox world, probably)

Edit Looking at the job, its looking for gazebo ros pkgs which it'll never find: https://build.ros2.org/job/Rbin_uN64__turtlebot3_gazebo__ubuntu_noble_amd64__binary/38/console

azeey · 2024-05-28T23:10:35Z

@azeey any thoughts on the get_package_share_directory('turtlebot3_gazebo')s? There's no turtlebot3-anything in 24.04 (except msgs, which I'm not ensurely sure what those are for)

I reached out to the maintainers two weeks ago about migrating to the new Gazebo, but no word so far. We've already created issues (ROBOTIS-GIT/turtlebot3_simulations#196, ROBOTIS-GIT/turtlebot3_simulations#179) and PRs (ROBOTIS-GIT/turtlebot3_simulations#180) to help make the transition, but there has been no updates.

SteveMacenski · 2024-05-28T23:14:16Z

Any thoughts about moving to the TB4? I know there's no Iron/Jazzy/Rolling versions yet, but I believe their humble is based on Ignition-Gazebo so it should be relatively less of a lift. I actually emailed Ryan at CPR today asking him about providing some engineering support to getting the TB4 for Jazzy/Rolling for this purpose. That could cut this problem out and upgrade to the more modern robot option which I prefer greatly and is still a 'community standard' / OSRF affiliated.

or do you know of another robot with a new-gazebo simulation that could be lifted to Jazzy/Rolling for our use here? Another option is that I can create a "simplified" TB4 that we control in the ros-navigation org that removes dependencies on icreate and other software packages that would need to be updated which we don't need. That's what we ended up doing with the TB3 which I was hoping to avoid here, but may be necessary.

I suppose we also need a new simulation environment since the sandbox is tied to TB3. We discussed some warehouses, I think that's a good move and I just need to look into the 'how's and refamiliarize myself with the threads on that.

Edit: good sign, CI is happy on this PR minus a test that is unrelated to this I need to look into. That's on 24.04/Rolling and was able to resolve all the dependencies then fine.

SteveMacenski · 2024-05-29T00:06:43Z

I was curious if it were possible to use a Fuel world as part of a description file and remove some element from it.

The AWS warehouse world model is great, but the manual forklift in the aisle isn't what I'd want for a default bringup to do more traditional demos. I actually prefer the Depot model, but the trusses and fan in the ceiling make it difficult for a user to see what's going on if they're setting the gazebo client on to watch the robot in simulation from above - so I'd want to remove those.

I don't recall any subtractive-operations in the SDF format, but maybe there are (or can override existing object names with new ones that I set its color properties to transparent?)

azeey · 2024-05-29T21:31:25Z

I was curious if it were possible to use a Fuel world as part of a description file and remove some element from it.

It is possible to include a model (only works with models not worlds) and remove some elements from it in SDFormat. The feature to do this is under an experimental XML namespace even though it's been around for a while. It should become a permanent feature, we just haven't gotten around to it yet. If you are okay with using that, then I can switch to the Depot model and remove the trusses and fan.

Any thoughts about moving to the TB4? I know there's no Iron/Jazzy/Rolling versions yet, but I believe their humble is based on Ignition-Gazebo so it should be relatively less of a lift.

I'm willing to provide support to make that happen. I looked to see if it would be quick but I realized it requires changes in https://github.com/iRobotEducation/create3_sim as well.

In the meantime, I have a fork of turtlebot3 that works on Rolling. We can build that from source if you'd like.

SteveMacenski · 2024-05-29T23:35:20Z

I looked to see if it would be quick but I realized it requires changes in https://github.com/iRobotEducation/create3_sim as well.

I realized the same thing which is what led me to: 'maybe we can just make a minimum nav2_tb4_description package that's self contained.' Too many cooks in the kitchen and too few have interest in keeping up with the latest in ROS 2. Doesn't need to be nav2 specific even, but figured I'd just open a new repo in ros-navigation with the minimal TB4 stuff and release that. I'm open to other more generic options though.

In the meantime, I have a fork of turtlebot3 that works on Rolling. We can build that from source if you'd like.

That's a good interim solution, I will test that after I get this comment out. I wonder if its better not to just replace it all in the first salvo, but this PR would be done if we could have that built with CI and workable. Then someone in parallel can work on using that for the system tests while I work on updating the robot model / world (or I'll get to it too). If you add your fork to https://github.com/ros-navigation/navigation2/blob/main/tools/underlay.repos (uncommented) then it'll be grabbed and built for CI.

The major question with that is: is that fork going to be able to make it upstream + released for Jazzy/Rolling? If so, good solution, lets do it. We can use that for our system tests so it has real value not to have rewrite those. If this won't get out in binary format for TB3, then its probably just worth stripping out TB3 up front since adding it here now won't really buy us anything in terms of unblocking Nav2's release in binary format. This PR gets merged (which isn't nothing for sure), but it seems more organizational since we'd immediately need to strip it out again for something that can be released.

SteveMacenski · 2024-05-29T23:46:18Z

@azeey I'm unable to get it to build with the following error:

--- stderr: turtlebot3_gazebo                                 
In file included from /jazzy_ws/src/turtlebot3_simulations/turtlebot3_gazebo/models/turtlebot3_dqn_world/obstacle_plugin/ObstacleAnimator.hh:19,
                 from /jazzy_ws/src/turtlebot3_simulations/turtlebot3_gazebo/models/turtlebot3_dqn_world/obstacle_plugin/obstacle2.cc:17:
/opt/ros/rolling/opt/gz_sim_vendor/include/gz/sim8/gz/sim/System.hh:28:10: fatal error: gz/transport/parameters/Registry.hh: No such file or directory
   28 | #include <gz/transport/parameters/Registry.hh>
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.
In file included from /jazzy_ws/src/turtlebot3_simulations/turtlebot3_gazebo/models/turtlebot3_dqn_world/obstacle_plugin/ObstacleAnimator.hh:19,
                 from /jazzy_ws/src/turtlebot3_simulations/turtlebot3_gazebo/models/turtlebot3_dqn_world/obstacle_plugin/obstacles.cc:17:
/opt/ros/rolling/opt/gz_sim_vendor/include/gz/sim8/gz/sim/System.hh:28:10: fatal error: gz/transport/parameters/Registry.hh: No such file or directory
   28 | #include <gz/transport/parameters/Registry.hh>
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.
gmake[2]: *** [CMakeFiles/obstacle2.dir/build.make:76: CMakeFiles/obstacle2.dir/models/turtlebot3_dqn_world/obstacle_plugin/obstacle2.cc.o] Error 1
gmake[1]: *** [CMakeFiles/Makefile2:195: CMakeFiles/obstacle2.dir/all] Error 2
gmake[1]: *** Waiting for unfinished jobs....
gmake[2]: *** [CMakeFiles/obstacles.dir/build.make:76: CMakeFiles/obstacles.dir/models/turtlebot3_dqn_world/obstacle_plugin/obstacles.cc.o] Error 1
gmake[1]: *** [CMakeFiles/Makefile2:221: CMakeFiles/obstacles.dir/all] Error 2
In file included from /jazzy_ws/src/turtlebot3_simulations/turtlebot3_gazebo/models/turtlebot3_dqn_world/obstacle_plugin/ObstacleAnimator.hh:19,
                 from /jazzy_ws/src/turtlebot3_simulations/turtlebot3_gazebo/models/turtlebot3_dqn_world/obstacle_plugin/obstacle1.cc:17:
/opt/ros/rolling/opt/gz_sim_vendor/include/gz/sim8/gz/sim/System.hh:28:10: fatal error: gz/transport/parameters/Registry.hh: No such file or directory
   28 | #include <gz/transport/parameters/Registry.hh>
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.

azeey · 2024-05-30T00:49:12Z

@azeey I'm unable to get it to build with the following error:

--- stderr: turtlebot3_gazebo                                 
In file included from /jazzy_ws/src/turtlebot3_simulations/turtlebot3_gazebo/models/turtlebot3_dqn_world/obstacle_plugin/ObstacleAnimator.hh:19,
                 from /jazzy_ws/src/turtlebot3_simulations/turtlebot3_gazebo/models/turtlebot3_dqn_world/obstacle_plugin/obstacle2.cc:17:
/opt/ros/rolling/opt/gz_sim_vendor/include/gz/sim8/gz/sim/System.hh:28:10: fatal error: gz/transport/parameters/Registry.hh: No such file or directory
   28 | #include <gz/transport/parameters/Registry.hh>
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.
In file included from /jazzy_ws/src/turtlebot3_simulations/turtlebot3_gazebo/models/turtlebot3_dqn_world/obstacle_plugin/ObstacleAnimator.hh:19,
                 from /jazzy_ws/src/turtlebot3_simulations/turtlebot3_gazebo/models/turtlebot3_dqn_world/obstacle_plugin/obstacles.cc:17:
/opt/ros/rolling/opt/gz_sim_vendor/include/gz/sim8/gz/sim/System.hh:28:10: fatal error: gz/transport/parameters/Registry.hh: No such file or directory
   28 | #include <gz/transport/parameters/Registry.hh>
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.
gmake[2]: *** [CMakeFiles/obstacle2.dir/build.make:76: CMakeFiles/obstacle2.dir/models/turtlebot3_dqn_world/obstacle_plugin/obstacle2.cc.o] Error 1
gmake[1]: *** [CMakeFiles/Makefile2:195: CMakeFiles/obstacle2.dir/all] Error 2
gmake[1]: *** Waiting for unfinished jobs....
gmake[2]: *** [CMakeFiles/obstacles.dir/build.make:76: CMakeFiles/obstacles.dir/models/turtlebot3_dqn_world/obstacle_plugin/obstacles.cc.o] Error 1
gmake[1]: *** [CMakeFiles/Makefile2:221: CMakeFiles/obstacles.dir/all] Error 2
In file included from /jazzy_ws/src/turtlebot3_simulations/turtlebot3_gazebo/models/turtlebot3_dqn_world/obstacle_plugin/ObstacleAnimator.hh:19,
                 from /jazzy_ws/src/turtlebot3_simulations/turtlebot3_gazebo/models/turtlebot3_dqn_world/obstacle_plugin/obstacle1.cc:17:
/opt/ros/rolling/opt/gz_sim_vendor/include/gz/sim8/gz/sim/System.hh:28:10: fatal error: gz/transport/parameters/Registry.hh: No such file or directory
   28 | #include <gz/transport/parameters/Registry.hh>
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.

That's really strange. I tried building it multiple times both on rolling and jazzy, but could not reproduce what you're getting.

SteveMacenski · 2024-05-30T01:29:05Z

I’m in a docker container based on ros-rolling-desktop-full & obtained gazebo from the ROS 2 integration instructions via apt install ros-rolling-ros-gz and then all the other rolling items related to Nav2. I pulled that base image in on Friday from dockerhub. What’s your setup?

azeey · 2024-05-30T02:55:39Z

Okay, I tracked it down to a missing dependency, cppzmq-dev. If you install that, the build should succeed.

What's strange is that if you start with a rolling-ros-core image and install ros-rolling-ros-gz, you'll get cppzmq-dev automatically. I assume it will be the same if you start with an ubuntu:noble image and install ros-rolling-desktop-full. I think the issue is that cppzmq-dev is a recommended package of libzmq3-dev, which gz-transport depends on. So, I wonder if the desktop-full docker image is created with apt --no-install-recommends.

Either way, gz-transport should declare a direct dependency on cppqmq. There is no rosdep key for cppzmq, so I'll have to create one and add that to gz-transport's package.xml. I'll plan on doing that tomorrow.

azeey · 2024-05-30T15:13:43Z

It turns out cppzmq-dev is a new package in Ubuntu Noble. Potential rosdep fix in ros/rosdistro#41413

SteveMacenski · 2024-05-30T17:22:16Z

So, I wonder if the desktop-full docker image is created with apt --no-install-recommends.

My docker image does the following as I'm dog fooding the Docker development tutorial on docs.nav2.org during this exercise

RUN apt update \
    && DEBIAN_FRONTEND=noninteractive apt install -y --no-install-recommends --no-install-suggests \
  ros-dev-tools \
  wget

A quick google search shows that in use all over the place in ROS-Docker-World and I never had reason before to question it. Should I remove that from my tutorial?

OK - I assume that new key will be added to some package(s) that will hit jazzy/rolling for the next sync?

That did make things work! I'm taking a look now and seeing the following:

Gazebo Classic: GUI 30% CPU, Server 105% CPU
Gazebo Harmonic: GUI 160% CPU, Server 120% CPU

I think the 105-120% on the server is fine enough not worth nitpicking, but I was surprised to see the GUI be 5x as consuming with these configs. Is there a way to reduce that to even just 2x?

I'll be keeping this up for a bit, sending goals, relaunching, etc to check stability but I don't expect to find anything to report. Mostly just for my own education.

^C[WARNING] [launch]: user interrupted with ctrl-c (SIGINT)
[parameter_bridge-2] [INFO] [1717089793.776311576] [rclcpp]: signal_handler(signum=2)
[ERROR] [ruby $(which gz) sim-4]: process has died [pid 176564, exit code -2, cmd 'ruby $(which gz) sim -r -s /tmp/nav2__ca33khs.sdf --force-version 8'].
[INFO] [robot_state_publisher-6]: process has finished cleanly [pid 176566]
[INFO] [parameter_bridge-2]: process has finished cleanly [pid 176562]
[ERROR] [ruby $(which gz) sim-5]: process has died [pid 176565, exit code -2, cmd 'ruby $(which gz) sim -v4 -g  --force-version 8'].

I don't love the crashing on shutdown, but I'm sure this is something the GZ team is aware of so I don't need to belabor it.

Thoughts on the steps forward for your TB3 fork? I think the path here boils down to if you think there's a reasonable chance that binaries will be possible. If so, then we can add this to our repos file and unlock the work for updating our system tests to use your work to get that part of our CI up and running. I'll probably still want to replace the default bringup before releasing to Jazzy with a new robot platform / environment, but that's non-blocking for this PR and squarely in the "Steve's problem" camp (though would love your reviews). The TB4 is the direction I think is most sensible unless anyone has a better idea.

azeey · 2024-05-30T21:00:20Z

My docker image does the following as I'm dog fooding the Docker development tutorial on docs.nav2.org during this exercise

The --no-install-recommends thing is why we didn't catch it in our testing. I'm not saying you should stop using it.

But the root cause of the issue is that the C++ headers were removed from libzmq3-dev in Noble and placed in their own package. My proposed fix is to update the libzmq3-dev rosdep key to include cppzmq-dev. If that gets merged, the fix would be available immediately (or when the docker image is rebuilt). The maintainers might decide it's better to add a new key for cppzmq. In that case it will be available in the next sync.

I think the 105-120% on the server is fine enough not worth nitpicking, but I was surprised to see the GUI be 5x as consuming with these configs. Is there a way to reduce that to even just 2x?

I think we'll need to profile the GUI to figure out what's happening. I don't see a quick fix for this.

I don't love the crashing on shutdown, but I'm sure this is something the GZ team is aware of so I don't need to belabor it.

This is caused by a bug in launch: ros2/launch#757. We're working on better launch integration (see gazebosim/ros_gz#552 and gazebosim/ros_gz#544) that will hopefully resolve this.

Thoughts on the steps forward for your TB3 fork?

Maybe @clalancette can chime in on this. I'm not sure if we are at the point of replacing the upstream package, but if we decide to go that route, we should probably move that repo into an OSRF org.

Honestly, the easiest thing to do would be to put the bits we need from turtlebot3_gazebo into nav2. I see y'all already have the files in https://github.com/ros-navigation/navigation2/tree/main/nav2_system_tests/models/turtlebot3_waffle and https://github.com/ros-navigation/navigation2/tree/main/nav2_system_tests/models/turtlebot3_burger, so I don't see why we couldn't do that more broadly instead of just for tests.

SteveMacenski · 2024-05-30T23:07:19Z

Honestly, the easiest thing to do would be to put the bits we need from turtlebot3_gazebo into nav2. I see y'all already have the files in https://github.com/ros-navigation/navigation2/tree/main/nav2_system_tests/models/turtlebot3_waffle and https://github.com/ros-navigation/navigation2/tree/main/nav2_system_tests/models/turtlebot3_burger, so I don't see why we couldn't do that more broadly instead of just for tests.

Good point. I'd be fine with that. Want to do that? That would then unblock re-enabling the system tests using it + GZ Harmonic. I don't love owning all of TB3 stack for Nav2 Bringup, but the systems test package is a little ... messy to begin with and as you point out we already have a bunch of files anyway. I'm not commiting you to updating all our system tests with Harmonic in this PR or ever, but certainly if you were willing to, I wouldn't object.

Our bringup still needs some work for getting a robot that is distributed in binary format, but I'm actively working on that now. I've created a set of minimal TB4 description + simulation files for Nav2 based on the CPR + iRobot files and your work here. Their repos have a bunch of other stuff we don't need which are blocking updates + releases for Jazzy/Rolling so I figure stripping out to just the bare minimum will help make it easy to forward port things in a distribution-agnostic way. Its in https://github.com/ros-navigation/nav2_min_tb4_sim and my plan is to release these in binary format then we can dep on that for nav2_bringup so that package doesn't have robot specific files in it but is able to bring up with the TB4's. Porting the existing TB4 packages to Jazzy involves porting a bunch of custom ignition plugins and panels which is more than I'm able to commit to, so this seems like a good way forward and when TB4 catches up, we can switch to that.

This way in the future, we're not dependent on companies supporting the bleeding edge, we can just have these packages in rolling and should always work (unless launch / URDF format changes).

SteveMacenski · 2024-06-03T21:46:13Z

@azeey what do you think? Should we replace all the existing TB3 files with your fork in the system tests and possibly port ~1 over to it as an example? Then I think we can merge this in and I or other community members can come in to work on porting the rest of the system tests + update the bringup package with a robot in binary format.

SteveMacenski · 2024-06-05T18:05:42Z

@ahcorde @azeey merging. I'm pretty far into a TB4 analog with the TB3 for both options in modern GZ

One thing I've noticed is that the simulation takes a long time to initialize with the depot world and the TB4. I haven't looked into it, but after the migration is done, that is worth me/someone looking into.

Path forward:

I'd really appreciate it if we could get a simulation test running with the TB3 as an example. With a blueprint, I can probably find a community contributor open to doing the rest of them if there's a template to start from
I'm going to release https://github.com/ros-navigation/nav2_minimal_turtlebot4_simulation/tree/main & add in the TB4 launch file
Clean up tickets Complete Port to modern Gazebo List #4399 Remove old Gazebo stuff once new gazebo lands #4386
Bringup time optimization (@bperseghetti related to your SDF / frame work perhaps?)

SteveMacenski · 2024-06-05T21:50:56Z

I refactored the TB4 into an independent simulation package in the minimal repository linked above. I also took the TB3 work done here and refactored that + the turtlebot3_gazebo assets into its own nav2_minimal_tb3_sim package so that it contains only the minimum things required without external dependencies so it can be released into rolling and available in all future distributions without work. I then refactored your work @ahcorde and @azeey in the PR linked below to move to that simulation package and link it back to the bringup in Nav2 to vendorize out the spawning of the specific robot platform in line with the TB4 example.

#4401

Now we can ship both, in binary form, without issues waiting for either of the turtlebot vendors! Phew, this was a lot of refactoring for a day...

stevedanomodolor · 2024-06-06T04:57:47Z

@ahcorde @azeey merging. I'm pretty far into a TB4 analog with the TB3 for both options in modern GZ

One thing I've noticed is that the simulation takes a long time to initialize with the depot world and the TB4. I haven't looked into it, but after the migration is done, that is worth me/someone looking into.

Path forward:

I'd really appreciate it if we could get a simulation test running with the TB3 as an example. With a blueprint, I can probably find a community contributor open to doing the rest of them if there's a template to start from

I'm going to release https://github.com/ros-navigation/nav2_minimal_turtlebot4_simulation/tree/main & add in the TB4 launch file

Clean up tickets Complete Port to modern Gazebo List #4399 Remove old Gazebo stuff once new gazebo lands #4386

Bringup time optimization (@bperseghetti related to your SDF / frame work perhaps?)

Is there an example on how to run this yet?, so I can start porting the system tests too.

SteveMacenski · 2024-06-06T15:33:22Z

If you clone the nav2 minimal turtlebot repo into your workspace and build them, the TB3/4 demos should work!

ahcorde mentioned this pull request Jun 16, 2023

Migrate to Ignition / TB4 simulation #2997

Closed

5 tasks

Initial support to the new Gazebo

c52b2b0

Signed-off-by: Alejandro Hernández Cordero <ahcorde@gmail.com>

ahcorde force-pushed the ahcorde/initial_support/gazebo branch from ad975c0 to c52b2b0 Compare June 16, 2023 15:03

SteveMacenski reviewed Jun 16, 2023

View reviewed changes

Added feedback

fd50e02

Signed-off-by: Alejandro Hernández Cordero <ahcorde@gmail.com>

SteveMacenski reviewed Jun 17, 2023

View reviewed changes

nav2_bringup/CMakeLists.txt Outdated Show resolved Hide resolved

nav2_bringup/urdf/gz_turtlebot3_waffle.urdf Outdated Show resolved Hide resolved

Fixed build

2500c8c

Signed-off-by: Alejandro Hernández Cordero <ahcorde@gmail.com>

ahcorde requested a review from SteveMacenski June 19, 2023 11:02

SteveMacenski requested changes Jun 19, 2023

View reviewed changes

nav2_bringup/launch/tb3_gz_simulation_launch.py Show resolved Hide resolved

nav2_bringup/launch/tb3_gz_simulation_launch.py Outdated Show resolved Hide resolved

SteveMacenski requested changes Jun 19, 2023

View reviewed changes

nav2_bringup/urdf/gz_turtlebot3_waffle.urdf Outdated Show resolved Hide resolved

Added feeback

5215768

Signed-off-by: Alejandro Hernández Cordero <ahcorde@gmail.com>

ahcorde requested a review from SteveMacenski June 20, 2023 11:13

ahcorde added 2 commits June 20, 2023 17:12

Fixed physics tag

d3c421a

Signed-off-by: Alejandro Hernández Cordero <ahcorde@gmail.com>

Fixed physics tag

975bc78

Signed-off-by: Alejandro Hernández Cordero <ahcorde@gmail.com>

SteveMacenski reviewed Jun 20, 2023

View reviewed changes

nav2_bringup/urdf/gz_turtlebot3_waffle.urdf Outdated Show resolved Hide resolved

Fix copyright and urdf

469c2a2

Signed-off-by: Alejandro Hernández Cordero <ahcorde@gmail.com>

SteveMacenski approved these changes Jun 20, 2023

View reviewed changes

min range cannot be zero

34f466e

Signed-off-by: Alejandro Hernández Cordero <ahcorde@gmail.com>

SteveMacenski mentioned this pull request Apr 25, 2024

Build warning BehaviorTree/BehaviorTree.CPP#811

Closed

Merge branch 'main' into ahcorde/initial_support/gazebo

5bf14e2

Signed-off-by: Steve Macenski <stevenmacenski@gmail.com>

azeey reviewed May 28, 2024

View reviewed changes

nav2_bringup/worlds/gz_world_only.sdf.xacro Outdated Show resolved Hide resolved

Update nav2_bringup/worlds/gz_world_only.sdf.xacro

cc5bde0

Co-authored-by: Addisu Z. Taddese <addisuzt@intrinsic.ai> Signed-off-by: Steve Macenski <stevenmacenski@gmail.com>

SteveMacenski mentioned this pull request May 30, 2024

Possible regression in rcl preshutdown callbacks - context invalid? ros2/rclcpp#2547

Open

SteveMacenski mentioned this pull request Jun 4, 2024

Remove old Gazebo stuff once new gazebo lands #4386

Closed

SteveMacenski merged commit bc6e3b2 into ros-navigation:main Jun 5, 2024
9 of 10 checks passed

SteveMacenski mentioned this pull request Jun 11, 2024

Complete Port to modern Gazebo List #4399

Open

5 tasks

Initial support to the new Gazebo #3634

Initial support to the new Gazebo #3634

Conversation

ahcorde commented Jun 16, 2023

Basic Info

Description of contribution in a few bullet points

Test it

mergify bot commented Jun 16, 2023

SteveMacenski commented Jun 16, 2023

mergify bot commented Jun 16, 2023

SteveMacenski left a comment • edited

Choose a reason for hiding this comment

SteveMacenski commented Jun 20, 2023

SteveMacenski left a comment

Choose a reason for hiding this comment

SteveMacenski commented Jun 20, 2023 • edited

ahcorde commented Jun 20, 2023

SteveMacenski commented Jun 20, 2023

ahcorde commented Jun 20, 2023

SteveMacenski commented Jun 20, 2023

SteveMacenski commented Jun 21, 2023 • edited

ahcorde commented Jun 21, 2023

azeey commented Apr 25, 2024

SteveMacenski commented Apr 25, 2024 • edited

SteveMacenski commented May 28, 2024 • edited

SteveMacenski commented May 28, 2024 • edited

azeey commented May 28, 2024

SteveMacenski commented May 28, 2024 • edited

SteveMacenski commented May 29, 2024

azeey commented May 29, 2024

SteveMacenski commented May 29, 2024 • edited

SteveMacenski commented May 29, 2024

azeey commented May 30, 2024

SteveMacenski commented May 30, 2024 • edited

azeey commented May 30, 2024

azeey commented May 30, 2024

SteveMacenski commented May 30, 2024 • edited

azeey commented May 30, 2024

SteveMacenski commented May 30, 2024 • edited

SteveMacenski commented Jun 3, 2024

SteveMacenski commented Jun 5, 2024

SteveMacenski commented Jun 5, 2024 • edited

stevedanomodolor commented Jun 6, 2024

SteveMacenski commented Jun 6, 2024

SteveMacenski left a comment •

edited

SteveMacenski commented Jun 20, 2023 •

edited

SteveMacenski commented Jun 21, 2023 •

edited

SteveMacenski commented Apr 25, 2024 •

edited

SteveMacenski commented May 28, 2024 •

edited

SteveMacenski commented May 28, 2024 •

edited

SteveMacenski commented May 28, 2024 •

edited

SteveMacenski commented May 29, 2024 •

edited

SteveMacenski commented May 30, 2024 •

edited

SteveMacenski commented May 30, 2024 •

edited

SteveMacenski commented May 30, 2024 •

edited

SteveMacenski commented Jun 5, 2024 •

edited