You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thank you so much for the reply :D
There is also another issue I am curious How you guys managed to run with Ros Rolling because if I follow your steps only (my python, ubuntu and ros version is all as you mention), I am faced with this error below:
`Starting >>> grid_map_rviz_plugin
--- stderr: grid_map_rviz_plugin
CMake Error at /usr/lib/x86_64-linux-gnu/cmake/Qt6Core/Qt6CoreVersionlessTargets.cmake:42 (message):
Some (but not all) targets in this export set were already defined.
Failed <<< grid_map_rviz_plugin [0.92s, exited with code 1]
`
This error first comes up with grid_map_rviz_plugin because rolling uses both qt5 and qt6 I believe, and if I continue-on-error, it continues to give similar errors on other qt related builds, how do I solve this?
I solved the issue of the colcon build errors and did successful run (with my own dataset). I can't access your dataset so I must train and test with my own dataset, but that faces an obstacle because of this nav_analyzer package is non-existent if I follow exactly your steps with success, can you help me @fmrico?
Hello @TanKarabudakk, please try this fork for now.
It is the most recent version of the project. I am currently finishing some updates and will open a pull request soon. I have also fixed the link to the bagfile there.
@midemig I was able to get your dataset and successfully run the demo. What I want to do is that I want to test my own dataset with your code. but to do so I must first perform training. I understand traversability updater is responsible for the training but there is no launch file for me to use
The VAE is already trained in genetic features, so there is no need to re-train for a different robot, it should work with your own dataset. Traversability_updater learns features of the terrain that the robot is in. By default, self.learning is set to true, so it should learn from your data. You can use the same launch file, and modify it with your dataset topics.
Hello @midemig, I am trying to fo as you say, and I did declared all the needed frames or topics correctly in core_params.yaml, but unlike cesped_00, I canmot see predictions done on grid map level, only the purple layer (no predictions), I would be happy if ypu cna provide me support
Activity
fmrico commentedon Dec 16, 2025
Hi @TanKarabudakk !! Thanks for your interest in this repo 🤓
@juanscelyg @midemig, May you check if this package should be available and isn't?
TanKarabudakk commentedon Dec 17, 2025
Thank you so much for the reply :D
There is also another issue I am curious How you guys managed to run with Ros Rolling because if I follow your steps only (my python, ubuntu and ros version is all as you mention), I am faced with this error below:
`Starting >>> grid_map_rviz_plugin
--- stderr: grid_map_rviz_plugin
CMake Error at /usr/lib/x86_64-linux-gnu/cmake/Qt6Core/Qt6CoreVersionlessTargets.cmake:42 (message):
Some (but not all) targets in this export set were already defined.
Targets Defined: Qt::Core
Targets not yet defined: Qt::CorePrivate
Call Stack (most recent call first):
/usr/lib/x86_64-linux-gnu/cmake/Qt6Core/Qt6CoreConfig.cmake:65 (include)
/usr/share/cmake-3.28/Modules/CMakeFindDependencyMacro.cmake:76 (find_package)
/usr/lib/x86_64-linux-gnu/cmake/Qt6/QtPublicDependencyHelpers.cmake:108 (find_dependency)
/usr/lib/x86_64-linux-gnu/cmake/Qt6Widgets/Qt6WidgetsDependencies.cmake:39 (_qt_internal_find_qt_dependencies)
/usr/lib/x86_64-linux-gnu/cmake/Qt6Widgets/Qt6WidgetsConfig.cmake:50 (include)
/usr/lib/x86_64-linux-gnu/cmake/Qt6/Qt6Config.cmake:167 (find_package)
/opt/ros/rolling/share/rviz_common/cmake/rviz_common-extras.cmake:34 (find_package)
/opt/ros/rolling/share/rviz_common/cmake/rviz_commonConfig.cmake:41 (include)
CMakeLists.txt:10 (find_package)
Failed <<< grid_map_rviz_plugin [0.92s, exited with code 1]
`
This error first comes up with grid_map_rviz_plugin because rolling uses both qt5 and qt6 I believe, and if I continue-on-error, it continues to give similar errors on other qt related builds, how do I solve this?
TanKarabudakk commentedon Dec 17, 2025
I solved the issue of the colcon build errors and did successful run (with my own dataset). I can't access your dataset so I must train and test with my own dataset, but that faces an obstacle because of this nav_analyzer package is non-existent if I follow exactly your steps with success, can you help me @fmrico?
midemig commentedon Dec 19, 2025
Hello @TanKarabudakk, please try this fork for now.
It is the most recent version of the project. I am currently finishing some updates and will open a pull request soon. I have also fixed the link to the bagfile there.
TanKarabudakk commentedon Dec 19, 2025
@midemig I was able to get your dataset and successfully run the demo. What I want to do is that I want to test my own dataset with your code. but to do so I must first perform training. I understand traversability updater is responsible for the training but there is no launch file for me to use
midemig commentedon Dec 23, 2025
The VAE is already trained in genetic features, so there is no need to re-train for a different robot, it should work with your own dataset. Traversability_updater learns features of the terrain that the robot is in. By default, self.learning is set to true, so it should learn from your data. You can use the same launch file, and modify it with your dataset topics.
TanKarabudakk commentedon Feb 24, 2026
Hello @midemig, I am trying to fo as you say, and I did declared all the needed frames or topics correctly in core_params.yaml, but unlike cesped_00, I canmot see predictions done on grid map level, only the purple layer (no predictions), I would be happy if ypu cna provide me support