Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Parameterization for narrow environments #2

Open
Taucrates opened this issue Jul 12, 2024 · 5 comments
Open

Parameterization for narrow environments #2

Taucrates opened this issue Jul 12, 2024 · 5 comments

Comments

@Taucrates
Copy link

Hi, good job here with this exploration framework, congratulations!!

I would like to test your algorithm in narrower environments such as the subt darpa final scenario or inside buildings, similar to the ones shown in the repository. However, I have checked different parameterizations for those environments (reducing resolution, reducing robot size, changing min_info_goal, changing depth, etc) but unfortunately it has not been possible to make the MAV explore more than a few meters. It seems that I am not able to parameterize the tree to expand in narrow spaces.
Could you please tell me which are the parameters I have to change to be able to expand the tree in such strait spaces?
Is there any parameter for setting the speed of the vehicle?

Thank you in advance!!

@aakapatel
Copy link
Member

Hi @Taucrates,
You can start with modifying UFOMap parameters in server.launch inside launch folder. Max range could be around 10m. also you can try different resolutions in UFOMap in the same file.
For E-RRT tuning, increase number of goals, nodes and iterations. Planning depth param could be set to 1 instead of 2 for narrow environment. Finally for path selection you can influence the exploration by changing k_info to make it greedy.

I hope this helps,
Akash

@Taucrates
Copy link
Author

Thanks for the quick response @aakapatel!
Currently I am doing tests in the cave scenario available in https://github.com/LTU-RAI/gazebo_cave_world. I have changed the parameters you told me in a lot of combinations, but I am not able to find the correct configuration to allow the software to find frontiers through the first corridor, shown in the image below. If it is possible to clarify me more precisely the parameters to tune I would be very grateful.
Screenshot from 2024-07-12 12-27-15

@aakapatel
Copy link
Member

Hi,
I will try to make a different config parameters file for narrow env scenarios and test them to find a suitable set of params that can be generalized.
Hopefully next week I will push the additional config.
Thanks,
Akash

@Taucrates
Copy link
Author

Thank you so much Akash!!

@Taucrates
Copy link
Author

Hi, I will try to make a different config parameters file for narrow env scenarios and test them to find a suitable set of params that can be generalized. Hopefully next week I will push the additional config. Thanks, Akash

Hi Akash, is it possible for you to upload that new config file for the narrow environments?

Thank you in advance!!

@Taucrates Taucrates reopened this Nov 4, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants