Global frontiers issue

[wlsdn9350]

Hi! Thank you for sharing your interesting research.
I'm running your algorithm in a tunnel environment, and it appears, as seen in the picture below, that there are still global frontiers left or new ones being created, even though the area has already been explored. Is there any way to address this by modifying certain parameters?

[HongbiaoZ]

Hi, First, my answer is no. These frontiers are not observable. But the relocation stage only checks the closest one that can be observed but not all of them. These useless frontiers will not be cleaned until all areas are covered. You can check and delete them by yourself with a separate thread.

…

On Sun, 12 Feb 2023 at 13:33, wlsdn9350 ***@***.***> wrote: Hi! Thank you for sharing your interesting research. I'm running your algorithm in a tunnel environment, and it appears, as seen in the picture below, that there are still global frontiers left or new ones being created, even though the area has already been explored. Is there any way to address this by modifying certain parameters? [image: Screenshot] <https://user-images.githubusercontent.com/51685358/218294770-6ef1f890-7b26-459c-883f-f3574ab9ad16.png> — Reply to this email directly, view it on GitHub <#9>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AO2OAXXEF6QESSCAHLXPKKTWXBY4PANCNFSM6AAAAAAUZDMVYE> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

-- Hongbiao Zhu The Field Robotics Center Robotics Institution Carnegie Mellon University