Open dummyjohn77 opened 1 year ago
So ideally, we wanted to do an adaptable system to the type of env which is in because this influences the computational capabilities of the robot (aka on the open field alignment takes more time rather than in an indoor, cave-like space), but what we ended up doing is to do dependency on the number of points in the cloud as a space monitor. But first idea is a good step to enhance the work :)
That would be a great enhancement. Thank you for the prompt response.
So, at the moment, setting b_sub_to_lsm would ensure adaptive voxel filter (using pointcloud size ) to control the rate, right?
Hi, In the LOCUS 2.0 paper, there is a reference to "Space Monitor" in Fig 2. It was however not comprehensively referred in the paper. Is it related to "moving window". I notice that in the code, "lo_settings.yaml" has two separate settings, to enable the space monitor and the moving window.
Can you please shed some light on this and how it would help in improving the mapping?