Gerardo Puga

This is Nav2 using the beam model: [beam_nav2.webm](https://github.com/user-attachments/assets/aa888b39-c9d7-4e9a-8609-6c83e52e8f9f)

I did some data gathering using data collected using the Robomaster sim, in three situations. Baseline (full map):  Robot starting in an "unknown" patch in the map:  Robot...

Beam model experiments: [beam_model_results.tar.gz](https://github.com/user-attachments/files/16753369/beam_model_results.tar.gz) Likelihood model experiments: [likelihood_model_results.tar.gz](https://github.com/user-attachments/files/16753371/likelihood_model_results.tar.gz)

I think we can remove the "bug" label from this, as everything seems to be working as expected in its design. It does open the question of whether the OpenLoris...

I wonder if, instead of implementing it, we should evaluate whether this actually makes a difference in performance. Is there empirical or bibliographic support for this feature? Does people use...

> but so is summing cubes and it works Indeed, but notice that we know the sum of cubes works because it's the default and therefore we can infer people...

> FTR I unearthed [ros-planning/navigation#238](https://github.com/ros-planning/navigation/pull/238). Interesting. @mikeferguson by any chance do you remember any context about the beam skipping feature in https://github.com/ros-planning/navigation/pull/238 ?

@DPR00 the base skeleton benchmark is already up and working https://github.com/Ekumen-OS/lambkin/pull/116 , feel free to take over. Notes: - At first sight, there's no difference betweeen using beam skipping or...

> As you mentioned, in one of the runs, LikelihoodProb tends to diverge near the end, but also LikelihoodProb+BeamSkipping when using the omni mode. Check that that are no "unknown"...

while the 24hs dataset is still running, the a broader set of datasets continues to confirm the trend that: - `likelihood_field_prob` performs consistently better than `likelihood_field` across the board, though...