Parallelising LIO-mapping

[lsdta]

Hi @hyye ,

Have you attempted to parallelise your mapping algorithm at all? I see you have linked to nvidia-docker so I imagine you have run this on the jetson TX2 or similar? Is there any reason why a GPU board is useful, or have you found that SLAM is inherently mostly sequential? (and thus CPU-dominant).

It seems that SLAM can be improved with a GPU, but by the looks of your code the GPU is only utilised for the visualisation?

Any ideas appreciated, I am planning on running this on an embedded board but not sure if it is worth utilising a GPU-accelerated board or not.

[hyye]

Hi @lsdta, I think GPU can be utilized in some parts of the SLAM algorithms, like correspondence finding, residual calculation, etc., where operations are the same for each point. But meanwhile, some other data structures/algorithms may require changes.