Suggestions to wiki multibeam sonar page CUDA installation instructions

[mabelzhang]

Per recent discussions, opening an issue with diff output to suggest edits to a wiki page.

I was able to get the CUDA library to work in the dockwater environment and was able to compile nps_uw_multibeam_sonar. There were some outdated commands though that should be updated.

• The dockwater flag for CUDA has changed to -c
• The pip commands now give a deprecation warning and say to use -m pip.

diff --git a/Multibeam-Forward-Looking-Sonar.md b/Multibeam-Forward-Looking-Sonar.md
index 0f0832b..9cc951d 100644
--- a/Multibeam-Forward-Looking-Sonar.md
+++ b/Multibeam-Forward-Looking-Sonar.md
@@ -99,11 +99,11 @@ The model is based on a ray-based spatial discretization of the model facets, be
 # Installation

 ## Option A. Use Docker
-The simplest way to prepare your machine with the CUDA library would be to use the Docker environment. Following commands include `-b cuda`, which provides the Cuda library.
+The simplest way to prepare your machine with the CUDA library would be to use the Docker environment. Following commands include `-c`, which provides the Cuda library.

Uninstall rocker

sudo apt-get remove python3-rocker -sudo pip uninstall rocker +sudo python3 -m pip uninstall rocker

Clone rocker fork by woensug choi

git clone git@github.com:woensug-choi/rocker.git @@ -113,10 +113,11 @@ cd rocker git checkout cuda-dev

install

-sudo pip install . +sudo python3 -m pip install .

change dockwater to cuda-dev branch

-cd ~/uuv_ws/src/dockwater +git clone git@github.com:Field-Robotics-Lab/dockwater.git +cd dockwater git checkout cuda-dev

Build and run docker image (This may take up to 30 min to finish at the first time)

[woensug-choi]

Thank you :) edited now.