6.1 Installation Steps

The following video will guide you through the steps outlined below to install the features required to optimisation in RoboCup

You still need to follow the General Installation Instructions for which ever distribution you will be working with. These include Ubunutu 18.04, 20.04 or Windows 10 WSL
Download the Optomisation containers The Real-Time optimisation container runs the server in real-time, where as the sped up container does not throttle the server speed. You can still use the viualiser for both, however, the sped up version will be difficult to see intricate details of the behaviour. We suggest if you are debugging with the visualiser to use the real-time container. These containers are seperate from your code so it is not an issue to swop between them.

Real-Time Optimisation : https://courses.ms.wits.ac.za/~branden/RoboCup/container/rc_optim_cluster_slow_real_time.img
Sped-Up Optimisation : https://courses.ms.wits.ac.za/~branden/RoboCup/container/rc_optim_cluster_fast.img

Download the corresponding shell script for your installation If you have installed RoboCup with the graphics card option being true, in other words you have a robocup-nvidia shell script after the base installation then you must download the corresponding nvidia shell for the particular optimisation container. If you said you didnt have a graphics card then you would of the seen a robocup-mesa shell script and therefore you should download the corresponding meas shell for the particular optimisation container. These shell scripts are simply copies of the base script however, if you open them you can see they point to the new img file downloaded in step 2.

nvidia-realtime : https://courses.ms.wits.ac.za/~branden/RoboCup/shell/robocup-nvidia-slow
nvidia-fast : https://courses.ms.wits.ac.za/~branden/RoboCup/shell/robocup-nvidia-fast
mesa-realtime :  https://courses.ms.wits.ac.za/~branden/RoboCup/shell/robocup-mesa-slow
meas-fast :  https://courses.ms.wits.ac.za/~branden/RoboCup/shell/robocup-mesa-fast