Refs: | |
1. http://os.archlinuxarm.org/os/ArchLinuxARM-aarch64-latest.tar.gz | |
2. https://dl-cdn.alpinelinux.org/alpine/v3.13/releases/aarch64/alpine-virt-3.13.5-aarch64.iso | |
3. https://wiki.alpinelinux.org/wiki/Replacing_non-Alpine_Linux_with_Alpine_remotely | |
4. https://wiki.archlinux.org/index.php/installation_guide#Configure_the_system | |
5. https://archlinuxarm.org/platforms/armv8/generic | |
Requirement: | |
Console access. | |
Refs: | |
1. http://mirror.cs.pitt.edu/archlinux/iso/2021.02.01/archlinux-bootstrap-2021.02.01-x86_64.tar.gz | |
2. https://dl-cdn.alpinelinux.org/alpine/v3.13/releases/x86_64/alpine-virt-3.13.1-x86_64.iso | |
3. https://wiki.alpinelinux.org/wiki/Replacing_non-Alpine_Linux_with_Alpine_remotely | |
4. https://wiki.archlinux.org/index.php/installation_guide#Configure_the_system | |
Requirement: | |
Console access. | |
1. In Ubuntu |
- 8.1.1 contains the 'holdout' feature that prevents from over training. The default version from ubuntu repository,7.3, | |
doesn't have this feature. |
"Unscented kalman filter (UKF) is a special type of kalman filter which performs better than a "Simple Kalman Filter (SKF)" or an "Extended Kalman Filter (EKF)" for non-linear systems. The UKF uses a "Unscented Transform" to pick a few sample points around the mean to propagate in later iterations instead of linearly propagating the whole gaussian probability space. As a result the UKF reflects the mean and covariance more correctly.
The auto-start scripts on the bot are all 'shell-scripts' placed under /home/teamauv/startup_scripts/
.
The scripts are run by the 'X server' on user login. Therefore it's crucial that auto-login is enabled on the bot (otherwise even though the bot boots the scripts won't be run). To enable auto-login refer the follwing :
http://askubuntu.com/questions/51086/how-do-i-enable-auto-login-in-lightdm
- To add your own scripts you must place your 'shell-scripts' under
~/startup_scripts/
and append the following line to~/.gnomerc