opaal/bco-app-adhoc repository overview

⁠Adhoc

This BCO App can be used to learn individual automation rules from humans manual control behavior. The internally used logic inductive programming rule learner is called Aleph and is implemented in Prolog. Adhoc can be deployed to any bco node. As requirement, the app needs permission to read all your unit states in order to perform the training.

• Aleph 5 - A Learning Engine for Proposing Hypotheses - Inductive Logic Programming (ILP) system for Prolog
  • http://friguzzi.github.io/aleph/doc/manual.html#IDX21⁠
  • https://www.cs.ox.ac.uk/activities/programinduction/Aleph/aleph.html⁠
  • https://www.swi-prolog.org/pack/list?p=aleph⁠
• Prolog SWI
  • https://www.swi-prolog.org⁠

⁠Build docker image

docker build .

⁠Installation

⁠Via Docker

sudo docker run \
    --name bco-app-adhoc \
    --net=host \
    --volume /etc/localtime:/etc/localtime:ro \
    --volume /etc/timezone:/etc/timezone:ro \
    --volume bco_data:/home/bco/data \
    --detach \
    --env USER_ID=$(id -u bco) \
    --env GROUP_ID=$(getent group bco | cut -d: -f3) \
    --restart=always \
    -t \
    opaal/bco-app-adhoc:latest

⁠Via Sourcecode

1. Install Prolog SWI
   • https://www.swi-prolog.org/Download.html⁠
2. Install Aleph (Rule Learner Engine)
   • after Prolog is installed, aleph can be installed via the swi package manager by executing swipl -g 'pack_install('aleph'), halt.'
3. Install Adhoc (BCO Adaptive Home Control)
   • switch into the project folder and execute ./install.sh
4. Start Adhoc
   • After Adhoc is installed to ${prefix}/bin and this directory is registered in your $PATH variable (echo $PATH) you should be able to start Adhoc via bco-app-adhoc.