buoncubi / sit_armor

The SIT algorithm injected in aRMOR

The SIT Algorithm Injected in the aRMOR service

A simple implementation of the Scene Identification and Tagging algorithm (SIT) injected in a ROS Multi Ontology References service (aRMOR).

How to run

• Run the service

rosrun sit_armor_injected execute it.emarolab.sitArmorInjected.SitArmorInjectedService

• or launch it from roslaunch

<node pkg="sit_armor" type="execute" name="sit_armor_injected" args="it.emarolab.sitArmorInjected.SitArmorInjectedService"/>

• load the ontology through aRMOR (where ros_ws is the ROS workspace in the home)

rosservice call /armor_interface_srv "armor_request:
  client_name: ''
  reference_name: 'ontoSIT'
  command: 'LOAD'
  primary_command_spec: 'FILE'
  secondary_command_spec: ''
  args: ['src/injected_armor_pkgs/scene_identification_tagging/sit/src/main/resources/simpleSIT.owl', 'http://www.emarolab.it/sit/simple', 'true', 'pellet', 'true']" 

• test some simple scenes

rosservice call /sit_armor_injected "ontoReference: 'ontoSIT'         
sceneElements:
- type: 'SPHERE'
  features: 
   - 0.0 
   - 0.0 
   - 0.0
- type: 'SPHERE'
  features: 
   - 0.8 
   - 0.0
   - 0.8
learningTreshold: 0.9" 

rosservice call /sit_armor_injected "ontoReference: 'ontoSIT'         
sceneElements:
- type: 'SPHERE'
  features: 
   - 0.0 
   - 0.0 
   - 0.0
- type: 'CONE'
  features: 
   - -0.8 
   - -0.8 
   - -0.0
   - 0
   - 0
   - 1
learningTreshold: 0.9" 

rosservice call /sit_armor_injected "ontoReference: 'ontoSIT'         
sceneElements:
- type: 'SPHERE'
  features: 
   - 0.0 
   - 0.0 
   - 0.0
- type: 'CONE'
  features: 
   - -0.8 
   - -0.8 
   - -0.0
   - 0
   - 0
   - 1
- type: 'SPHERE'
  features: 
   - 0.8 
   - 0.0 
   - 0.8
learningTreshold: 0.9" 

Contacts

For comment, discussions or support refer to this git repository open issues before to contact us

• [e-mail].