(!) Please ask about problems and questions regarding this tutorial on answers.ros.org. Don't forget to include in your question the link to this page, the versions of your OS & ROS, and also add appropriate tags.

Using the Control script

Description: We use the control script to set a startposition/scene configuration.

Tutorial Level: BEGINNER

Description

We use the control script to set a startposition/scene configuration.

Setup

  • run rosrun asr_resources_for_active_scene_recognition start_modules_sim.sh.
  • make sure all modules are working, the tmux "control" window should show the python repl prompt ">>>".

  • make sure the state_machine's mode is set to 2/indirect search only.

Tutorial

  • CAUTION: make sure you saved parameters/launch files in asr_recognizer_prediction_ism, asr_world_model and asr_fake_object_recognition, since this might overwrite your settings. In the "control" window run setConfig(cfg_may25) to change the scene configuration, this will automatically restart the required tmux panes/modules.
  • (optional) with getCurrentConfig() you can get the currently set config, if a config of cfgs is set. the startPositios attribute is a list with some start poses to set.
  • ensure that the tmux "ptu_mock" window is working and the ptu client (cli_controll.py) asks for "Pan?", if you get ptu errors later, wait here a bit more.
  • run setDefaultPos() to move the robot to the first startposition of the current config.
  • (optional) run startRecognition() to start all recognizers and to find out which objects can be found in the current view. Something like "Smacks': 7, 'VitalisSchoko': 7}" should be returned. The result contains a number of recognitions per object type.

  • run startSM to start the state machine.
  • (optional) the waitSMFinished() function waits/blocks until the state machine terminates.
  • getSMResult() returns the result which can be seen in the "state_machine" tmux window, when the state_machine is finished.

Wiki: asr_resources_for_active_scene_recognition/UsingControl (last edited 2019-11-11 08:26:40 by FelixMarek)