Cognitive Robotics

Final project - Pepper list assistant

In this file you can see how to launch all things described in the file report.pdf.

Application Demo

Authors:

• Alessio Pepe - 0622701463 - [email protected]
• Alfonso Comentale - 0622701438 - [email protected]
• Giuseppina Di Paolo - 0622701510 - [email protected]
• Teresa Tortorella - 0622701507 - [email protected]

How to run the application

1. Find the michrophone index using the code

   import sounddevice as sd
   sd.query_devices()
   

2. Set the index just found in the config file (line 41) as:

   # pyAudio microphone idex. That need to be changed basing on the device. 
   MICROPHONE_INDEX = 2
   

3. Launch the chatbot server. This take about 60 seconds to start.

   # go in your ros workspace
   source devel/setup.bash
   roslaunch ros_chatbot chatbot.launch
   

4. Launch the pepper nodes.

   # go in your ros workspace
   source devel/setup.bash
   roslaunch pepper_nodes pepper.launch
   

5. Launch the core application (wait until the other two are ready).

   # go in your ros workspace
   source devel/setup.bash
   roslaunch fp_audio core.launch
   

Alternatively to the step 3, 4, and 5, you can launch as

# go in your ros workspace
source devel/setup.bash
roslaunch fp_audio all.launch

How to run the application (simulating pepper)

1. Set the index as default in the config file (line 41) as:

   # pyAudio microphone idex. That need to be changed basing on the device. 
   MICROPHONE_INDEX = None
   

   and, at line 45,

   # Flag: If False, the services who need Pepper are not called. Useful for debug.
   # -- In the use must be True.
   ON_PEPPER = False
   

2. Launch the chatbot server. This take about 60 seconds to start.

   # go in your ros workspace
   source devel/setup.bash
   roslaunch ros_chatbot chatbot.launch
   

3. Launch the core application that simulate the pepper webcam (wait until the other two are ready).

   # go in your ros workspace
   source devel/setup.bash
   roslaunch fp_audio core_s.launch
   

How to run the tests

To run the re-identification test, use the following commands:

# go in your ros workspace
source devel/setup.bash
roslaunch fp_audio test_reidentification.launch

To run the detector test, use the following commands:

# go in your ros workspace
source devel/setup.bash
roslaunch fp_audio test_detector.launch

To run th integration test ROS and RASA

# go in your ros workspace
source devel/setup.bash
roslaunch ros_chatbot dialogue.xml