The repository was built as additional material of the paper entitled "Robot's inner speech and emotions" by Arianna Pipitone, Sophia Corvaia and Antonio Chella .

The framework provides the robot with the ability to appraise the context by the inner speech (the self-dialogue implemented by the inner speech cognitive architecture [1]) and to elicit an emotion related to the appraised context. The emotion emerges by projecting the appraisal variables (that are mathematically formalized as described in the paper) into the Russell's space.

The repository contains the Python scripts implementing the proposed model of the tight link between inner speech and emotions, and the needed resources for running it (the knowledge model and the xml description of the parameters for the appraisal variables).

Please, refer to the paper for more details, and to the files in the module folder of this repository for comments about the code.

To run the provided framework, the NaoQi 2.5 have to be installed, and you can deploy the code on the robots by Aldebaran, specifically Pepper or NAO. Not other robotics platforms are tested.

[1] Pipitone A., Chella A. What robots want? Hearing the inner voice of a robot iScience, Volume 24, Issue 4, 2021, Article 102371

Operative systems:

• Ubuntu 16.04 LTS or latest

• Windows with Windows Subsystem for Linux (WSL) with Ubuntu 16.04 LTS or latest (you can follow the official instructions available at https://docs.microsoft.com/en-us/windows/wsl/install-manual for enabling WSL)

• Install additional libraries

The project require some additional libraries to work.

• Execnet: The project use a Python Bridge between the two based python interpreter (python 2.7 and every version of python 3), so you may install the execnet library: you can do it throught the command:

  pip install execnet

You can find more information on the following site: https://pypi.org/project/execnet/

• NAOqi: The project require the Naoqi library to manage robot of Aldebaran.

  pip install naoqi

You can follow the instruction available at http://doc.aldebaran.com/2-5/dev/python/install_guide.html. You have to check your NAOqi version on your robot before doing the installation.

• PyQt5: To execute the simulation graphic interface, you may install the PyQt5 library to manage the graphic part of the project. You can install it throught the command:

  pip install PyQt5

You can find more information about at https://pypi.org/project/PyQt5/ If you want to make changes in the simulation GUI, you have to install the Qt Designer Interface. You can find here at site https://www.qt.io/download

Pay attention to the Python Bridge: the project require python 2.7 to make robot alive, because the naoqi API run over that python version. So, you have to configure your application environment to python 2.7 and put there the model directory.

The graphic interface of the simulation run over python 3 (every version) because PyQt5 require that python version. So, you have to setting up your environment to that python version.

Once you have done it, you can make the communication between the two interpreter with the PythonBridge class. You can find it in:

/utilities/Utils.py

If you want to change the ontology in the project, you may change the file at:

/resources/kb/onto_tavola_eng.owl

Pay attention: the project requires an OWL ontology file to be execute. With the ontology, you have to change the pixels point of the dishes in the simulation table. You can find the file at:

/resources/points.txt

The project can evaluate all 28 emotion from the Russell's Circumplex Model, but this version of the model considers only the 5 primary emotion by Ekman's theory. If you want to evaluate other emotions, you have to view the CSV file of Russell's emotions available at:

/resources/Russell_Ekman_emotion.csv
/resources/Russell_emotion_label.csv

You can select the emotion you interest and run the project.

Launch the script project by the following command:

python3 simulation.py

and see your Pepper/NAO robot interacts to you while reasoning about its emotion by inner speech.

If you want to check a simple version of the project, considering only the modal modal cycle, launch the script project like this:

python modal_modal.py

Please, pay attention to the argument passed to the function.

Enjoy!