Affect-Driven Learning of Robot Behaviour for Collaborative Human-Robot Interactions

Aziz NA, Abd. Aziz A. Augmenting ECG Data with Multiple Filters for a Better Emotion Recognition System

A physiological-based emotion recognition system (ERS) with a unimodal approach such as an electrocardiogram (ECG) is not as popular compared to a multimodal approach. However, a single modality has the advantage of lower development and computational cost. Therefore, this study focuses on a unimodal ECG-based ERS. The ECG-based ERS has the potential to become the next mass-adopted consumer application due to the wide availability of wearable and mobile ECG devices in the market. Currently, ECG-inclusive affective datasets are limited, and many of the existing datasets have small sample sizes. Hence, ECG-based ERS studies are stunted by the lack of quality data. A novel multi-filtering augmentation technique is proposed here to increase the sample size of the ECG data. This technique augments the ECG signals by cleaning the data in different ways. Three small ECG datasets labelled according to emotion state are used in this study. The benefit of the proposed augmentation techniques is measured using the classification accuracy of five machine learning algorithms; k-nearest neighbours (KNN), support vector machine, decision tree, random forest and multilayer perceptron. The results show that with the proposed technique, there is a significant improvement in performance for all the datasets and classifiers. KNN classifier improved the most with the augmented data and the reported classification accuracies of over 90%.

Towards a Personality AI for Robots: Potential Colony Capacity of a Goal-Shaped Generative Personality Model When Used for Expressing Personalities via Non-Verbal Behaviour of Humanoid Robots

Engineering robot personalities is a challenge of multiple folds. Every robot that interacts with humans is an individual physical presence that may require their own personality. Thus, robot personalities engineers face a problem that is the reverse of that of personality psychologists: robot personalities engineers need to make batches of identical robots into individual personalities, as oppose to formulating comprehensive yet parsimonious descriptions of individual personalities that already exist. The robot personality research so far has been fruitful in demonstrating the positive effects of robot personality but unfruitful in insights into how robot personalities can be engineered in significant quantities. To engineer robot personalities for mass-produced robots we need a generative personality model with a structure to encode a robot’s individual characteristics as personality traits and generate behaviour with inter- and intra-individual differences that reflect those characteristics. We propose a generative personality model shaped by goals as part of a personality AI for robots towards which we have been working, and we conducted tests to investigate how many individual personalities the model can practically support when it is used for expressing personalities via non-verbal behaviour on the heads of humanoid robots.