Increasing Transparency of Reinforcement Learning using Shielding for Human Preferences and Explanations

The adoption of Reinforcement Learning (RL) in several human-centred applications provides robots with autonomous decision-making capabilities and adaptability based on the observations of the operating environment. In such scenarios, however, the learning process can make robots' behaviours unclear and unpredictable to humans, thus preventing a smooth and effective Human-Robot Interaction (HRI). As a consequence, it becomes crucial to avoid robots performing actions that are unclear to the user. In this work, we investigate whether including human preferences in RL (concerning the actions the robot performs during learning) improves the transparency of a robot's behaviours. For this purpose, a shielding mechanism is included in the RL algorithm to include human preferences and to monitor the learning agent's decisions. We carried out a within-subjects study involving 26 participants to evaluate the robot's transparency in terms of Legibility, Predictability, and Expectability in different settings. Results indicate that considering human preferences during learning improves Legibility with respect to providing only Explanations, and combining human preferences with explanations elucidating the rationale behind the robot's decisions further amplifies transparency. Results also confirm that an increase in transparency leads to an increase in the safety, comfort, and reliability of the robot. These findings show the importance of transparency during learning and suggest a paradigm for robotic applications with human in the loop.

AGAR: Attention Graph-RNN for Adaptative Motion Prediction of Point Clouds of Deformable Objects

This paper focuses on motion prediction for point cloud sequences in the challenging case of deformable 3D objects, such as human body motion. First, we investigate the challenges caused by deformable shapes and complex motions present in this type of representation, with the ultimate goal of understanding the technical limitations of state-of-the-art models. From this understanding, we propose an improved architecture for point cloud prediction of deformable 3D objects. Specifically, to handle deformable shapes, we propose a graph-based approach that learns and exploits the spatial structure of point clouds to extract more representative features. Then we propose a module able to combine the learned features in an adaptative manner according to the point cloud movements. The proposed adaptative module controls the composition of local and global motions for each point, enabling the network to model complex motions in deformable 3D objects more effectively. We tested the proposed method on the following datasets: MNIST moving digits, the Mixamo human bodies motions, JPEG and CWIPC-SXR real-world dynamic bodies. Simulation results demonstrate that our method outperforms the current baseline methods given its improved ability to model complex movements as well as preserve point cloud shape. Furthermore, we demonstrate the generalizability of the proposed framework for dynamic feature learning, by testing the framework for action recognition on the MSRAction3D dataset and achieving results on-par with state-of-the-art methods

Proceeding of the 1st Workshop on Social Robots Personalisation At the crossroads between engineering and humanities (CONCATENATE)

Nowadays, robots are expected to interact more physically, cognitively, and socially with people. They should adapt to unpredictable contexts alongside individuals with various behaviours. For this reason, personalisation is a valuable attribute for social robots as it allows them to act according to a specific user's needs and preferences and achieve natural and transparent robot behaviours for humans. If correctly implemented, personalisation could also be the key to the large-scale adoption of social robotics. However, achieving personalisation is arduous as it requires us to expand the boundaries of robotics by taking advantage of the expertise of various domains. Indeed, personalised robots need to analyse and model user interactions while considering their involvement in the adaptative process. It also requires us to address ethical and socio-cultural aspects of personalised HRI to achieve inclusive and diverse interaction and avoid deception and misplaced trust when interacting with the users. At the same time, policymakers need to ensure regulations in view of possible short-term and long-term adaptive HRI. This workshop aims to raise an interdisciplinary discussion on personalisation in robotics. It aims at bringing researchers from different fields together to propose guidelines for personalisation while addressing the following questions: how to define it - how to achieve it - and how it should be guided to fit legal and ethical requirements.

An Application of a Runtime Epistemic Probabilistic Event Calculus to Decision-making in e-Health Systems

We present and discuss a runtime architecture that integrates sensorial data and classifiers with a logic-based decision-making system in the context of an e-Health system for the rehabilitation of children with neuromotor disorders. In this application, children perform a rehabilitation task in the form of games. The main aim of the system is to derive a set of parameters the child's current level of cognitive and behavioral performance (e.g., engagement, attention, task accuracy) from the available sensors and classifiers (e.g., eye trackers, motion sensors, emotion recognition techniques) and take decisions accordingly. These decisions are typically aimed at improving the child's performance by triggering appropriate re-engagement stimuli when their attention is low, by changing the game or making it more difficult when the child is losing interest in the task as it is too easy. Alongside state-of-the-art techniques for emotion recognition and head pose estimation, we use a runtime variant of a probabilistic and epistemic logic programming dialect of the Event Calculus, known as the Epistemic Probabilistic Event Calculus. In particular, the probabilistic component of this symbolic framework allows for a natural interface with the machine learning techniques. We overview the architecture and its components, and show some of its characteristics through a discussion of a running example and experiments. Under consideration for publication in Theory and Practice of Logic Programming (TPLP).

The Road to a Successful HRI: AI, Trust and ethicS-TRAITS

The aim of this workshop is to foster the exchange of insights on past and ongoing research towards effective and long-lasting collaborations between humans and robots. This workshop will provide a forum for representatives from academia and industry communities to analyse the different aspects of HRI that impact on its success. We particularly focus on AI techniques required to implement autonomous and proactive interactions, on the factors that enhance, undermine, or recover humans' acceptance and trust in robots, and on the potential ethical and legal concerns related to the deployment of such robots in human-centred environments. Website: https://sites.google.com/view/traits-hri-2022

A ROS Architecture for Personalised HRI with a Bartender Social Robot

BRILLO (Bartending Robot for Interactive Long-Lasting Operations) project has the overall goal of creating an autonomous robotic bartender that can interact with customers while accomplishing its bartending tasks. In such a scenario, people's novelty effect connected to the use of an attractive technology is destined to wear off and, consequently, it negatively affects the success of the service robotics application. For this reason, providing personalised natural interaction while accessing its services is of paramount importance for increasing users' engagement and, consequently, their loyalty. In this paper, we present the developed three-layers ROS architecture integrating a perception layer managing the processing of different social signals, a decision-making layer for handling multi-party interactions, and an execution layer controlling the behaviour of a complex robot composed of arms and a face. Finally, user modelling through a beliefs layer allows for personalised interaction.

I Know What You Would Like to Drink: Benefits and Detriments of Sharing Personal Info with a Bartender Robot

This paper introduces benefits and detriments of a robot bartender that is capable of adapting the interaction with human users according to their preferences in drinks, music, and hobbies. We believe that a personalised experience during a human-robot interaction increases the human user's engagement with the robot and that such information will be used by the robot during the interaction. However, this implies that the users need to share several personal information with the robot. In this paper, we introduce the research topic and our approach to evaluate people's perceptions and consideration of their privacy with a robot. We present a within-subject study in which participants interacted twice with a robot that firstly had not any previous info about the users, and, then, having a knowledge of their preferences. We observed that less than 60\% of the participants were not concerned about sharing personal information with the robot.

Spatio-temporal Graph-RNN for Point Cloud Prediction

In this paper, we propose an end-to-end learning network to predict future frames in a point cloud sequence. As main novelty, an initial layer learns topological information of point clouds as geometric features, to form representative spatio-temporal neighborhoods. This module is followed by multiple Graph-RNN cells. Each cell learns points dynamics (i.e., RNN states) by processing each point jointly with the spatio-temporal neighbouring points. We tested the network performance with a MINST dataset of moving digits, a synthetic human bodies motions and JPEG dynamic bodies datasets. Simulation results demonstrate that our method outperforms baseline ones that neglect geometry features information.

Towards Transparency of TD-RL Robotic Systems with a Human Teacher

The high request for autonomous and flexible HRI implies the necessity of deploying Machine Learning (ML) mechanisms in the robot control. Indeed, the use of ML techniques, such as Reinforcement Learning (RL), makes the robot behaviour, during the learning process, not transparent to the observing user. In this work, we proposed an emotional model to improve the transparency in RL tasks for human-robot collaborative scenarios. The architecture we propose supports the RL algorithm with an emotional model able to both receive human feedback and exhibit emotional responses based on the learning process. The model is entirely based on the Temporal Difference (TD) error. The architecture was tested in an isolated laboratory with a simple setup. The results highlight that showing its internal state through an emotional response is enough to make a robot transparent to its human teacher. People also prefer to interact with a responsive robot because they are used to understand their intentions via emotions and social signals.

Emotional Distraction for Children Anxiety Reduction During Vaccination

Social assistive robots are starting to be widely used in pediatric health-care environments with the aim of distracting and entertaining children, and so of reducing a possible state of anxiety. In this paper, we present some initial results of a study (N=69) conducted in a Health-Vaccines Center, where the distraction role of a social robot, which interacts with a child showing an emotional behavior, is compared with the same not showing any emotional social cue. Outcome criteria for the evaluation of the intervention included the parents reported level of anxiety before, during and after the procedure.