Frontiers in Robotics and AI

The evolving field of human-robot interaction (HRI) necessitates that we better understand how social robots operate and interact with humans. This scoping review provides an overview of about 300 research works focusing on the use of the NAO robot from 2010 to 2020. This study presents one of the most extensive and inclusive pieces of evidence on the deployment of the humanoid NAO robot and its global reach. Unlike most reviews, we provide both qualitative and quantitative results regarding how NAO is being used and what has been achieved so far. We analyzed a wide range of theoretical, empirical, and technical contributions that provide multidimensional insights, such as general trends in terms of application, the robot capabilities, its input and output modalities of communication, and the human-robot interaction experiments that featured NAO (e.g. number and roles of participants, design, and the length of interaction). Lastly, we derive from the review some research gaps in current state-of-the-art and provide suggestions for the design of the next generation of social robots.

Social robots are increasingly being used as a mediator between a therapist and a child in autism therapy studies. In this context, most behavioural interventions are typically short-term in nature. This paper describes a long-term study that was conducted with 11 children diagnosed with either Autism Spectrum Disorder (ASD) or ASD in co-occurrence with Attention Deficit Hyperactivity Disorder (ADHD). It uses a quantitative analysis based on behavioural measures, including engagement, valence, and eye gaze duration. Each child interacted with a robot on several occasions in which each therapy session was customized to a child’s reaction to robot behaviours. This paper presents a set of robot behaviours that were implemented with the goal to offer a variety of activities to be suitable for diverse forms of autism. Therefore, each child experienced an individualized robot-assisted therapy that was tailored according to the therapist’s knowledge and judgement. The statistical analyses showed that the proposed therapy managed to sustain children’s engagement. In addition, sessions containing familiar activities kept children more engaged compared to those sessions containing unfamiliar activities. The results of the interviews with parents and therapists are discussed in terms of therapy recommendations. The paper concludes with some reflections on the current study as well as suggestions for future studies.

Natural motion types found in skeletal and muscular systems of vertebrate animals inspire researchers to transfer this ability into engineered motion, which is highly desired in robotic systems. Dielectric elastomer actuators (DEAs) have shown promising capabilities as artificial muscles for driving such structures, as they are soft, lightweight, and can generate large strokes. For maximum performance, dielectric elastomer membranes need to be sufficiently pre-stretched. This fact is challenging, because it is difficult to integrate pre-stretched membranes into entirely soft systems, since the stored strain energy can significantly deform soft elements. Here, we present a soft robotic structure, possessing a bioinspired skeleton integrated into a soft body element, driven by an antagonistic pair of DEA artificial muscles, that enable the robot bending. In its equilibrium state, the setup maintains optimum isotropic pre-stretch. The robot itself has a length of 60 mm and is based on a flexible silicone body, possessing embedded transverse 3D printed struts. These rigid bone-like elements lead to an anisotropic bending stiffness, which only allows bending in one plane while maintaining the DEA's necessary pre-stretch in the other planes. The bones, therefore, define the degrees of freedom and stabilize the system. The DEAs are manufactured by aerosol deposition of a carbon-silicone-composite ink onto a stretchable membrane that is heat cured. Afterwards, the actuators are bonded to the top and bottom of the silicone body. The robotic structure shows large and defined bimorph bending curvature and operates in static as well as dynamic motion. Our experiments describe the influence of membrane pre-stretch and varied stiffness of the silicone body on the static and dynamic bending displacement, resonance frequencies and blocking forces. We also present an analytical model based on the Classical Laminate Theory for the identification of the main influencing parameters. Due to the simple design and processing, our new concept of a bioinspired DEA based robotic structure, with skeletal and muscular reinforcement, offers a wide range of robotic application.

Control theory provides engineers with a multitude of tools to design controllers that manipulate the closed-loop behavior and stability of dynamical systems. These methods rely heavily on insights into the mathematical model governing the physical system. However, in complex systems, such as autonomous underwater vehicles performing the dual objective of path following and collision avoidance, decision making becomes nontrivial. We propose a solution using state-of-the-art Deep Reinforcement Learning (DRL) techniques to develop autonomous agents capable of achieving this hybrid objective without having a priori knowledge about the goal or the environment. Our results demonstrate the viability of DRL in path following and avoiding collisions towards achieving human-level decision making in autonomous vehicle systems within extreme obstacle configurations.