As robots are becoming increasingly common within social settings, understanding the drivers of effective Human-Robot Interaction (HRI) becomes crucial. One factor emerging as a potential key variable in HRI is social presence, which is the extent to which a person feels connected to, or aware of, the presence of another. In the context of HRI, social presence has been identified as a factor that shapes human responses to robots, impacting outcomes such as attachment, trust, and social influence. Social presence itself appears to be influenced by factors including the nature of the embodiment. However, HRI research examining social presence is limited by the lack of consistent operationalization of social presence and conflation with the variance in embodiment of the robot. The current pilot study is designed to tease apart embodiment and social presence by exploring the relative impacts of interactions with two matched representations of a social robot (physical and virtual), finding that even controlling for size, proximity and response mechanisms, physical embodiment is associated with higher reported social presence. The study also identified both consistencies and inconsistencies in the measurement of social presence across assessments. The findings provide a foundation for further study into social presence in HRI, by clarifying mechanisms of quantifying and experimentally manipulating social presence, allowing insight into ways in which this factor drives HRI.

In the early days of the Open Source Robotics Foundation, a lesser-known project aimed to design an “app-able robot”, leading to the creation of the “Capabilities” package for the Robot Operating System (ROS). Over a decade later, formulating robot capabilities remains a significant technical hurdle in bringing robots from the lab into everyday life. This paper introduces Capabilities2, a successor to the original Capabilities package, now reimagined for ROS2. Capabilities2 enhances the original design by enabling advancements in skill-based control techniques and offering a more efficient, extensible framework for defining and utilising robot capabilities. We delve into its application in new real-world scenarios, with a particular focus on human-robot interactions and the deployment of collaborative mobile robots in human-centric environments. Capabilities2 addresses challenges in implementing intuitive, collaborative robots by introducing an abstracted database handler, an object-relational mapping for capability models, and a plugin architecture for capability execution. These features support dynamic capability representation, runtime adaptability, and integration with modern AI techniques for skill-based task planning. By providing a standardised yet flexible framework, Capabilities2 reduces the integration effort required to develop top-level controls for real-world scenarios, facilitating rapid development and deployment. Our contributions include the reimplementation of the Capabilities package in ROS2, enhancements to support contemporary robotic applications, and demonstrations of new use cases enabled by Capabilities2. We believe that Capabilities2 significantly advances the field of robotics by equipping developers with tools to create more capable, adaptable, and interactive robots. Capabilities2 is available at https://github.com/CollaborativeRoboticsLab/capabilities2