Search

Continual learning (CL) has emerged as an important avenue of research in recent years, at the intersection of Machine Learning (ML) and Human–Robot Interaction (HRI), to allow robots to continually learn in their environments over long-term interactions with humans. Most research in CL, however, has been robot-centered to develop CL algorithms that can quickly learn new information on systematically collected static datasets. In this article, we take a human-centered approach to CL, to understand how humans interact with, teach, and perceive CL robots over the long term, and if there are variations in their teaching styles. We developed a socially guided CL system that integrates CL models for object recognition with a mobile manipulator robot and allows humans to directly teach and test the robot in real time over multiple sessions. We conducted an in-person study with 60 participants who interacted with the CL robot in 300 sessions with 5 sessions per participant. In this between-participant study, we used three different CL models deployed on a mobile manipulator robot. An extensive qualitative and quantitative analysis of the data collected in the study shows that there is significant variation among the teaching styles of individual users indicating the need for personalized adaptation to their distinct teaching styles. Our analysis shows that the constrained experimental setups that have been widely used to test most CL models are not adequate, as real users interact with and teach CL robots in a variety of ways. Finally, our analysis shows that although users have concerns about CL robots being deployed in our daily lives, they mention that with further improvements CL robots could assist older adults and people with disabilities in their homes. [ABSTRACT FROM AUTHOR]

This paper presents a biologically inspired flocking-based aggregation behaviour of a swarm of mobile robots. Aggregation behaviour is essential to many swarm systems, such as swarm robotics systems, in order to accomplish complex tasks that are impossible for a single agent. In this work, we developed a robot controller using Reynolds' flocking rules to coordinate the movements of multiple e-puck robots during the aggregation process. To improve aggregation behaviour among these robots and address the scalability issues in current flocking-based aggregation approaches, we proposed using a K-means algorithm to identify clusters of agents. Using the developed controller, we simulated the aggregation behaviour among the swarm of robots. Five experiments were conducted using Webots simulation software. The performance of the developed system was evaluated under a variety of environments and conditions, such as various obstacles, agent failure, different numbers of robots, and arena sizes. The results of the experiments demonstrated that the proposed algorithm is robust and scalable. Moreover, we compared our proposed algorithm with another implementation of the flocking-based self-organizing aggregation behaviour based on Reynolds' rules in a swarm of e-puck robots. Our algorithm outperformed this method in terms of cohesion performance and aggregation completion time. [ABSTRACT FROM AUTHOR]

Ever since Varela and Maturana proposed the concept of autopoiesis as the minimal requirement for life, there has been a focus on cellular systems that erect topological boundaries to separate themselves from their surrounding environment. Here, we reconsider whether the existence of such a spatial boundary is strictly necessary for self-producing entities. This work presents a novel computational model of a minimal autopoietic system inspired by dendrites and molecular dynamic simulations in three-dimensional space. A series of simulation experiments where the metabolic pathways of a particular autocatalytic set are successively inhibited until autocatalytic entities that could be considered autopoietic are produced. These entities maintain their distinctness in an environment containing multiple identical instances of the entities without the existence of a topological boundary. This gives rise to the concept of a metabolic boundary which manifests as emergent self-selection criteria for the processes of self-production without any need for unique identifiers. However, the adoption of such a boundary comes at a cost, as these autopoietic entities are less suited to their simulated environment than their autocatalytic counterparts. Finally, this work showcases a generalized metabolism-centered approach to the study of autopoiesis that can be applied to both physical and abstract systems alike. [ABSTRACT FROM AUTHOR]

To be successful, robots that can learn new tasks from humans should interact effectively with them while being trained, and humans should be able to trust the robots' abilities after teaching. Typically, when human learners make mistakes, their teachers tolerate those errors, especially when students exhibit acceptable progress overall. But how do errors and appearance of a trainee robot affect human teachers' trust while the robot is generally improving in performing a task? First, an online survey with 173 participants investigated perceived severity of robot errors in performing a cooking task. These findings were then used in an interactive online experiment with 138 participants, in which the participants were able to remotely teach their food preparation preferences to trainee robots with two different appearances. Compared with an untidy-looking robot, a tidy-looking robot was rated as more professional, without impacting participants' trust. Furthermore, while larger errors at the end of iterative training had a greater impact, even a small error could significantly reduce trust in a trainee robot performing the domestic physical task of food preparation, regardless of the robot's appearance. The present study extends human–robot interaction knowledge about teachers' perception of trainee robots, particularly when teachers observe them accomplishing domestic physical tasks. [ABSTRACT FROM AUTHOR]

How do we perceive robots practising a task that we have taught them? While learning, human trainees usually provide nonverbal cues that reveal their level of understanding and interest in the task. Similarly, nonverbal social cues of trainee robots that can be interpreted naturally by humans can enhance robot learning. In this article, we investigated a scenario in which a robot is practising a physical task in front of the human teachers (i.e., participants), who were asked to assume that they had previously taught the robot to perform that task. Through an online experiment with 167 participants, we examined the effects of different gaze patterns and arm movements with multiple speeds and various kinds of pauses on human teachers' perception of different attributes of the robot. We found that the perception of a trainee robot's attributes (e.g., confidence and eagerness to learn) can be systematically affected by its behaviours. Findings of this study can inform designing more successful nonverbal social interactions for intelligent robots. [ABSTRACT FROM AUTHOR]

Being able to measure time, whether directly or indirectly, is a significant advantage for an organism. It allows for timely reaction to regular or predicted events, reducing the pressure for fast processing of sensory input. Thus, clocks are ubiquitous in biology. In the present article, we consider minimal abstract pure clocks in different configurations and investigate their characteristic dynamics. We are especially interested in optimally time-resolving clocks. Among these, we find fundamentally diametral clock characteristics, such as oscillatory behavior for purely local time measurement or decay-based clocks measuring time periods on a scale global to the problem. We include also sets of independent clocks (clock bags), sequential cascades of clocks, and composite clocks with controlled dependence. Clock cascades show a condensation effect, and the composite clock shows various regimes of markedly different dynamics. [ABSTRACT FROM AUTHOR]

The flow semigroup, introduced by Rhodes, is an invariant for digraphs and a complete invariant for graphs. After collecting previous partial results together, we refine and prove Rhodes's conjecture on the structure of the maximal groups in the flow semigroup for finite, antisymmetric, strongly connected digraphs. Building on this result, we investigate and fully describe the structure and actions of the maximal subgroups of the flow semigroup acting on all but points for all finite digraphs and graphs for all . A linear algorithm (in the number of edges) is presented to determine these so-called 'defect groups' for any finite (di)graph. Finally, we prove that the complexity of the flow semigroup of a 2-vertex connected (and strongly connected di)graph with vertices is , completely confirming Rhodes's conjecture for such (di)graphs. [ABSTRACT FROM AUTHOR]

Co-development of action, conceptualization and social interaction mutually scaffold and support each other within a virtuous feedback cycle in the development of human language in children. Within this framework, the purpose of this article is to bring together diverse but complementary accounts of research methods that jointly contribute to our understanding of cognitive development and in particular, language acquisition in robots. Thus, we include research pertaining to developmental robotics, cognitive science, psychology, linguistics and neuroscience, as well as practical computer science and engineering. The different studies are not at this stage all connected into a cohesive whole; rather, they are presented to illuminate the need for multiple different approaches that complement each other in the pursuit of understanding cognitive development in robots. Extensive experiments involving the humanoid robot iCub are reported, while human learning relevant to developmental robotics has also contributed useful results. Disparate approaches are brought together via common underlying design principles. Without claiming to model human language acquisition directly, we are nonetheless inspired by analogous development in humans and consequently, our investigations include the parallel codevelopment of action, conceptualization and social interaction. Though these different approaches need to ultimately be integrated into a coherent, unified body of knowledge, progress is currently also being made by pursuing individual methods. [ABSTRACT FROM AUTHOR]

This article presents results from a multidisciplinary research project on the integration and transfer of language knowledge into robots as an empirical paradigm for the study of language development in both humans and humanoid robots. Within the framework of human linguistic and cognitive development, we focus on how three central types of learning interact and co-develop: individual learning about one's own embodiment and the environment, social learning (learning from others), and learning of linguistic capability. Our primary concern is how these capabilities can scaffold each other's development in a continuous feedback cycle as their interactions yield increasingly sophisticated competencies in the agent's capacity to interact with others and manipulate its world. Experimental results are summarized in relation to milestones in human linguistic and cognitive development and show that the mutual scaffolding of social learning, individual learning, and linguistic capabilities creates the context, conditions, and requisites for learning in each domain. Challenges and insights identified as a result of this research program are discussed with regard to possible and actual contributions to cognitive science and language ontogeny. In conclusion, directions for future work are suggested that continue to develop this approach toward an integrated framework for understanding these mutually scaffolding processes as a basis for language development in humans and robots. [ABSTRACT FROM AUTHOR]

In this paper, we use empowerment, a recently introduced biologically inspired measure, to allow an AI player to assign utility values to potential future states within a previously unencountered game without requiring explicit specification of goal states. We further introduce strategic affinity, a method of grouping action sequences together to form “strategies,” by examining the overlap in the sets of potential future states following each such action sequence. We also demonstrate an information-theoretic method of predicting future utility. Combining these methods, we extend empowerment to soft-horizon empowerment which enables the player to select a repertoire of action sequences that aim to maintain anticipated utility. We show how this method provides a proto-heuristic for nonterminal states prior to specifying concrete game goals, and propose it as a principled candidate model for “intuitive” strategy selection, in line with other recent work on “self-motivated agent behavior.” We demonstrate that the technique, despite being generically defined independently of scenario, performs quite well in relatively disparate scenarios, such as a Sokoban-inspired box-pushing scenario and in a Pac-Man-inspired predator game, suggesting novel and principle-based candidate routes toward more general game-playing algorithms. [ABSTRACT FROM PUBLISHER]

The article offers information on the Biological and Mathematical Basis of Interaction Computing (BIOMICS) project. It informs that the aim of the project is to develop new model of bottom-up computation with equal applicability to Interaction Computing (IC) which includes computer science and system biology. It further informs that the BIOMICS project framework has been developed from statistical physics, dynamical systems and logical point of view.