A better understanding of how humans perceive robot personality variables could enable the design of more socially acceptable robots. In this exploratory study, we examined whether manipulations of an iCub robot’s voice and movements affected human participants’ perceptions of the robot’s personality. We programmed the robot to behave in different ways during a teaching scenario in which it played either a teaching, learning, or collaborative role, shown in recorded videos of human–robot interactions. A total of 240 participants in an Amazon Mechanical Turk study watched these videos and completed a series of questionnaires assessing their perceptions of the robot. Participants perceived the iCub as more extroverted when it spoke faster, with a higher pitch, and performed larger-amplitude movements. It was determined that participants’ personality dimensions were more influential in their perceptions of the robot’s TIPI and RoSAS personality dimensions than the robot’s social role and personality manipulations. Participants’ self-rated extroversion, emotional stability, and conscientiousness repeatedly appeared as significant factors affecting their perceptions of the robot’s personality. Interestingly, we observed strong perceiver effects, whereby participants’ perceptions of the robot’s personality traits were correlated with their own self-rated personality traits.

To be able to learn effectively, robots sometimes will need to select more suitable human teachers. We propose an attribute in human teachers for robots that learn through visual observations, namely human teachers’ awareness of and attention to the robot’s visual capabilities and constraints, and explore how it affects robot learning outcomes. In an in-person experiment involving 72 participants who taught three physical tasks to an iCub humanoid robot, we manipulated teachers’ awareness of the robot’s visual constraints by offering the visual perspective of the robot in one of the experimental conditions. Participants who were able to see the robot’s vision output paid increased attention to ensuring task objects were visible to the robot when providing demonstrations of physical tasks. This emphasis on attention to the robot’s view resulted in better learning outcomes for the robot, as indicated by lower perception error rates and higher learning scores. This study contributes to understanding factors in human teachers that lead to better learning outcomes for robots.

Social robots are increasingly employed as personalized coaches in educational settings, offering new opportunities for applications such as public speaking training. In this domain, emotional self-regulation plays a crucial role, especially for students presenting in a non-native language. This study proposes a novel pipeline for detecting public speaking anxiety (PSA) using multimodal emotion recognition. Unlike traditional datasets that typically rely on acted emotions, we consider spontaneous data from students interacting naturally with a social robot coach. Emotional labels are generated through knowledge distillation, enabling the creation of soft labels that reflect the emotional valence of each presentation. We introduce a lightweight multimodal model that integrates speech prosody and body posture to classify speakers by anxiety level, without relying on linguistic content. Evaluated on a collected dataset of student presentations, the system achieves 74.67% accuracy and an F1-score of 0.64. The model can operate completely disconnected from the transmission network on an NVIDIA Jetson board, safeguarding data privacy and demonstrating its feasibility for real-world deployment.

The evolution of living beings with continuous and consistent progress toward adaptation and ways to model evolution along principles as close as possible to Darwin's are important areas of focus in Artificial Life. Though genetic algorithms and evolutionary strategies are good methods for modeling selection, crossover, and mutation, biological systems are undeniably spatially distributed processes in which living organisms interact with locally available individuals rather than with the entire population at once. This work presents a model for the survival of organisms during a change in the environment to a less favorable one, putting them at risk of extinction, such as many organisms experience today under climate change or local habitat loss or fragmentation. Local spatial structure of resources and environmental quality also impacts the capacity of an evolving population to adapt. The problem is considered on a probabilistic cellular automaton with update rules based on the principles of genetic algorithms. To carry out simulations according to the described model, the Darwinian cellular automata are introduced, and the software has been designed with the code available open source. An experimental evaluation of the behavioral characteristics of the model was carried out, completed by a critical evaluation of the results obtained, parametrically describing conditions and thresholds under which extinction or survival of the population may occur.

August 16 2024 Editorial: Special Issue "The Distributed Ghost"—Cellular Automata, Distributed Dynamical Systems, and Their Applications to Intelligence In Special Collection: CogNet Stefano Nichele, Stefano Nichele Østfold University College, NorwayOslo Metropolitan University, Norway Search for other works by this author on: This Site Google Scholar Hiroki Sayama, Hiroki Sayama Binghamton University, USAWaseda University, Japan Search for other works by this author on: This Site Google Scholar Eric Medvet, Eric Medvet University of Trieste, Italy Search for other works by this author on: This Site Google Scholar Chrystopher Nehaniv, Chrystopher Nehaniv University of Waterloo, Canada Search for other works by this author on: This Site Google Scholar Mario Pavone Mario Pavone University of Catania, Italy Search for other works by this author on: This Site Google Scholar Author and Article Information Stefano Nichele Østfold University College, NorwayOslo Metropolitan University, Norway Hiroki Sayama Binghamton University, USAWaseda University, Japan Eric Medvet University of Trieste, Italy Chrystopher Nehaniv University of Waterloo, Canada Mario Pavone University of Catania, Italy Online ISSN: 1530-9185 Print ISSN: 1064-5462 © 2024 Massachusetts Institute of Technology2024Massachusetts Institute of Technology Artificial Life 1–3. https://doi.org/10.1162/artl_e_00450 Cite Icon Cite Permissions Share Icon Share Facebook Twitter LinkedIn Email Views Icon Views Article contents Figures & tables Video Audio Supplementary Data Peer Review Search Site Citation Stefano Nichele, Hiroki Sayama, Eric Medvet, Chrystopher Nehaniv, Mario Pavone; Editorial: Special Issue "The Distributed Ghost"—Cellular Automata, Distributed Dynamical Systems, and Their Applications to Intelligence. Artif Life 2024; doi: https://doi.org/10.1162/artl_e_00450 Download citation file: Ris (Zotero) Reference Manager EasyBib Bookends Mendeley Papers EndNote RefWorks BibTex toolbar search Search Dropdown Menu toolbar search search input Search input auto suggest filter your search All ContentAll JournalsArtificial Life Search Advanced Search This content is only available as a PDF. © 2024 Massachusetts Institute of Technology2024Massachusetts Institute of Technology Article PDF first page preview Close Modal You do not currently have access to this content.

The year 2024 marks the 25th anniversary of the publication of evoloops, an evolutionary variant of Chris Langton's self-reproducing loops, which proved constructively that Darwinian evolution of self-reproducing organisms by variation and natural selection is possible within deterministic cellular automata. Over the last few decades, this line of Artificial Life research has since undergone several important developments. Although it experienced a relative dormancy of activity for a while, the recent rise of interest in open-ended evolution and the success of continuous cellular automata models have brought researchers' attention back to how to make spatiotemporal patterns self-reproduce and evolve within spatially distributed computational media. This article provides a review of the relevant literature on this topic over the past 25 years and highlights the major accomplishments made so far, the challenges being faced, and promising future research directions.

As robots increasingly permeate diverse domains like healthcare, education, service industries and homes, accurately understanding humans’ responses to and behaviour towards robots is crucial. While many human-robot interaction (HRI) studies focus on either quantitative or qualitative approaches, we advocate a mixed-method approach. This study investigated robot norm violations by implementing a scenario where a mobile manipulator robot and a human, in-person, carry out a physical, competitive task. Sixty-two participants were recruited and randomly assigned to either an experimental or a control condition (balanced for age/gender). The scenario was a competitive scavenger hunt game where participants took turns with a robot. We investigated the robot behaviours’ effects on trust, discomfort, competence, enjoyment, participant behaviour and physiological changes. The mixed-method approach integrated physiological measurements, behavioural observations and qualitative responses, thus offering a comprehensive account of HRI dynamics in the context of norm violations. Questionnaire results reveal significant shifts in human perceptions and attitudes when social norms are violated by robots, compared to a norm-compliant control condition. Specifically, trust and enjoyment decrease, discomfort increases and the robot’s perceived competence is compromised. These findings are extended through additional analyses of participants’ physiological changes, behaviours and responses to open-ended questions. Behavioural observations indicated increased verbal engagement and emotional responses, while physiological data showed elevated stress levels in the experimental group. Our study highlights the advantage of a mixed-methods approach combining different qualitative and quantitative data, providing a more comprehensive picture of participants’ perceptions of a robot, and how they react and respond to robot norm violations.

As robots integrate more into daily life, socially aware robots with specific social attributes and behaviors are necessary. This review aims to explore how social norms in Human–Robot Interaction (HRI) impact robot design and human perception. We searched for relevant articles in the following databases: ACM Digital Library, IEEE Digital Library, Scopus, Springer Link, and PsycINFO. After applying inclusion and exclusion criteria, a final set of 69 articles were included in the review. These articles were categorized based on whether they examined norm conformity or norm violations, and were further sorted into 12 categorical norm labels to assist in analysis and comparison. By examining the existing literature, this review uncovers how social norms impact aspects of HRIs like trust, acceptance, and comfort while highlighting the importance of aligning robot design with user expectations. It reveals design challenges such as accounting for cultural variations, context-specific norms, and evolving norms over time. Addressing these challenges has the potential to improve user experiences, promote broader acceptance of robots, and foster successful integration of robots into various domains. The findings contribute to the ongoing discussion on the role of social norms in HRI, offering valuable insights and a foundation for future research.

For robots to perform assistive tasks in unstructured home environments, they must learn and reason on the semantic knowledge of the environments. Despite a resurgence in the development of semantic reasoning architectures, these methods assume that all the training data is available a priori. However, each user’s environment is unique and can continue to change over time, which makes these methods unsuitable for personalized home service robots. Although research in continual learning develops methods that can learn and adapt over time, most of these methods are tested in the narrow context of object classification on static image datasets. In this paper, we combine ideas from continual learning, semantic reasoning, and interactive machine learning literature and develop a novel interactive continual learning architecture for continual learning of semantic knowledge in a home environment through human-robot interaction. The architecture builds on core cognitive principles of learning and memory for efficient and real-time learning of new knowledge from humans. We integrate our architecture with a physical mobile manipulator robot and perform extensive system evaluations in a laboratory environment over two months. Our results demonstrate the effectiveness of our architecture to allow a physical robot to continually adapt to the changes in the environment from limited data provided by the users (experimenters), and use the learned knowledge to perform object fetching tasks.

For social robots to succeed in places such as homes, they must learn new skills from various people and act in a manner desirable to different users. We introduce a novel biologically inspired approach for robot learning through program-level imitation, inspired by the way primates, including humans, understand and perform complex actions. Our approach enables robots to discover the hierarchical structure of tasks by identifying sequential regularities and sub-goals from diverse human demonstrations. To do so, human-provided demonstrations, which can be obtained by a robot through different modalities (such as kinesthetic teaching, behavioural observation, and verbal instruction), are processed by an algorithm that discovers multiple possibilities for arranging observed sub-goals to achieve a final goal. Prior to acting, the available sequences are evaluated based on user-defined criteria, through mental simulation of the task by the robot, to find the optimal sequence of actions. As a proof-of-concept, we implemented our system on an iCub humanoid robot and present here how our method allowed the robot to adapt its action sequences for task execution when starting the task from different states, incorporating user preference for finishing the task as fast as possible. Our envisaged system is meant to accommodate variations in human teaching styles and is expected to help a robot perform tasks with greater flexibility and efficiency. This work contributes by proposing a framework for robots to learn from humans at an abstract level, opening the way to more adaptable and intelligent robotic assistants in everyday tasks.

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.

The successful integration of domestic service robots into home environments can bring significant services and convenience to the general population and possibly mitigate important societal issues, such as care provision for older adults. However, home environments are complex, dynamic and object-rich. It is, thus, very probable that service robots will encounter ambiguity while interacting with household items. To enable service robots to be more adaptive, we proposed a learning so-cial referencing computational framework and experimentally evaluated the framework on a mobile manipulator robot, Fetch, in object selection scenarios. The framework allows the robot to (1) detect and analyze the ambiguity level based on the robot's view and user's command, (2) assess the human's attention level and attract their attention, (3) disambiguate references to objects using human feedback and (4) learn novel objects after clarification from the user. System evaluation results are presented. The framework is modular and can be applied to different robotic platforms.

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 continual learning, however, has been robot-centered to develop continual learning algorithms that can quickly learn new information on static datasets. In this paper, we take a human-centered approach to continual learning, to understand how humans teach continual learning robots over the long term and if there are variations in their teaching styles. We conducted an in-person study with 40 participants that interacted with a continual learning robot in 200 sessions. In this between-participant study, we used two different CL models deployed on a Fetch 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. The results also show that although there is a difference in the teaching styles between expert and non-expert users, the style does not have an effect on the performance of the continual learning robot. Finally, our analysis shows that the constrained experimental setups that have been widely used to test most continual learning techniques are not adequate, as real users interact with and teach continual learning robots in a variety of ways. Our code is available at https://github. com/aliayub7/c1-hri.

For robots to assist users with household tasks, they must first learn about the tasks from the users. Further, performing the same task every day, in the same way, can become boring for the robot’s user(s), therefore, assistive robots must find creative ways to perform tasks in the household. In this paper, we present a cognitive architecture for a household assistive robot that can learn personalized breakfast options from its users and then use the learned knowledge to set up a table for breakfast. The architecture can also use the learned knowledge to create new breakfast options over a longer period of time. The proposed cognitive architecture combines state-of-the-art perceptual learning algorithms, computational implementation of cognitive models of memory encoding and learning, a task planner for picking and placing objects in the household, a graphical user interface (GUI) to interact with the user and a novel approach for creating new breakfast options using the learned knowledge. The architecture is integrated with the Fetch mobile manipulator robot and validated, as a proof-of-concept system evaluation in a large indoor environment with multiple kitchen objects. Experimental results demonstrate the effectiveness of our architecture to learn personalized breakfast options from the user and generate new breakfast options never learned by the robot

Social robots have the potential to serve as coaches for public speaking training. To design successful social robots, it is important to understand the expectations and perceptions of prospective users of such robots. In this paper, we present thematic analyses of comments made by 168 participants in an online study where participants watched videos of agents in the role of a public speaking coach. The study had a between-participant design with three conditions: two conditions with a humanoid social robot in either (1) active listening mode, i.e., using non-verbal backchanneling, or (2) passive listening mode, and (3) a voice assistant agent. The themes identified and discussed can contribute to the development of social robots and other agents as public speaking coaches.

This study aimed to investigate the effects of norm-violating behaviour on human perception and attitudes towards robots. Specifically, we examined the impact of a robot performing social norm violations in the context of a competitive scavenger hunt game. During the game, the robot was programmed to engage in predefined behaviours considered as social norm violations, including both injunctive and descriptive norm violations (e.g., cheating, and making loud noises). The study used an experimental and control group, with participants either exposed to norm-violating behaviour or not, respectively. The results indicated that participants in the experimental group had a strong awareness of the norm-violating behaviour according to self-reported assessments. Additionally, post-questionnaire results revealed a significant difference in trust, overall enjoyment, and discomfort between the two groups. These findings show that in our study, participants expected robots to abide by both types of social norms (i.e., injunctive and descriptive) and that violations of them negatively impacted participants’ perceptions and attitudes towards robots. This further emphasizes the importance of considering social norms in the design and programming of robots for human-robot interactions.

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.

One of the fundamental modes of learning in children is through curiosity. Children (and adults) interact with new people, learn about novel objects, activities and other stimuli through curiosity and other intrinsic motivations. Creating autonomous robots that learn continually through intrinsic curiosity may result in breakthroughs in artificial intelligence. Such robots could continue to learn about themselves and the world around them through curiosity, thus improving their abilities over their ‘lifetime’. Although recent works on curiosity in different fields have produced significant results, most of these works have focused on constrained simulated environments which do not involve human interaction. However, in real-world applications such as healthcare, home-assistance etc., robots generally have to interact with humans on a regular basis. In these scenarios, it is imperative that curiosity is directed towards seeking out and learning important information from the humans when needed rather than simply learning in an unsupervised manner. Further, there is limited work on how humans perceive such curious robots and whether humans prefer curious robots that adapt over time to other robots that simply perform their assigned tasks. In this workshop, our goal is to bring together researchers and practitioners in different multidisciplinary fields to discuss the role of robot curiosity in real-world applications and its implications in human-robot interaction (HRI).

When humans make a mistake, they often try to employ some strategies to manage the situation and possibly mitigate the negative effects of the mistake. Robots that operate in the real world will also make errors and therefore might benefit from such recovery strategies. In this work, we studied how different verbal expression strategies of a trainee humanoid robot when committing an error after learning a task influence participants’ intention to use it. We performed a virtual experiment in which the expression modes of the robot were as follows: (1) being silent; (2) verbal expression but ignoring any errors; or (3) verbal expression while mentioning any error by apologizing, as well as acknowledging and justifying the error. To simulate teaching, participants remotely demonstrated their preferences to the robot in a series of food preparation tasks; however, at the very end of the teaching session, the robot made an error (in two of the three experimental conditions). Based on data collected from 176 participants, we observed that, compared to the mode where the robot remained silent, both modes where the robot utilized verbal expression could significantly enhance participants’ intention to use the robot in the future if it made an error in the last practice round. When no error occurred at the end of the practice rounds, a silent robot was preferred and increased participants’ intention to use.

When human students practise new skills with a teacher, they often display nonverbal behaviours (e.g., head and limb movements, gaze, etc.) to communicate their level of understanding and expressing their interest in the task. Similarly, a student robot's capability to provide human teachers with social signals to express its internal state might improve learning outcomes. This could also lead to a more successful social interactions between intelligent robots and human teachers. However, to design successful nonverbal communication for a robot, we first need to understand how human teachers interpret such nonverbal cues when watching a trainee robot practising a task. Therefore, in this paper, we study the effects of different gaze behaviours as well as manipulating speed and smoothness of arm movement on human teachers' perception of a robot's (a) confidence, (b) eagerness to learn, and (c) attention to the task. In an online experiment, we asked the 167 participants (as teachers) to rate the behaviours of a trainee robot in the context of learning a physical task. The results suggest that splitting the robot's gaze between the teacher and the task not only affects the perceived attention, but can also make the robot appear to be more eager to learn. Furthermore, perceptions of all three attributes tested were systematically affected by varying parameters of the robot's arm movement trajectory while performing task actions.

It is anticipated that intelligent robots will gain the ability to learn from humans how to perform tasks, and will assist them in many contexts such as with household chores in the near future; therefore, people should have the confidence to trust these robots after teaching them how to do a task. Like most machines, robots may sometimes behave in an erroneous manner and such errors can easily undermine trust in the robots, depending on their severity. Nevertheless, when a robot has been taught a task by humans, we hypothesize that the teachers may ignore small mistakes made by the robot, if it shows significant improvements while practising the task. We first conducted a study with 173 participants in which the perceived severity of different robot errors in a household chore (preparing food) was investigated. We then used the results to create scenarios of different levels of severity and conducted a second study with 138 participants to investigate the impact of error severity on trust. Participants remotely taught their preferences in food preparation tasks to robots. Over several practice rounds, robots’ behaviour improved, but the robots made either (a) no errors, (b) a small, or (c) a big error at the end, depending on the experimental condition. Small errors significantly affected trust and big errors had an even more adverse impact. Trust in the robot was found to be correlated with personality traits of the participants as well as with their disposition to trust other people.

Motor resonance, the activation of an observer’s motor control system by another actor’s movements, has been claimed to be an indicator for quality of interaction. Motor interference as one of the consequences of the presence of resonance can be detected by analyzing an actor’s spatial movements. It has therefore been used as an indicator for the presence of motor resonance. Unfortunately, the experimental paradigm in which motor interference has been shown to be detectable is ecologically implausible both in terms of the types of movements employed and the number of repetitions required. In the presented experiment, we tested whether some of these experimental constraints can be relaxed or modified toward a more naturalistic behavior without losing the ability to detect the interference effect. In the literature, spatial variance has been analytically quantified in many different ways. This study found these analytical variations to be nonequivalent by implementing them. Back-and-forth transitive movements were tested for motor interference; the effect was found to be more robust than with left-right movements, although the direction of interference was opposite to that reported in the literature. We conclude that motor interference, when measured by spatial variation, lacks promise for embedding in naturalistic interaction scenarios because the effect sizes were small.

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.

“No” is one of the first ten words used by children and embodies the first form of linguistic negation. Despite its early occurrence, the details of its acquisition remain largely unknown. The circumstance that “no” cannot be construed as a label for perceptible objects or events puts it outside the scope of most modern accounts of language acquisition. Moreover, most symbol grounding architectures will struggle to ground the word due to its non-referential character. The presented work extends symbol grounding to encompass affect and motivation. In a study involving the child-like robot iCub, we attempt to illuminate the acquisition process of negation words. The robot is deployed in speech-wise unconstrained interaction with participants acting as its language teachers. The results corroborate the hypothesis that affect or volition plays a pivotal role in the acquisition process. Negation words are prosodically salient within prohibitive utterances and negative intent interpretations such that they can be easily isolated from the teacher’s speech signal. These words subsequently may be grounded in negative affective states. However, observations of the nature of prohibition and the temporal relationships between its linguistic and extra-linguistic components raise questions over the suitability of Hebbian-type algorithms for certain types of language grounding.

The question of what to imitate is pivotal for imitation learning in robotics. When the robot's tutor is a naive user, it is very difficult for the embodied agent to account for the unpredictability of the tutor's behaviour. Preliminary results from a previous study suggested that the phenomenon of temporal emphasis, i.e., that tutors tend to keep the goal state of the demonstrated task stationary longer than the sub-states, can be used to recognise that task. In the present paper, the previous study is expanded and the existence of the phenomenon is investigated further. An improved experimental setup, using the iCub humanoid robot and naive users, was implemented. Analysis of the data showed that the phenomenon was detected in the majority of the cases, with a strongly significant result. In the few cases that the end state was not the one with the longest time span, it was a borderline second. Then, a very simple algorithm using a single binary criterion was used to show that the phenomenon exists and can be detected easily. That leads to the argument that humans may also be able to detect this phenomenon and use it for recognizing, as learners or emphasizing and teaching as tutors, the end goal, at least for tasks with clear and separate sub-goal sequences. A robot that implements this behavior could be able to perform better both as a tutor and as a learner when interacting with naive users.

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.

We overview how sensorimotor experience can be operationalized for interaction scenarios in which humanoid robots acquire skills and linguistic behaviours via enacting a “form-of-life” in interaction games (following Wittgenstein) with humans. The enactive paradigm is introduced which provides a powerful framework for the construction of complex adaptive systems, based on interaction, habit, and experience. Enactive cognitive architectures (following insights of Varela, Thompson and Rosch) that we have developed support social learning and robot ontogeny by harnessing information-theoretic methods and raw uninterpreted sensorimotor experience to scaffold the acquisition of behaviours. The success criterion here is validation by the robot engaging in ongoing human-robot interaction with naive participants who, over the course of iterated interactions, shape the robot's behavioural and linguistic development. Engagement in such interaction exhibiting aspects of purposeful, habitual recurring structure evidences the developed capability of the humanoid to enact language and interaction games as a successful participant.

Music in general and drumming in specific has been used several times in human-robot interaction studies. We present a drumming robotic system based on the iCub humanoid robotic platform using all four limbs to play a full-sized drum set and the exploratory experimental setups used to assess it. The social aspect of the system uses the concept of music as a communicative gesture to transfer information and facilitate the interaction. The first implementation is focused on assessing the imitative capabilities of a robot on a full drum set at different levels of speed and complexity. By treating the rhythmic content as communication, the second implementation allows for improvisation of both the robot and the human in a kind of musical dialogue. While keeping the structure of turn-taking fixed and predefined, the focus of the setup is on the actual content of each turn and how it affects the other participant's communicative exchange. The resulting platform was able to imitate the drumming of the humans within the timing constraints of the rhythm and tempo. The second setup was used to evaluate the improvisation capabilities of the platform and also led to observations about the potential behaviour of the human participant in such an interaction. The robotic drumming system produced can be used for more complex rhythmical interactions between humans and robots. These will facilitate the study of not only the structure of the interaction but also its content.

Mutual gaze is an important aspect of face-to-face communication that arises from the interaction of the gaze behavior of two individuals. In this dual eye-tracking study, gaze data was collected from human conversational pairs with the goal of gaining insight into what characteristics of the conversation partners influence this behavior. We investigate the link between personality, familiarity and mutual gaze. The results found indicate that mutual gaze behavior depends on the characteristics of both partners rather than on either individual considered in isolation. We discuss the implications of these findings for the design of socially appropriate gaze controllers for robots that interact with people.

The paper investigates the effects of a robot's “on-line” feedback during a tutoring situation with a human tutor. Analysis is based on a study conducted with an iCub robot that autonomously generates its feedback (gaze, pointing gesture) based on the system's perception of the tutor's actions using the idea of reciprocity of actions. Sequential micro-analysis of two opposite cases reveals how the robot's behavior (responsive vs. non-responsive) pro-actively shapes the tutor's conduct and thus co-produces the way in which it is being tutored. A dialogic and a monologic tutoring style are distinguished. The first 20 seconds of an encounter are found to shape the user's perception and expectations of the system's competences and lead to a relatively stable tutoring style even if the robot's reactivity and appropriateness of feedback changes.

In this paper we describe the design of an approach to evolve Sims's creatures with morphology and behaviour similar to biped animals. Our hypothesis is that biases in morphology that encourage limb specialisation, combined with rewards for successful locomotion and carrying at the same time and realistic, physics-based penalties for falling in fitness function, would lead to creatures capable of bipedal locomotion. We present experimental results demonstrating successful evolution of biped morphology and stepping gait.

Shannon's information theory can be used to quantify morphological and topological features of a collective of agents in an arbitrary environment. In particular the ability of individual agents to extract information locally about global features of the collective can be quantified. Here, we considered chains of agents in a grid world. The agents are equipped with local sensors. We then quantified the amount of information the sensors contain about certain features global to the chain. Furthermore, we compared the amount of locally available information to the amount of information the whole collective could in principle acquire about a feature in different contexts.

This paper describes a system that simulates word form acquisition without meaning as a preliminary stage in language acquisition, through interaction with a human teacher. The system extracts phonemes from the teacher's speech and supplies them to a linguistically enabled synthetic agent. The agent babbles initially but gradually, words begin to emerge as the agent biases its babble towards the speech of the teacher. Experiments are conducted in real-time with a human teacher interacting with the agent embodied in the iCub humanoid robot.

We demonstrate how autonomous agents can be evolved whose controllers are programs in the dynamical systems-oriented, behaviour-based PDL language introduced by Luc Steels. Controllers evolved using a physics simulator solved the problem of parallel parking a car without colliding into obstacles on three sides of a parking area.

This article involves the study of inheritable mutations in a von Neumann self-reproducing automaton, making use of the GOLLY cellular automata simulator. Multi-point mutations were done on the tape which holds the description of the self-reproducer so as to observe the effects over subsequent generations. Von Neumann said that `Self reproduction includes the ability to undergo inheritable mutations as well as the ability to make another organism like the original.' One of the central models used to study self-reproduction is cellular automata. Evolution relies on sources of variability, such as the results of mutation. This occurs when a system reproduces itself with inheritable variation, resulting in possibly more complex offspring. Subsequent offspring generations may inherit complex attributes from the parent generation during reproduction, and this is shown here constructively.

This article investigates the role of interaction kinesics in human–robot interaction (HRI). We adopted a bottom-up, synthetic approach towards interactive competencies in robots using simple, minimal computational models underlying the robot's interaction dynamics. We present two empirical, exploratory studies investigating a drumming experience with a humanoid robot (KASPAR) and a human. In the first experiment, the turn-taking behaviour of the humanoid is deterministic and the non-verbal gestures of the robot accompany its drumming to assess the impact of non-verbal gestures on the interaction. The second experiment studies a computational framework that facilitates emergent turn-taking dynamics, whereby the particular dynamics of turn-taking emerge from the social interaction between the human and the humanoid. The results from the HRI experiments are presented and analysed qualitatively (in terms of the participants’ subjective experiences) and quantitatively (concerning the drumming performance of the human–robot pair). The results point out a trade-off between the subjective evaluation of the drumming experience from the perspective of the participants and the objective evaluation of the drumming performance. A certain number of gestures was preferred as a motivational factor in the interaction. The participants preferred the models underlying the robot's turn-taking which enable the robot and human to interact more and provide turn-taking closer to ‘natural’ human–human conversations, despite differences in objective measures of drumming behaviour. The results are consistent with the temporal behaviour matching hypothesis previously proposed in the literature which concerns the effect that the participants adapt their own interaction dynamics to the robot's.

We created a simple evolutionary system, F-sexyloop, on a deterministic twelve-state five-neighbour cellular automaton (CA) where self-reproducing loops have the capability of sex. This work was based on the sexyloop which was transformed by adding two new states and new rules. In the F-sexyloop, the loops can carry a sex gene used to facilitate the transfer of genetic material from a loop to another. This gene is analogous to the F factor plasmid in bacterial conjugation which confers the capacity to act as a donor of genetic material (including the gene itself). Therefore, the sex gene could potentially be maintained in the population during evolution or disappear. We show that in a wide variety of cases, the sex gene persists over evolutionary time and is present in the genomes of the dominant species.

It is thought that meaning may be grounded in early childhood language learning via the physical and social interaction of the infant with those around him or her, and that the capacity to use words, phrases and their meaning are acquired through shared referential dasiainferencepsila in pragmatic interactions. In order to create appropriate conditions for language learning by a humanoid robot, it would therefore be necessary to expose the robot to similar physical and social contexts. However in the early stages of language learning it is estimated that a 2-year-old child can be exposed to as many as 7,000 utterances per day in varied contextual situations. In this paper we report on the issues behind and the design of our currently ongoing and forthcoming experiments aimed to allow a robot to carry out language learning in a manner analogous to that in early child development and which effectively dasiashort cutspsila holophrase learning. Two approaches are used: (1) simulated babbling through mechanisms which will yield basic word or holophrase structures and (2) a scenario for interaction between a human and the humanoid robot where shared dasiaintentionalpsila referencing and the associations between physical, visual and speech modalities can be experienced by the robot. The output of these experiments, combined to yield word or holophrase structures grounded in the robot's own actions and modalities, would provide scaffolding for further proto-grammatical usage-based learning. This requires interaction with the physical and social environment involving human feedback to bootstrap developing linguistic competencies. These structures would then form the basis for further studies on language acquisition, including the emergence of negation and more complex grammar.

Understanding the evolvability of simple differentiating multicellular systems is a fundamental problem in the biology of genetic regulatory networks and in computational applications inspired by the metaphor of growing and developing networks of cells. We compare the evolvability of a static network model to a more realistic regulatory model with dynamic structure. In the former model, each regulatory protein-binding site is always influenced by exactly one gene product. In the latter model, binding is only more likely to occur the better the match between site and gene product is (smooth binding) and, in addition, affinity dynamically changes under the action of specificity factors during a cellpsilas lifetime. On evolutionary timescales, this means that often the strength of influences between nodes is perturbed instead of direct changes being made to network connectivity. A main result is that for evolutionary search spaces of increasing sizes evolved performance drops much more strongly in the classical network model as compared to the smooth binding model. This effect was even greater in the case of using smooth binding together with specificity factors.

In this paper, we explore some conceptual issues, relevant for the design of robotic systems aimed at interacting with humans in domestic environments. More specifically, we study the role of the robot's feedback (positive or negative acknowledgment of understanding) on a human teacher's demonstration of a routine home task (laying a table). Both the human and the system's perspectives are considered in the analysis and discussion of results from a human-robot user study, highlighting some important conceptual and practical issues. These include the lack of explicitness and consistency on people's demonstration strategies. Furthermore, we discuss the need to investigate design strategies to elicit people's knowledge about the task and also successfully advertize the robot's abilities in order to promote people's ability to provide appropriate demonstrations.

This paper presents results of a novel study on interaction kinesics where 18 children interacted with a humanoid child-sized robot called KASPAR. Based on findings in psychology and social sciences we propose the temporal behaviour matching hypothesis which predicts that children will adapt to and match the robot's temporal behaviour. Each child took part in six experimental trials involving two games in which the dynamics of interactions played a key part: a body expression imitation game, where the robot imitated expressions demonstrated by the children, and a drumming game where the robot mirrored the children's drumming. In both games KASPAR responded either with or without a delay. Additionally, in the drumming game, KASPAR responded with or without exhibiting facial/gestural expressions. Individual case studies as well as statistical analysis of the complete sample are presented. Results show that a delay of the robot's drumming response lead to larger pauses (with and without robot nonverbal gestural expressions) and longer drumming durations (with nonverbal gestural expressions only). In the imitation game, the robot's delay lead to longer imitation eliciting behaviour with longer pauses for the children, but systematic individual differences are observed in regards to the effects on the children's pauses. Results are generally consistent with the temporal behaviour matching hypothesis, i.e. children adapted the timing of their behaviour, e.g. by mirroring to the robot's temporal behaviour.

We present results from an empirical study investigating emergent turn-taking in a drumming experience involving Kaspar, a humanoid child-sized robot, and adult participants. In this work, our aim is to have turn-taking and role switching which is not deterministic but emerging from the social interaction between the human and the humanoid. Therefore the robot is not just dasiafollowingpsila and imitating the human, but could be the leader in the game and being imitated by the human. Data from the first implementation of a human-robot interaction experiment are presented and analysed qualitatively (in terms of participantspsila subjective experiences) and quantitatively (concerning the drumming performance of the human-robot pair). Results are analysed statistically and show significant differences for the three games (with different probabilistic models) where the models enabling more interaction and more ldquonaturalrdquo turn-taking were preferred by the human participants.

Beyond complexity measures, sometimes it is worthwhile in addition to investigate how complexity changes structurally, especially in artificial systems where we have complete knowledge about the evolutionary process. Hierarchical decomposition is a useful way of assessing structural complexity changes of organisms modeled as automata, and we show how recently developed computational tools can be used for this purpose, by computing holonomy decompositions and holonomy complexity. To gain insight into the evolution of complexity, we investigate the smoothness of the landscape structure of complexity under minimal transitions. As a proof of concept, we illustrate how the hierarchical complexity analysis reveals symmetries and irreversible structure in biological networks by applying the methods to the lac operon mechanism in the genetic regulatory network of Escherichia coli.

We study the evolvability and dynamics of artificial genetic regulatory networks (GRNs), as active control systems, realizing simple models of biological clocks that have evolved to respond to periodic environmental stimuli of various kinds with appropriate periodic behaviors. GRN models may differ in the evolvability of expressive regulatory dynamics. A new class of artificial GRNs with an evolvable number of complex cis-regulatory control sites--each involving a finite number of inhibitory and activatory binding factors--is introduced, allowing realization of complex regulatory logic. Previous work on biological clocks in nature has noted the capacity of clocks to oscillate in the absence of environmental stimuli, putting forth several candidate explanations for their observed behavior, related to anticipation of environmental conditions, compartmentation of activities in time, and robustness to perturbations of various kinds or to unselected accidents of neutral selection. Several of these hypotheses are explored by evolving GRNs with and without (Gaussian) noise and blackout periods for environmental stimulation. Robustness to certain types of perturbation appears to account for some, but not all, dynamical properties of the evolved networks. Unselected abilities, also observed for biological clocks, include the capacity to adapt to change in wavelength of environmental stimulus and to clock resetting.

At the heart of the development of fertilized eggs into fully formed organisms and the adaptation of cells to changed conditions are genetic regulatory networks (GRNs). In higher multicellular organisms, signal selection and multiplexing are performed at the cis-regulatory domains of genes, where combinations of transcription factors (TFs) regulate the rates at which the genes are transcribed into mRNA. To be able to act as activators or repressors of gene transcription, TFs must first bind to target sequences on the regulatory domains. Two TFs that act in concert may bind entirely independently of each other, but more often binding of the first one will alter the affinity of the other for its binding site. This article presents a systematic investigation into the effect of TF binding dependences on the predicted regulatory function of this bio-logic. Four extreme scenarios, commonly used to classify enzyme activation and inhibition patterns, for the binding of two TFs were explored: independent (the TFs bind without affecting each other's affinities), competitive (the TFs compete for the same binding site), ordered (the TFs bind in a compulsory order), and joint binding (the TFs either bind as a preformed complex, or binding of one is virtually impossible in the absence of the other). The conclusions are: (1) the laws of combinatorial logic hold only for systems with independently binding TFs; (2) systems formed according to the other scenarios can mimic the functions of their Boolean logical counterparts, but cannot be combined or decomposed in the same way; and (3) the continuously scaled output of systems consisting of competitively binding activators and repressors can be controlled more robustly than that of single TF or (quasi-)logical multi-TF systems.

In this paper, we discuss the concept of autobiographic agent and how memory may extend an agent's temporal horizon and increase its adaptability. These concepts are applied to an implementation of a scenario where agents are interacting in a complex virtual artificial life environment. We present computational memory architectures for autobiographic virtual agents that enable agents to retrieve meaningful information from their dynamic memories which increases their adaptation and survival in the environment. The design of the memory architectures, the agents, and the virtual environment are described in detail. Next, a series of experimental studies and their results are presented which show the adaptive advantage of autobiographic memory, i.e. from remembering significant experiences. Also, in a multi-agent scenario where agents can communicate via stories based on their autobiographic memory, it is found that new adaptive behaviours can emerge from an individual's reinterpretation of experiences received from other agents whereby higher communication frequency yields better group performance. An interface is described that visualises the memory contents of an agent. From an observer perspective, the agents’ behaviours can be understood as individually structured, and temporally grounded, and, with the communication of experience, can be seen to rely on emergent mixed narrative reconstructions combining the experiences of several agents. This research leads to insights into how bottom-up story-telling and autobiographic reconstruction in autonomous, adaptive agents allow temporally grounded behaviour to emerge. The article concludes with a discussion of possible implications of this research direction for future autobiographic, narrative agents.

For robots to be more capable interaction partners they will necessarily need to adapt to the needs and requirements of their human companions. One way that the human could aid this adaptation may be by teaching the robot new ways of doing things by physically demonstrating different behaviours and tasks such that the robot learns new skills by imitating the learnt behaviours in appropriate contexts. In human–human teaching, the concept of scaffolding describes the process whereby the teacher guides the pupil to new competence levels by exploiting and extending existing competencies. In addition, the idea of event structuring can be used to describe how the teacher highlights important moments in an overall interaction episode. Scaffolding and event structuring robot skills in this way may be an attractive route in achieving robot adaptation; however, there are many ways in which a particular behaviour might be scaffolded or structured and the interaction process itself may have an effect on the robot's resulting performance. Our overall research goal is to understand how to design an appropriate human–robot interaction paradigm where the robot will be able to intervene and elicit knowledge from the human teacher in order to better understand the taught behaviour. In this article we examine some of these issues in two exploratory human–robot teaching scenarios. The first considers task structuring from the robot's viewpoint by varying the way in which a robot is taught. The experimental results illustrate that the way in which teaching is carried out, and primarily how the teaching steps are decomposed, has a critical effect on the efficiency of human teaching and the effectiveness of robot learning. The second experiment studies the problem from the human's viewpoint in an attempt to study the human teacher's spontaneous levels of event segmentation when analysing their own demonstrations of a routine home task to a robot. The results suggest the existence of some individual differences regarding the level of granularity spontaneously considered for the task segmentation and for those moments in the interaction which are viewed as most important.

We explore how information theoretic quantities such as potential information flow (empowerment) can be used as a drive toward complex collective behaviour in the context of multi-agent systems. In a first experiment, we investigate the empowerment of two agents interacting in a grid world. We show that some conditions lead to higher empowerment than others, depending on the amount of interaction and the amount of information shared by the agents. We then investigate more deeply the tradeoff between freedom of the agents and the constraints they impose on each other. We show that there exist a trade-off between these where empowerment is maximized. In a third experiment, we show that agents behaving so as to maximize potential information transfer over time generate a wide range of complex collective behaviours. We then discuss how these notions can be compared to what happens in natural systems

Networks that abstractly model natural genetic regulatory networks (GRNs) are evolved to show a range of dynamical behaviors. Specifically one group was evolved to show differentiation, i.e. to be able to perform an additional behavior as compared to the original, single target behavior. These GRNs are then analyzed and compared with measures used in the biological sciences. Having huge numbers of GRNs available for not only analysis but also "metabolic" inspection, we find that evolutionary niches (target functions) do not necessarily mold network structure uniquely. Our results suggest that variability operators can have a stronger influence on network topologies than selection pressures, especially when many topologies can create similar dynamics. Furthermore, damaging the most significantly represented motif (whether in differentiating or non-differentiating GRNs) is found not to have a significantly bigger impact on function than random lesions, suggesting that particular motifs are not as important in the robust functioning of networks as might perhaps be expected

We created a simple evolutionary system, sexyloop, on a deterministic ten-state five-neighbour cellular automaton (CA) where self-reproducing loops have the capability of sex. With this ability, the loops are capable of transferring genetic material into other loops. This work was based on Sayama's evoloop which was transformed by adding a new state and new rules. The evoloop model showed an emergent evolutionary process only due to an adaptation of the loops to interaction in the environment; and after a certain time, all the individuals capable of self-reproduction belonged to the smallest species 4 which reproduced the fastest. We created two different models of self-reproducing loops with sex, sexyloop M1 and M2, in order to study the possibility of sex in self-reproducing automata and to assess the impact of sex on the evolutionary process via comparing between the evoloop and the sexyloop variants. In the sexyloop M1 and M2, the diversity of the whole population was different from that found in the evoloop and the evolutionary process was quite different too. The sexyloops also created smaller and bigger species than the evoloops, and exhibited greater diversity and faster evolution than their non-sexual counterparts. The most interesting model was the sexyloop M2 whose evolutionary dynamics had a very different longterm behaviour than the evoloop and sexyloop M1. The most surprising and intriguing phenomenon in the sexyloop M2 was that the evolutionary process was selecting quickly a bigger species than in the evoloop and sexyloop M1: the species 5. In fact, the individuals from this species needed more time to reproduce than those from the species 4. So it appears that in the sexyloop M2, the fittest individual was not one that could reproduce the fastest but surely one that reproduced fast and which was more adapted to propagate in an environment where sex with individuals of the same and other types could occur. These results give the first examples in cellular automata of evolution in a population of self-replicators where sex plays an important role

Teaching a robot new skills may require that the teacher scaffolds the teaching experience appropriately. However, due to inherent assumptions made by a human teacher the scaffolding process may in some circumstances fail to effectively teach the robot. Here we illustrate this issue in two simple robot teaching exploratory studies and examine the assumptions made by the teacher when teaching the robot. In the first study the human teacher had to reason about robot perceived states in order to provide suitable teaching. In the second study the human teachers had to understand the perceptual constraints of the robot based on the instructions given beforehand by the experimenter and subsequently adapt the guidance given. The results suggest that although the two tasks are quite distinct in their level of complexity a common thread can be observed: people tend to underspecify their teaching. It seems that steps of the explanation are assumed to be known and skipped or not even considered at all. We reflect on the possibility that one of the major challenges in designing robots that are capable interaction partners in these teaching situations is to be able to make them communicate their internal state and current capabilities effectively. Furthermore, we also reflect on designing appropriate behavioral primitives for the robot, corresponding implications on the level of task description and for benefiting from human teaching.

Sensor evolution in nature aims at improving the acquisition of information from the environment and is intimately related with selection pressure toward adaptivity and robustness. Our work in the area indicates that information theory can be applied to the perception-action loop. This letter studies the perception-action loop of agents, which is modeled as a causal Bayesian network. Finite state automata are evolved as agent controllers in a simple virtual world to maximize information flow through the perception-action loop. The information flow maximization organizes the agent's behavior as well as its information processing. To gain more insight into the results, the evolved implicit representations of space and time are analyzed in an information-theoretic manner, which paves the way toward a principled and general understanding of the mechanisms guiding the evolution of sensors in nature and provides insights into the design of mechanisms for artificial sensor evolution.

This paper addresses the problem of body mapping in robotic imitation where the demonstrator and imitator may not share the same embodiment [degrees of freedom (DOFs), body morphology, constraints, affordances, and so on]. Body mappings are formalized using a unified (linear) approach via correspondence matrices, which allow one to capture partial, mirror symmetric, one-to-one, one-to-many, many-to-one, and many-to-many associations between various DOFs across dissimilar embodiments. We show how metrics for matching state and action aspects of behavior can be mathematically determined by such correspondence mappings, which may serve to guide a robotic imitator. The approach is illustrated and validated in a number of simulated 3-D robotic examples, using agents described by simple kinematic models and different types of correspondence mappings.

Imitation is a powerful learning tool when humans and robots interact in a social context. A series of experimental runs and a small pilot user study were conducted to evaluate the performance of a system designed for robot imitation. Performance assessments of similarity of imitative behaviours were carried out by machines and by humans: the system was evaluated quantitatively (from a machine-centric perspective) and qualitatively (from a human perspective) in order to study the reconciliation of these views. The experimental results presented here illustrate how the number of exceptions can be used as a performance measure by a robotic or software imitator of an object manipulation behaviour. (In this context, exceptions are events when the optimal displacement and/or rotation that minimize the dissimilarity metrics used to generate a corresponding imitative behaviour cannot be directly achieved in the particular context.) Results of the user study giving similarity judgments on imitative behaviours were used to examine how the quantitative measure of the number of exceptions (from a robot's perspective) corresponds to the qualitative evaluation of similarity (from a human's perspective) for the imitative behaviours generated by the jabberwocky system. Results suggest that there is a good alignment between this quantitive system centered assessment and the more qualitative human-centered assessment of imitative performance.

As robots enter everyday life and start to interact with ordinary people [5]the question of their appearance becomes increasingly important. A user's perception of a robot can be strongly influenced by its facial appearance [6]. The dimensions and issues of face design are illustrated in the design rationale, details of construction and intended uses of a new minimal expressive robot called KASPAR.

This paper presents the combined results of two studies that investigated how a robot should best approach and place itself relative to a seated human subject. Two live Human Robot Interaction (HRI) trials were performed involving a robot fetching an object that the human had requested, using different approach directions. Results of the trials indicated that most subjects disliked a frontal approach, except for a small minority of females, and most subjects preferred to be approached from either the left or right side, with a small overall preference for a right approach by the robot. Handedness and occupation were not related to these preferences. We discuss the results of the user studies in the context of developing a path planning system for a mobile robot.

Programming robots to carry out useful tasks is both a complex and non-trivial exercise. A simple and intuitive method to allow humans to train and shape robot behaviour is clearly a key goal in making this task easier. This paper describes an approach to this problem based on studies of social animals where two teaching strategies are applied to allow a human teacher to train a robot by moulding its actions within a carefully scaffolded environment. Within these enviroments sets of competences can be built by building stateslash action memory maps of the robot's interaction within that environment. These memory maps are then polled using a k-nearest neighbour based algorithm to provide a generalised competence. We take a novel approach in building the memory models by allowing the human teacher to construct them in a hierarchical manner. This mechanism allows a human trainer to build and extend an action-selection mechanism into which new skills can be added to the robot's repertoire of existing competencies. These techniques are implemented on physical Khepera miniature robots and validated on a variety of tasks.

This paper addresses the problem of body mapping in robotic imitation where the demonstrator and imitator may not share the same embodiment (degrees of freedom (DOFs), body morphology, constraints, affordances and so on). Body mappings are formalized using a unified (linear) approach via correspondence matrices, which allow one to capture partial, mirror symmetric, one-to-one, one-to-many, many-to-one and many-to-many associations between various DOFs across dissimilar embodiments. We show how metrics for matching state and action aspects of behaviour can be mathematically determined by such correspondence mappings, which may serve to guide a robotic imitator. The approach is illustrated in a number of examples, using agents described by simple kinematic models and different types of correspondence mappings. Also, focusing on aspects of displacement and orientation of manipulated objects, a selection of metrics are presented, towards a characterization of the space of effect metrics

As robots enter everyday life and start to interact with ordinary people the question of their appearance becomes increasingly important. Our perception of a robot can be strongly influenced by its facial appearance. Synthesizing relevant ideas from narrative art design, the psychology of face recognition, and recent HRI studies into robot faces, we discuss effects of the uncanny valley and the use of iconicity and its relationship to the self-other perceptive divide, as well as abstractness and realism, classifying existing designs along these dimensions. A new expressive HRI research robot called KASPAR is introduced and the results of a preliminary study on human perceptions of robot expressions are discussed

This paper presents an approach for human activity recognition focusing on gestures in a teaching scenario, together with the setup and results of user studies on human gestures exhibited in unconstrained human-robot interaction (HRI). The user studies analyze several aspects: the distribution of gestures, relations, and characteristics of these gestures, and the acceptability of different gesture types in a human-robot teaching scenario. The results are then evaluated with regard to the activity recognition approach. The main effort is to bridge the gap between human activity recognition methods on the one hand and naturally occuring or at least acceptable gestures for HRI on the other. The goal is two-fold: to provide recognition methods with information and requirements on the characteristics and features of human activities in HRI, and to identify human preferences and requirements for the recognition of gestures in human-robot teaching scenarios

This paper describes our general framework for the investigation of how human gestures can be used to facilitate the interaction and communication between humans and robots. More specifically, a study was carried out to reveal which "naturally occurring" gestures can be observed in a scenario where users had to explain to a robot how to perform a specific home task. The study followed a within-subjects design where ten participants had to demonstrate how to lay a table for two people using two different methods for their explanation: utilizing only gestures or gestures and speech. The experiments also served to validate a new coding scheme for human gestures in human-robot interaction, with good inter-rater reliability. Moreover, annotated video corpus was produced and characteristics such as frequency, duration, and co-occurrence of the different gestural classes have been gathered in order to capture requirements for the designers of HRI systems. The results regarding the frequencies of the different gestural types suggest an interaction between the order of presentation of the two methods and the actual type of gestures produced. Moreover, the results also suggest that there might be an interaction between the type of task and the type of gestures produced

An evolutionary predecessor to observational imitation may have been self-imitation. Self-imitation is where an agent is able to learn and replicate actions it has experienced through the manipulation of its body by another. This form of imitative learning has the advantage of avoiding some of the complexities encountered in observational learning such as the correspondence problem. We investigate how a system using self-imitation can be constructed with reference to psychological models of motor control including ideomotor theory and ideas from social scaffolding seen in animals to allow us to construct a robotic control system. The system allows a human trainer to teach a robot new skills and modify existing skills. Additionally the system allows the robot to notify the trainer when it is being taught skills it already possesses. We argue that this mechanism may be the first step towards the transformation from self-imitation to observational imitation. We demonstrate the system on a physical Pioneer robot with a 5-DOF arm and pan/tilt camera which is taught using self-imitation to track and point to coloured objects

This article describes a developmental system based on information theory implemented on a real robot that learns a model of its own sensory and actuator apparatus. There is no innate knowledge regarding the modalities or representation of the sensory input and the actuators, and the system relies on generic properties of the robot’s world, such as piecewise smooth effects of movement on sensory changes. The robot develops the model of its sensorimotor system by first performing random movements to create an informational map of the sensors. Using this map, the robot then learns what effects the different possible actions have on the sensors. After this developmental process, the robot can perform basic visually guided movement.

We compare the performance of autonomous agents with three different behavior selection architectures (static-threshold, winner-takes-all and voting-based) in terms of survival in a large and complex dynamic virtual environment. Experiment results indicate both advantages and disadvantages when applying each architecture in such environmental conditions, and also shows that the performance of voting-based architecture is significantly sensitive to the balance of rates at which various environmental resources can serve to satisfy an agent's physiological needs. Some problems with behavior selection architectures are identified and possible solutions are proposed.

In this paper, we extend our previous work in investigating the performance of different autobiographic memory control architectures which are developed based on a basic subsumption control architecture for artificial life autonomous agents surviving in a dynamic virtual environment. In our previous work we showed how autonomous agents' survival in a static virtual environment can benefit from autobiographic memory, with a kind of communication of experiences in multi-agent experiments. In the current work we extend the existing memory architecture by enhancing its functionalities and introducing long-term autobiographic memory, which is derived from the inspiration of human memory schema - categorical rules or scripts that psychologists in human memory research believe all humans possess to interpret the word, A large-scale and dynamic virtual environment is created to compare the performance of various types of agents with various memory control architectures, and each agent's behaviour is observed and analyzed together with lifespan measurements. Results confirm our previous research hypothesis that autobiographic memory can prove beneficial - indicating increases in the lifespan of an autonomous, autobiographic, minimal agent. Furthermore, the utility of combining long-term memory with short-term memory is established. We finally discuss the environmental factors influencing the performance of each architecture and the areas for future work

The classical approach to using utility functions suffers from the drawback of having to design and tweak the functions on a case by case basis. Inspired by examples from the animal kingdom, social sciences and games we propose empowerment, a rather universal function, defined as the information-theoretic capacity of an agent's actuation channel. The concept applies to any sensorimotor apparatus. Empowerment as a measure reflects the properties of the apparatus as long as they are observable due to the coupling of sensors and actuators via the environment. Using two simple experiments we also demonstrate how empowerment influences sensor-actuator evolution

The potential for animals to inflict damage on one another whilst competing for indivisible resources is a factor of crucial importance when determining pay-offs to such animals and consequent likelihood of adopting an aggressive resource procurement strategy, a less costly display-based alternative or just a retreat response. Using computer simulations of evolving agents, we assessed the effects of degree of damage potential on social aggression and resource procuring strategies. Furthermore, we assessed the effects of evolving ritualized displays used in contests over resources relative to damage potential. Our results showed that aggressive interactions increased in frequency when the degree of damage potential was low. Ritualized displays tended to reduce aggressive approaches and mortality rate where damage potential was high and low but this was not the case at the intermediate level where aggression and mortality rate increased.

Information distance is used to measure how similar sensorimotor experience is to past experience within a certain temporal horizon. Applied to groups of sensors this gives a mathematical metric on sensorimotor experience over time. We show that for complex data from a robot, large scale similarity of experience can be discovered from the robot perspective, providing a means of building an experiential interaction history

We introduce metrics on sensorimotor experience at various temporal scales based on information-theory. Sensorimotor variables through which the experience of an agent flows are modeled as information sources in the sense of Shannon information theory. Information distance between the constellation of an embodied agent's sensorimotor variables at different moments in time can be taken variable-by-variable or between entire sets of such variables to yield two classes of metrics on sensorimotor experience: the temporal experiential information distance and the Hausdorff metric on experience. Unlike mutual information, these measures each satisfy the metric axioms and thus induce a geometry on the space of experiences with the same temporal scope. Continuity of maps between experiential spaces as well as robotic applications and extensions are discussed

Imitation is a powerful mechanism whereby knowledge may be transferred between agents (both biological and artificial). Key problems on the topic of imitation have emerged in various areas close to artificial intelligence, including the cognitive and social sciences, animal behavior, robotics, human-computer interaction, embodied intelligence, software engineering, programming by example and machine learning. Artificial systems used to study imitation can both test models of imitation derived from observational or neurobiological data on imitation in animals and then apply them to different kinds of nonbiological systems ranging from robots to software agents. A crucial problem in imitation is the correspondence problem, mapping action sequences of the demonstrator and the imitator agent. This problem becomes particularly obvious when the two agents do not share the same embodiment and affordances. This paper describes a new general imitation mechanism called ALICE (action learning for imitation via correspondence between embodiments) that specifically addresses the correspondence problem. The mechanism is implemented and its efficacy illustrated on the "chessworld" testbed that was created to study imitation from an agent-based perspective, i.e., by a particular agent in a particular environment.

Imitation is a powerful mechanism for efficient learning of novel behaviors that both supports and takes advantage of sociality. A fundamental problem for imitation is to create an appropriate (partial) mapping between the body of the system being imitated and the imitator. By considering for each of these two systems an associated automaton (respectively, transformation semigroup) structure, attempts at such mapping can be considered (partial) relational homomorphisms. This article shows how mathematical techniques can be applied to characterize how far a behavior is from a successful imitation and how to evaluate attempts at imitation arising from a particular correspondence between the imitator and model. For the imitator and the imitated, affordances in the agent-environment structural coupling are likely to be different, all the more so in the case of dissimilar embodiment. We argue that the use of what is afforded to the imitator to attain corresponding effects or, as in dance, sequences of effects, is necessary and sufficient for successful imitation. However, the judged degree of success or failure of an attempted behavioral match depends on some externally imposed or in the case ofautonomous agents internally determined criteria on effects of the attempted imitative behavior (including effects attained successively as well as final effects). These criteria correspond to metrics measures of difference which can guide the evaluation of a correspondence, the learning of a correspondence, or learning how to apply one. Metrics on states and sequences of action events in the system-environment coupling allow judgment of similarity for observer-dependent' purposes. This allows one to formally define successful imitation with respect to such criteria. The resulting measures can be used to compare various candidate mappings (e.g., body plan or perception-action correspondences). Additionally, this may be applied in the automated construction and learning of mappings to be used in imitation for artificial, hardware, and software systems.

We develop the rigorous notion of a model for understanding state transition systems by hierarchical coordinate systems. Using this we motivate an algebraic definition of the complexity of biological systems, comparing it to other candidates such as genome size and number of cell types. We show that our complexity measure is the unique maximal complexity measure satisfying a natural set of axioms. This reveals a strong relationship between hierarchical complexity in biological systems and the area of algebra known as global semigroup theory. We then study the rate at which hierarchical complexity can evolve in biological systems assuming evolution is "as slow as possible" from the perspective of computational power of organisms. Explicit bounds on the evolution of complexity are derived showing that, although the evolutionary changes in hierarchical complexity are bounded, in some circumstances complexity may more than double in certain "genius jumps" of evolution. In fact, examples show that our bounds are sharp. We sketch the structure where such complexity jumps are known to occur and note some similarities to previously identified mechanisms in biological evolutionary transitions. We also address the question of, How fast can complexity evolve over longer periods of time? Although complexity may more than double in a single generation, we prove that in a smooth sequence of t "inclusion" steps, complexity may grow at most from N to (N + 1)t + N, a linear function of number of generations t, while for sequences of "mapping" steps it increases by at most t. Thus, despite the fact that there are major transitions in which complexity jumps are possible, over longer periods of time, the growth of complexity may be broken into maximal intervals on which it is bounded above in the manner described.

January 01 2000 The Right Stuff: Appropriate Mathematics for Evolutionary and Developmental Biology (Editors' Introduction to the Special Issue) Chrystopher L. Nehaniv, Chrystopher L. Nehaniv Faculty of Engineering & Information Sciences University of Hertfordshire Hatfield, Hertfordshire AL10 9AB United Kingdom Search for other works by this author on: This Site Google Scholar Günter P. Wagner Günter P. Wagner Department of Ecology and Evolutionary Biology Yale University New Haven, CT 06520 U.S.A. Search for other works by this author on: This Site Google Scholar Author and Article Information Chrystopher L. Nehaniv Faculty of Engineering & Information Sciences University of Hertfordshire Hatfield, Hertfordshire AL10 9AB United Kingdom Günter P. Wagner Department of Ecology and Evolutionary Biology Yale University New Haven, CT 06520 U.S.A. Online Issn: 1530-9185 Print Issn: 1064-5462 © 2000 Massachusetts Institute of Technology2000 Artificial Life (2000) 6 (1): 1–2. https://doi.org/10.1162/106454600568285 Cite Icon Cite Permissions Share Icon Share MailTo Twitter LinkedIn Views Icon Views Article contents Figures & tables Video Audio Supplementary Data Peer Review Search Site Citation Chrystopher L. Nehaniv, Günter P. Wagner; The Right Stuff: Appropriate Mathematics for Evolutionary and Developmental Biology (Editors' Introduction to the Special Issue). Artif Life 2000; 6 (1): 1–2. doi: https://doi.org/10.1162/106454600568285 Download citation file: Ris (Zotero) Reference Manager EasyBib Bookends Mendeley Papers EndNote RefWorks BibTex toolbar search Search Dropdown Menu toolbar search search input Search input auto suggest filter your search All ContentAll JournalsArtificial Life Search Advanced Search This content is only available as a PDF. © 2000 Massachusetts Institute of Technology2000 Article PDF first page preview Close Modal You do not currently have access to this content.

When prototyping the requirement specifications of a large and complex system, it is very difficult to express them in a single model containing all the features. It is often disastrous if the requirement specifications are prototyped in a single level or single window of a single model. Hence multimodel based hierarchical requirement specifications are desirable in order to fully specify large and complex systems. But multimodel based hierarchical requirement specifications tools are then confronted with verification problems for inter model consistency and syntactic correctness. We introduce a predefined graphical structural syntax checker called Model Hierarchy Definition (MHD) which provides interactive hierarchy syntax checking during modeling the structured requirement specifications and predefined lowest level graphical formalisms called Model Hierarchy Model (MHM), defining the meta rules, the graphical diagramming entity and the hierarchical relationship for MHD embedding in the prototyping models. Having MHD and the MHM in the prototyping environment, one obtains higher quality structured requirement specifications when building hierarchical multimodel based prototypes for complex system analysis, and greatly reduces the prototyping time and costs. The MHD and MHM were implemented to support an integrated graphical prototyping system, which we overview, called the Prototyper's Workbench (PWB) for rapid prototyping and analysis of software requirement specifications of target systems.

Computation and representation of data and knowledge have algebraic structure. Using results of C.L. Nehaniv and Krohn-Rhodes, it is possible to introduce mathematical coordinate systems appropriate for the algebra of each type of computation. Such a coordinate system provides a formal model of understanding of the given computational domain, and is characterized by strict hierarchical representation of data, process and knowledge; and exhibits also the advantages of object-oriented systems for representing knowledge-not only at the object-but also process level.>