Volume 6, Issue 1

This paper introduces a benchmark framework to evaluate the performance of reaching motion generation approaches that learn from demonstrated examples. The system implements ten different performance measures for typical generalization tasks in robotics using open source MATLAB software. Systematic comparisons are based on a default training data set of human motions, which specify the respective ground truth. In technical terms, an evaluated motion generation method needs to compute velocities, given a state provided by the simulation system. This however is agnostic to how this is done by the method or how the methods learns from the provided demonstrations. The framework focuses on robustness, which is tested statistically by sampling from a set of perturbation scenarios. These perturbations interfere with motion generation and challenge its generalization ability. The benchmark thus helps to identify the strengths and weaknesses of competing approaches, while allowing the user the opportunity to configure the weightings between different measures.

The use of service robots to assist ageing people in their own homes has the potential to allow people to maintain their independence, increasing their health and quality of life. In many assistive applications, robots perform tasks on people’s behalf that they are unable or unwilling to monitor directly. It is important that users be given useful and appropriate information about task progress. People being assisted in homes and other realworld environments are likely be engaged in other activities while they wait for a service, so information should also be presented in an appropriate, nonintrusive manner. This paper presents a human-robot interaction experiment investigatingwhat type of feedback people prefer in verbal updates by a service robot about distributed assistive services. People found feedback about time until task completion more useful than feedback about events in task progress or no feedback. We also discuss future research directions that involve giving non-expert users more input into the task planning process when delays or failures occur that necessitate replanning or modifying goals.

The quality of life of people with special needs, such as residents of healthcare facilities, may be improved through operating social telepresence robots that provide the ability to participate in remote activities with friends or family. However, to date, such platforms do not exist for this population. Methodology: Our research utilized an iterative, bottomup, user-centered approach, drawing upon our assistive robotics experiences. Based on the findings of our formative user studies, we developed an augmented reality user interface for our social telepresence robot. Our user interface focuses primarily on the human-human interaction and communication through video, providing support for semi-autonomous navigation. We conducted a case study (n=4) with our target population in which the robot was used to visit a remote art gallery. Results: All of the participants were able to operate the robot to explore the gallery, form opinions about the exhibits, and engage in conversation. Significance: This case study demonstrates that people from our target population can successfully engage in the active role of operating a telepresence robot.

In thiswork, a real-time system able to automatically recognize soft-biometric traits is introduced and used to improve the capability of a humanoid robot to interact with humans. In particular the proposed system is able to estimate gender and age of humans in images acquired from the embedded camera of the robot. This knowledge allows the robot to properly react with customized behaviors related to the gender/age of the interacting individuals. The system is able to handle multiple persons in the same acquired image, recognizing the age and gender of each person in the robot’s field of view. These features make the robot particularly suitable to be used in socially assistive applications.

In this article, we present the user-centered development of the service robot IURO. IURO’s goal is to find the way to a designated place in town without any previous map knowledge, just by retrieving information from asking pedestrians for directions. We present the 3-years development process,which involved a series of studies on its appearance, communication model, feedback modalities, and social navigation mechanisms. Our main contribution lies within the final field trial.With the autonomous IURO platform, we performed a series of six way-finding runs (over 24 hours of run-time in total) in the city center of Munich, Germany. The robot interacted with approximately 100 pedestrians of which 36 interactions included a full route dialogue. A variety of empirical methods was used to explore reactions of primary users (pedestrians who actually interacted with the robot) and secondary users (bystanders who observed others interacting). The gathered data provides insights into usability, user experience, and acceptance of IURO and allowed us deriving recommendations for the development of other socially interactive robots.

Action affordance learning based on visual sensory information is a crucial problem within the development of cognitive agents. In this paper, we present a method for learning action affordances based on basic visual features, which can vary in their granularity, order of combination and semantic content. The method is provided with a large and structured set of visual features, motivated by the visual hierarchy in primates and finds relevant feature action associations automatically. We apply our method in a simulated environment on three different object sets for the case of grasp affordance learning. For box objects,we achieve a 0.90 success probability, 0.80 for round objects and up to 0.75 for open objects, when presented with novel objects. In thiswork,we demonstrate, in particular, the effect of choosing appropriate feature representations. We demonstrate a significant performance improvement by increasing the complexity of the perceptual representation. By that, we present important insights in how the design of the feature space influences the actual learning problem.

Future applications of robotic technologies will involve interactions with non-expert humans as machines will assume the role of companions, teachers or healthcare assistants. In all those tasks social behavior is a key ability that needs to be systematically investigated and modelled at the lowest level, as even a minor inconsistency of the robot’s behavior can greatly affect the way humans will perceive it and react to it. Here we propose an integrated architecture for generating a socially competent robot.We validate our architecture using a humanoid robot, demonstrating that gaze, eye contact and utilitarian emotions play an essential role in the psychological validity or social salience of Human-Robot Interaction (HRI). We show that this social salience affects both the empathic bonding between the human and a humanoid robot and, to a certain extent, the attribution of a Theory of Mind (ToM). More specifically, we investigate whether these social cues affect other utilitarian aspects of the interaction such as knowledge transfer within a teaching context.

Enabling robots to seamlessly operate as part of smart spaces is an important and extended challenge for robotics R&D and a key enabler for a range of advanced robotic applications, such as AmbientAssisted Living (AAL) and home automation. The integration of these technologies is currently being pursued from two largely distinct view-points: On the one hand, people-centred initiatives focus on improving the user’s acceptance by tackling human-robot interaction (HRI) issues, often adopting a social robotic approach, and by giving to the designer and - in a limited degree – to the final user(s), control on personalization and product customisation features. On the other hand, technologically-driven initiatives are building impersonal but intelligent systems that are able to pro-actively and autonomously adapt their operations to fit changing requirements and evolving users’ needs, but which largely ignore and do not leverage human-robot interaction and may thus lead to poor user experience and user acceptance. In order to inform the development of a new generation of smart robotic spaces, this paper analyses and compares different research strands with a view to proposing possible integrated solutions with both advanced HRI and online adaptation capabilities.

The use of Smart Environments in the delivery of pervasive care is a research topic that has witnessed increasing interest in recent years. These environments aim to deliver pervasive care through ubiquitous sensing by monitoring the occupants Activities of Daily Living. In order for these environments to succeed in achieving their goal, it is crucial that sensors deployed in the environment perform faultlessly. In this research we investigate addressing anomalous sensor behavior through the utilization of a mobile robot. The robot’s role is twofold; it must provide substitution in the presence of suspected sensor faults and act as an observer of anomalous sensor behavior in order to understand the changes that occur in the behavior of sensors deployed within the environment over time. The aim of this work is to explore a paradigm shift to the use of Autonomic Ambient Assisted Living.We have discovered that the use of a mobile robot is a viable means of introducing this paradigm to a Smart Environment.

This paper presents a comprehensive and detailed analysis of the elementary building blocks of neurally inspired architectures for cognitive robotics. It provides a brief outline of the fundamental principles by which biological nervous systems link to the environment in terms of perception, cognition, and behavior. We describe a class of dynamic neural activation variable based on these principles. We show that these dynamic neurons have the appropriate stability properties.Adding even simple connections between a small number of nodes is sufficient to constitute systems that make important decisions. Going through these mechanisms in detail, this paper should facilitate the design of neurally inspired architectures for behavior generation in robotic agents.

Parsing of action sequences is the process of segmenting observed behavior into individual actions. In robotics, this process is critical for imitation learning from observation and for representing an observed behavior in a form that may be communicated to a human. In this paper, we develop a model for action parsing, based on our understanding of principles of grounded cognitive processes, such as perceptual decision making, behavioral organization, and memory formation.We present a neural-dynamic architecture, in which action sequences are parsed using a mathematical and conceptual framework for embodied cognition—the Dynamic Field Theory. In this framework, we introduce a novel mechanism, which allows us to detect and memorize actions that are extended in time and are parametrized by the target object of an action. The core properties of the architecture are demonstrated in a set of simple, proof-of-concept experiments.

In order to proceed along an action sequence, an autonomous agent has to recognize that the intended final condition of the previous action has been achieved. In previous work, we have shown how a sequence of actions can be generated by an embodied agent using a neural-dynamic architecture for behavioral organization, in which each action has an intention and condition of satisfaction. These components are represented by dynamic neural fields, and are coupled to motors and sensors of the robotic agent.Here,we demonstratehowthemappings between intended actions and their resulting conditions may be learned, rather than pre-wired.We use reward-gated associative learning, in which, over many instances of externally validated goal achievement, the conditions that are expected to result with goal achievement are learned. After learning, the external reward is not needed to recognize that the expected outcome has been achieved. This method was implemented, using dynamic neural fields, and tested on a real-world E-Puck mobile robot and a simulated NAO humanoid robot.

This paper describes a method for the detection of textureless objects. Our target objects include furniture and home appliances, which have no rich textural features or characteristic shapes. Focusing on the ease of application, we define a model that represents objects in terms of three-dimensional edgels and surfaces. Object detection is performed by superimposing input data on the model. A two-stage algorithm is applied to bring out object poses. Surfaces are used to extract candidates fromthe input data, and edgels are then used to identify the pose of a target object using two-dimensional template matching. Experiments using four real furniture and home appliances were performed to show the feasibility of the proposed method.We suggest the possible applicability in occlusion and clutter conditions.