Matt Berlin

The huggable: a therapeutic robotic companion for relational, affective touch

Numerous studies have shown the positive benefits of companion animal therapy. Unfortunately, companion animals are not always available. The Huggable is a new type of robotic companion being designed specifically for such cases. It features a full body “sensitive skin” for relational affective touch, silent, muscle-like, voice coil actuators, an embedded PC with data collection and networking capabilities. In this paper we briefly describe the Huggable and propose a live demonstration of the robot.

Using perspective taking to learn from ambiguous demonstrations

Abstract This paper addresses an important issue in learning from demonstrations that are provided by “naive” human teachers—people who do not have expertise in the machine learning algorithms used by the robot. We therefore entertain the possibility that, whereas the average human user may provide sensible demonstrations from a human’s perspective, these same demonstrations may be insufficient, incomplete, ambiguous, or otherwise “flawed” from the perspective of the training set needed by the learning algorithm to generalize properly. To address this issue, we present a system where the robot is modeled as a socially engaged and socially cognitive learner. We illustrate the merits of this approach through an example where the robot is able to correctly learn from “flawed” demonstrations by taking the visual perspective of the human instructor to clarify potential ambiguities.

An Embodied Cognition Approach to Mindreading Skills for Socially Intelligent Robots

Future applications for personal robots motivate research into developing robots that are intelligent in their interactions with people. Toward this goal, in this paper we present an integrated socio-cognitive architecture to endow an anthropomorphic robot with the ability to infer mental states such as beliefs, intents, and desires from the observable behavior of its human partner. The design of our architecture is informed by recent findings from neuroscience and embodies cognition that reveals how living systems leverage their physical and cognitive embodiment through simulation-theoretic mechanisms to infer the mental states of others. We assess the robots mindreading skills on a suite of benchmark tasks where the robot interacts with a human partner in a cooperative scenario and a learning scenario. In addition, we have conducted human subjects experiments using the same task scenarios to assess human performance on these tasks and to compare the robots performance with that of people. In the process, our human subject studies also reveal some interesting insights into human behavior.

An embodied computational model of social referencing

Social referencing is the tendency to use the emotional reaction of another to help form ones own affective appraisal of a novel situation, which is then used to guide subsequent behavior. It is an important form of emotional communication and is a developmental milestone for human infants in their ability to learn about their environment through social means. In this paper, we present a biologically-inspired computational model of social referencing for our expressive, anthropomorphic robot that consists of three interacting systems: emotional empathy through facial imitation, a shared attention mechanism, and an affective memory system. This model presents opportunities for understanding how these mechanisms might interact to enable social referencing behavior in humans.

Action parsing and goal inference using self as simulator

The ability to understand a teammates actions in terms of goals and other mental states is an important element of cooperative behavior. Simulation theory argues in favor of an embodied approach whereby humans reuse parts of their cognitive structure for not only generating behavior, but also for simulating the mental states responsible for generating that behavior in others. We present our simulation-theoretic approach and demonstrates its performance in a collaborative task scenario. The robot offers its human teammate assistance by either inferring the humans belief states to anticipate their informational needs, or inferring the humans goal states to physically help the human achieve those goals.

Mobile, dexterous, social robots for mobile manipulation and human-robot interaction

The purpose of this platform is to support research and education goals in human-robot interaction and mobile manipulation with applications that require the integration of these abilities. In particular, our research aims to develop personal robots that work with people as capable teammates to assist in eldercare, healthcare, domestic chores, and other physical tasks that require robots to serve as competent members of human-robot teams. The robot’s small, agile design is particularly well suited to human-robot interaction and coordination in human living spaces. Our collaborators include the Laboratory for Perceptual Robotics at the University of Massachusetts at Amherst, Xitome Design, Meka Robotics, and digitROBOTICS.

Tega: A Social Robot

Tega is a new expressive “squash and stretch”, Android-based social robot platform, designed to enable long-term interactions with children.

Spatial scaffolding cues for interactive robot learning

Spatial scaffolding is a naturally occurring human teaching behavior, in which teachers use their bodies to spatially structure the learning environment to direct the attention of the learner. Robotic systems can take advantage of simple, highly reliable spatial scaffolding cues to learn from human teachers. We present an integrated robotic architecture that combines social attention and machine learning components to learn tasks effectively from natural spatial scaffolding interactions with human teachers. We evaluate the performance of this architecture via a human subjects experiment which examines our humanoid robotpsilas ability to learn from live interactions with human teachers in a secret-constraint task domain. This evaluation provides quantitative evidence for the utility of spatial scaffolding cues to systems that learn from natural human teaching behavior.

Teaching Robots via Natural Nonverbal Cues

As personal robots enter the social environments of our workplaces and homes, it will be important for them to be able to learn from a wide demographic of people. Our research seeks to identify simple, natural, and prevalent human teaching cues that are useful for directing the attention of robot learners so that robots can learn efficiently and effectively from these interactions.