Machines Like Us

The Nico project focuses on the development of anthropomorphic robots that interact with people using natural social cues. Socially-competent robots would have great practical impact; naïve users could interact with these robots in a more natural and effortless way, could command them through social instruction, and could integrate them into daily life. The project seeks to address both the technical challenges involved in constructing these robots and the ways in which they can be used as tools to study human social development.

The technical challenges of building social robots are substantial. The design and construction of a robot that can produce gestures and utterances which can be easily interpreted by a human observer is a challenging mechanical design problem. A more difficult technical challenge will be to build machines that can recognize human social cues such as pointing gestures, direction of gaze, and tone of voice. Existing research succeeds in recognizing a few of these cues in structured situations, requiring visual scenes that have a constant background or audio signals that contain only the voice of a single speaker. This project proposes to build on existing work by integrating techniques from multiple sensory modalities and using models of human social development as a roadmap for constructing more complex social behaviors. A final engineering challenge will be the implementation of a computational infrastructure to support these algorithms. For example, a social robot currently under construction generates over 200 megabytes of sensory data per second. Novel software algorithms and tools for a distributed network of parallel processors will be required to rapidly process this volume of data.

The construction of social robots will be more than a technical accomplishment; social robots can be unique tools to study social skills in humans. Specifically, we propose to use these social robots to study methods for quantifying social behavior, to evaluate models of how infants acquire social skills, and to investigate children's concepts of identity. If we can construct a robot that is capable of perceptually identifying social cues, then this recognition system will provide a quantitative metric of social response. This metric could be a useful diagnostic tool for social development disorders such as autism. Additionally, these robots could be used to evaluate models of how children acquire social skills. Just as simulations of neural networks have been useful in evaluating the applicability of models of neural function, these robots can serve as a test-bed for evaluating models of human development. While we do not assume that the results from our robot implementation will necessarily dictate what happens in the human brain, this novel experiment has the potential to offer insight into the requirements for a valid model. Finally, these robots represent interesting borderline cases in traditional issues of human identity. By observing human children and adults interacting with these robots, we can ask questions about what it is to be alive, to be animate, and to be human.

The Social Robotics Lab at Yale University