The key idea of this article is to use events as the fundamental structures for the semantic representations of a robot. Events are modeled in terms of conceptual spaces and mappings between spaces. It is shown how the semantics of major word classes can be described with the aid of conceptual spaces in a way that is amenable to computer implementations. An event is represented by two vectors, one force vector representing an action and one result vector representing the effect of the action. The two-vector model is then extended with the thematic roles so that an event is built up from an agent, an action, a patient, and a result. It is shown how the components of an event can be put together to semantic structures that represent the meanings of sentences.
HRI researchers have made major strides in developing robotic architectures that are capable of reading a limited set of social cues and producing behaviors that enhance their likeability and feeling of comfort amongst humans. However, the cues in these models are fairly direct and the interactions largely dyadic. To capture the normative qualities of interaction more robustly, we propose consent as a distinct, critical area for HRI research. Convening important insights in existing HRI work around topics like touch, proxemics, gaze, and moral norms, the notion of consent reveals key expectations that can shape how a robot acts in social space. By sorting various kinds of consent through social and legal doctrine, we delineate empirical and technical questions to meet consent challenges faced in major application domains and robotic roles. Attention to consent could show, for example, how extraordinary, norm-violating actions can be justified by agents and accepted by those around them. We argue that operationalizing ideas from legal scholarship can better guide how robotic systems might cultivate and sustain proper forms of consent.
Robots are expected to possess similar capabilities that humans exhibit during close proximity dyadic interaction. Humans can easily adapt to each other in a multitude of scenarios, ensuring safe and natural interaction. Even though there have been attempts to mimic human motions for robot control, understanding the motion patterns emerging during dyadic interaction has been neglected. In this work, we analyze close proximity human-human interaction and derive an ontology that describes a broad range of possible interaction scenarios by abstracting tasks, and using insights from attention theory. This ontology enables us to group interaction behaviors into separate cases, each of which can be represented by a graph. Using inverse reinforcement learning, we find unique interaction models, represented as combination of cost functions, for each case. The ontology offers a unified, and generic approach to categorically analyze and learn close proximity interaction behaviors that can enhance natural human-robot collaboration.
Many recent studies have shown that behaviors and interaction logic for social robots can be learned automatically from natural examples of human-human interaction by machine learning algorithms, with minimal input from human designers. Oftentimes in human-human interactions, a person?s action is dependent upon actions that occurred much earlier in the interaction. However, in previous approaches to data-driven social robot behavior learning, the robot decides its next action based only on a narrow window of the interaction history. In this work, an analysis of the types of memory-setting and memory-dependent actions that occur in a camera shop interaction scenario are presented. Furthermore, a robot behavior system with a recurrent neural network architecture with gated recurrent units is proposed to tackle the problem of learning relevant memories to enable a shopkeeper robot to perform appropriate memory-dependent actions. In an offline evaluation the proposed system performed appropriate memory-dependent actions at a significantly higher rate than a baseline system without memory. Finally, an analysis of how the recurrent neural network with gated recurrent units learned to represent memories is presented.
Curiosity did not kill the robot: A curiosity-based learning system for a shopkeeper robot
this paper, we present a multi-level time scales framework for analysis of human-robot interaction (HRI). Such framework allows HRI scientists to model the inter-relation between measures and factors of an experiment. Our final goal with introducing this framework is to unify scientific practice in HRI community for better reproduciblility. Our new approach transposes Newell's framework of human actions to model human-robot interaction. Measures from the interaction are sorted in categories (time scales) corresponding to the temporal constraints proposed by Newell. According to this sorting, a bottom-up or top-bottom analysis can then be performed to correlate variables that allows to better explain the interaction. Utilization of this method within a child-robot interaction involving 2 robots and one child playing a memory game is then presented to illustrate the potential and use of this method. Finally we introduce clear guidelines to re-use the framework.