The deployment of robots in environments shared with humans implies that they must be able to justify or explain their behavior to nonexpert users when the user, or the situation itself, requires it. We propose a framework for robots to generate personalized explanations of their behavior by integrating cause-and-effect structures, social roles, and natural language queries. Robot events are stored as cause-effect pairs in a causal log. Given a human natural language query, the system uses machine learning to identify the matching cause-and-effect entry in the causal log and determine the social role of the inquirer. An initial explanation is generated and is then further refined by a large language model (LLM) to produce linguistically diverse responses tailored to the social role and the query. This approach maintains causal and factual accuracy while providing language variation in the generated explanations. Qualitative and quantitative experiments show that combining the causal information with the social role and the query when generating the explanations yields the most appreciated explanations.

In recent years, there has been growing interest in developing robots capable of explaining their behavior, thereby improving their acceptance by humans with whom they share their environment. Proposed software designs are typically based on the advances being made in conversational systems built on deep learning techniques. However, apart from the ability to formulate explanations, the robot also needs an internal episodic memory, where it stores information from the continuous stream of experiences. Most previous proposals are designed to deal with short streams of episodic data (several minutes long). With the aim of managing larger experiences, we propose in this work a high-level episodic memory, where relevant events are abstracted to natural language concepts. The proposed framework is intimately linked to a software architecture in which the explanations, whether externalized or not, are shaped internally in a collaborative process involving the task-oriented software agents that make up the architecture. The core of this process is a runtime knowledge model, employed as working memory whose evolution allows for capturing the causal events stored in the episodic memory. We present several use cases that illustrate how the suggested framework allows an autonomous robot to generate correct and relevant explanations of its actions and behavior.

This study investigates user experiences of interactions with two types of robots: Pepper, a social humanoid robot, and Double 3, a self-driving telepresence robot. Conducted in a controlled setting with a specific participant group, this research aims to understand how the design and functionality of these robots influence user perception, interaction patterns, and emotional responses. The findings reveal diverse participant reactions, highlighting the importance of adaptability, effective communication, autonomy, and perceived credibility in robot design. Participants showed mixed responses to human-like emotional displays and expressed a desire for robots capable of more nuanced and reliable behaviors. Trust in robots was influenced by their perceived functionality and reliability. Despite limitations in sample size, the study provides insights into the ethical and social considerations of integrating AI in public and professional spaces, offering guidance for enhancing user-centered designs and expanding applications for social and telepresence robots in society.

The paper reports an empirical study of the effect of human treatment of a robot on the social perception of the robot's behavior. The study employed an enacted interaction between an anthropomorphic "waiter" robot and two customers. The robot and one of the customers (acted out by a researcher) were following four different interaction scripts, representing all combinations of mutual politeness and impoliteness of the robot and the customer. The participants (N=24, within-subject design) were assigned the role of an "included observer", that is, a fellow customer who was present in the situation without being actively involved in the interactions. The participants assessed how they experienced the interaction scenarios by providing Likert scale scores and free-text responses. The results indicate that while impolite robots' behavior was generally assessed negatively, it was commonly perceived as more justifiable and fairer if the robot was treated impolitely by the human. Politeness reciprocity expectations in the context of the social perception of robots are discussed.

This paper presents a taxonomy of agents’ embodiment in physical and virtual environments. It categorizes embodiment based on five entities: the agent being embodied, the possible mediator of the embodiment, the environment in which sensing and acting take place, the degree of body, and the intertwining of body, mind, and environment. The taxonomy is applied to a wide range of embodiment of humans, artifacts, and programs, including recent technological and scientific innovations related to virtual reality, augmented reality, telepresence, the metaverse, digital twins, and large language models. The presented taxonomy is a powerful tool to analyze, clarify, and compare complex cases of embodiment. For example, it makes the choice between a dualistic and non-dualistic perspective of an agent’s embodiment explicit and clear. The taxonomy also aided us to formulate the term “embodiment by proxy” to denote how seemingly non-embodied agents may affect the world by using humans as “extended arms”. We also introduce the concept “off-line embodiment” to describe large language models’ ability to create an illusion of human perception.

The latest developments of chatbots driven by Large Language Models (LLMs), more specifically ChatGPT, have shaken the foundations of how text is created, and may drastically reduce and change the need, ability, and valuation of human writing. Furthermore, our trust in the written word is likely to decrease, as an increasing proportion of all written text will be AI-generated – and potentially incorrect. In this essay, I discuss these implications and possible scenarios for us humans, and for AI itself.

Causality is relatively unexplored in robotics even if it is highly relevant, in several respects. In this paper, we study how a robot’s causal understanding can be improved by allowing the robot to ask humans causal questions. We propose a general algorithm for selecting direct causal effects to ask about, given a partial causal representation (using partially directed acyclic graphs, PDAGs) obtained from observational data. We propose three versions of the algorithm inspired by different causal discovery techniques, such as constraint-based, score-based, and interventions. We evaluate the versions in a simulation study and our results show that asking causal questions improves the causal representation over all simulated scenarios. Further, the results show that asking causal questions based on PDAGs discovered from data provides a significant improvement compared to asking questions at random, and the version inspired by score-based techniques performs particularly well over all simulated experiments.

When a robot is asked to verbalize its plan it can do it in many ways. For example, a seemingly natural strategy is incremental, where the robot verbalizes its planned actions in plan order. However, an important aspect of this type of strategy is that it misses considerations on what is effectively informative to communicate, because not considering what the user knows prior to explanations. In this paper we propose a verbalization strategy to communicate robot plans informatively, by measuring the information gain that verbalizations have against a second-order theory of mind of the user capturing his prior knowledge on the robot. As shown in our experiments, this strategy allows to understand the robot’s goal much quicker than by using strategies such as increasing or decreasing plan order. In addition, following our formulation we hint to what is informative and why when a robot communicates its plan.

In this paper we propose a method to augment a Reinforcement Learning agent with legibility. This method is inspired by the literature in Explainable Planning and allows to regularize the agent's policy after training, and without requiring to modify its learning algorithm. This is achieved by evaluating how the agent's optimal policy may produce observations that would make an observer model to infer a wrong policy. In our formulation, the decision boundary introduced by legibility impacts the states in which the agent's policy returns an action that is non-legible because having high likelihood also in other policies. In these cases, a trade-off between such action, and legible/sub-optimal action is made. We tested our method in a grid-world environment highlighting how legibility impacts the agent's optimal policy, and gathered both quantitative and qualitative results. In addition, we discuss how the proposed regularization generalizes over methods functioning with goal-driven policies, because applicable to general policies of which goal-driven policies are a special case.

In this paper we illustrate a novel architecture generating interpretable behavior and explanations. We refer to this architecture as the Mirror Agent Model because it defines the observer model, that is the target of explicit and implicit communications, as a mirror of the agent's. With the goal of providing a general understanding of this work, we firstly show prior relevant results addressing the informative communication of agents intentions and the production of legible behavior. In the second part of the paper we furnish the architecture with novel capabilities for explanations through off-the-shelf saliency methods, followed by preliminary qualitative results.

Wizard-of-Oz experiments play a vital role in Human-Robot Interaction (HRI), as they allow for quick and simple hypothesis testing. Still, a publicly available general tool to conduct such experiments is currently not available in the research community, and researchers often develop and implement their own tools, customized for each individual experiment. Besides being inefficient in terms of programming efforts, this also makes it harder for non-technical researchers to conduct Wizard-of-Oz experiments. In this paper, we present a general and easy-to-use tool for the Pepper robot, one of the most commonly used robots in this context. While we provide the concrete interface for Pepper robots only, the system architecture is independent of the type of robot and can be adapted for other robots. A configuration file, which saves experiment-specific parameters, enables a quick setup for reproducible and repeatable Wizard-of-Oz experiments. A central server provides a graphical interface via a browser while handling the mapping of user input to actions on the robot. In our interface, keyboard shortcuts may be assigned to phrases, gestures, and composite behaviors to simplify and speed up control of the robot. The interface is lightweight and independent of the operating system. Our initial tests confirm that the system is functional, flexible, and easy to use. The interface, including source code, is made commonly available, and we hope that it will be useful for researchers with any background who want to conduct HRI experiments.

This paper describes a recently initiated research project aiming at supporting development of computerised dialogue systems that handle breaches of conversational norms such as the Gricean maxims, which describe how dialogue participants ideally form their utterances in order to be informative, relevant, brief, etc. Our approach is to model dialogue and norms with co-operating distributed grammar systems (CDGSs), and to develop methods to detect breaches and to handle them in dialogue systems for verbal human-robot interaction.

The ability to understand the surrounding environment and being able to communicate with interacting humans are important functionalities for many automated systems where visual input (e.g., images, video) and natural language input (speech or text) have to be related to each other. Possible applications are automatic image caption generation, interactive surveillance systems, or human robot interaction. In this paper, we propose algorithms for automatic responses to natural language queries about an image. Our approach uses a predefined neural net for detection of bounding boxes and objects in images, spatial relations between bounding boxes are modeled with a neural net, the queries are analyzed with a syntactic parser, and algorithms to map natural language to properties in the images are introduced. The algorithms make use of semantic similarity and antonyms. We evaluate the performance of our approach with test users assessing the quality of our system’s generated answers.

In this article, we investigate the role of causal reasoning in robotics research. Inspired by a categorization of human causal cognition, we propose a categorization of robot causal cognition. For each category, we identify related earlier work in robotics and also connect to research in other sciences. While the proposed categories mainly cover the sense–plan–act level of robotics, we also identify a number of higher-level aspects and areas of robotics research where causation plays an important role, for example, understandability, machine ethics, and robotics research methodology. Overall, we conclude that causation underlies several problem formulations in robotics, but it is still surprisingly absent in published research, in particular when it comes to explicit mentioning and using of causal concepts and terms. We discuss the reasons for, and consequences of, this and hope that this article clarifies the broad and deep connections between causal reasoning and robotics and also by pointing at the close connections to other research areas. At best, this will also contribute to a “causal revolution” in robotics.

Autonomous agents may be required to form an understanding of other agents for which they don’t possess a model. In such cases, they must rely on their previously gathered knowledge of agents, and ground the observed behaviors in the models this knowledge describes by theory of mind reasoning. To give flesh to this process, in this paper we propose an algorithm to ground observations on a combination of priorly possessed Belief-Desire-Intention models, while using rationality to infer unobservable variables. This allows to jointly infer beliefs, goals and intentions of an unknown observed agent by using only available models.

In public media as well as in scientific publications, the term \emph{bias} is used in conjunction with machine learning in many different contexts, and with many different meanings. This paper proposes a taxonomy of these different meanings, terminology, and definitions by surveying the, primarily scientific, literature on machine learning. In some cases, we suggest extensions and modifications to promote a clear terminology and completeness. The survey is followed by an analysis and discussion on how different types of biases are connected and depend on each other. We conclude that there is a complex relation between bias occurring in the machine learning pipeline that leads to a model, and the eventual bias of the model (which is typically related to social discrimination). The former bias may or may not influence the latter, in a sometimes bad, and sometime good way.

This paper describes ongoing work on the development of a service robot for serving drinks to people sitting at tables, for example in the recreation room of a care-house. The robot, denoted the Tr ...

The term understandable robots refers to robots making their actions and intentions understandable (or explainable) to humans. To support understandability of a team of collaborating robots we use natural language to let the robots verbalize what they do and plan to do. Our solution is based on Cooperating Distributed Grammar Systems for plan derivation and a Multi-agent algorithm for coordination of robot actions. We implemented and evaluated our solution on a team of three Pepper robots that work collaboratively to move an object on a table, thereby coordinating their capabilities and actions and verbalizing their actions and intentions. In a series of experiments, our solution not only successful demonstrated collaboration and task fulfilment, but also considerable variation, both regarding actions and generated natural language utterances.