What Questions Should Robots Be Able to Answer? A Dataset of User Questions for Explainable Robotics

Lennart Wachowiak, Andrew Coles, Gerard Canal, Oya Celiktutan

24 Oct 2025 3 min read

AI-generated image, based on the article abstract

Quick Insight

What Questions Should Your Home Robot Be Able to Answer?

Ever wondered what you’d actually ask a robot that helps with chores? Researchers have gathered almost 2,000 everyday questions from real people, revealing the curious minds behind our future helpers. Imagine a kitchen robot that not only chops veggies but can also explain, “Why did I pause?” or “What will I do if the pan spills?” The study shows most folks care about simple details—like “Did you finish washing the dishes?”—but the most important queries are the “what‑if” scenarios that keep us safe. Think of it like asking a friend for directions: you want both the exact turn‑by‑turn steps and reassurance that they’ll avoid traffic jams. The data also uncovered a surprise: beginners ask about basic facts, while seasoned tech fans probe deeper strategies. This treasure trove helps engineers design robots that speak our language, log the right info, and give answers we truly need. Imagine a home where your robot not only works but also chats with confidence, making daily life smoother and more trustworthy. That’s the future we’re building—one question at a time.

Short Review

Overview

This article presents a comprehensive dataset of 1,893 user questions aimed at enhancing the interaction between humans and household robots. Collected from 100 participants, the dataset categorizes questions into 12 main categories and 70 subcategories, focusing on various aspects of robot behavior. The study highlights the significance of understanding user inquiries, particularly regarding safety and hypothetical scenarios, which are crucial for developing effective conversational interfaces. The methodology involved creating video and text stimuli to elicit natural language questions, providing valuable insights into user expectations and experiences.

Critical Evaluation

Strengths

The primary strength of this study lies in its systematic approach to data collection and analysis. By employing a diverse participant pool and utilizing both video and text stimuli, the researchers ensured a rich variety of user questions that reflect real-world interactions with robots. The categorization of questions into distinct types, such as execution details and robot capabilities, offers a structured framework for understanding user needs. Furthermore, the application of statistical modeling to assess the importance of different question types adds rigor to the findings, allowing for nuanced insights into user behavior.

Weaknesses

Despite its strengths, the study has notable limitations. The reliance on a specific participant demographic may affect the generalizability of the findings, as cultural and contextual factors can influence user questions. Additionally, the study acknowledges potential ambiguities in user intent, which could complicate the interpretation of results. The focus on a limited set of question categories may overlook other relevant inquiries that could arise in more complex human-robot interactions.

Implications

The implications of this research are significant for the field of explainable robotics. By identifying the types of questions users prioritize, roboticists can better design systems that meet user expectations and enhance the overall interaction experience. This dataset serves as a foundational resource for benchmarking question-answering modules and developing explanation strategies that align with user needs, ultimately contributing to more effective human-robot collaboration.

Conclusion

In summary, this article provides a valuable contribution to the understanding of user interactions with household robots through its extensive dataset of user questions. The findings underscore the importance of tailoring robotic responses to user inquiries, particularly in terms of safety and functionality. As robots become increasingly integrated into daily life, this research lays the groundwork for future studies aimed at improving human-robot interaction and enhancing the usability of robotic systems.