Magic in Human-Robot Interaction (HRI)

TL;DR

The paper investigates magic as a design lens for human-robot interaction, proposing experiential, supernatural, and illusory dimensions to enrich user experiences beyond utilitarian robot performance. It conducts a rapid literature review across these dimensions, followed by speculative ideation and a practical Baxter prototype that performs four magic tricks using multispectral sensing and ROS-based pipelines. Key contributions include a taxonomy of magic in HRI, design ideas for magical artifacts and behaviors, and a demonstrated prototype that earned a robotics competition award. The work highlights potential benefits for affect, engagement, and well-being in home assistants and autonomous vehicles, while acknowledging challenges around perception, transparency, and ethical considerations, and calls for further research to advance practical implementations.

Abstract

"Magic" is referred to here and there in the robotics literature, from "magical moments" afforded by a mobile bubble machine, to "spells" intended to entertain and motivate children--but what exactly could this concept mean for designers? Here, we present (1) some theoretical discussion on how magic could inform interaction designs based on reviewing the literature, followed by (2) a practical description of using such ideas to develop a simplified prototype, which received an award in an international robot magic competition. Although this topic can be considered unusual and some negative connotations exist (e.g., unrealistic thinking can be referred to as magical), our results seem to suggest that magic, in the experiential, supernatural, and illusory senses of the term, could be useful to consider in various robot design contexts, also for artifacts like home assistants and autonomous vehicles--thus, inviting further discussion and exploration.

Figures (4)

• Figure 1: Basic Concept: Robots can incorporate "magic" in its (a) experiential sense (e.g., in play and affection), (b) supernatural sense (e.g., in emulating fantastical creatures and objects), and (c) illusory sense (e.g., in tricks and Wizard-of-Oz scenarios).
• Figure 2: Some "provocatype"-like ideas about how magic could be incorporated into interactions with robots: a) a HA surprises its owner with a present (money it has earned helping others in its spare time), b) HAs accompany, entertain, and protect a couple on a date, c) an AV projects a fake pattern of a face on its window to defeat unlawful face recognition technology, d) an AV tells its passenger's fortune
• Figure 3: The system set-up (components in gray were not used in the tricks).
• Figure 4: Scenes from the magic performance: a) the robot opens its "wand" to reveal a message, b) the robot sees through a black box using a thermal camera, c) the robot detects sunscreen on a hand via an UV camera, and d) the robot pretends it knew which color a human was thinking of using a standard "cold reading"/"rainbow ruse"-style magic trick.