Encouraging Bystander Assistance for Urban Robots: Introducing Playful Robot Help-Seeking as a Strategy

TL;DR

The paper addresses the challenge of urban delivery robots needing bystander assistance for tasks beyond their capabilities, proposing playful help-seeking as a novel strategy to foster spontaneous bystander collaboration.Using a VR-based within-subject study, the authors compare three strategies—playful help-seeking, verbal help-seeking (work), and emotional expression (care)—across three urban failure scenarios, evaluating unambiguity, politeness, appropriateness, effectiveness, mood, and attitudes toward the robot.Key findings show playful help-seeking increases willingness to help, especially for physically demanding tasks, and enhances acceptance and cognitive trust, while verbal help-seeking is often perceived as impolite and emotionally expressive cues can reduce perceived competence.The study provides design implications and cautions for gameful interfaces in public spaces, highlighting opportunities and limitations for deploying bystander-friendly, game-inspired help-seeking in real urban environments.

Abstract

Robots in urban environments will inevitably encounter situations beyond their capabilities (e.g., delivery robots unable to press traffic light buttons), necessitating bystander assistance. These spontaneous collaborations possess challenges distinct from traditional human-robot collaboration, requiring design investigation and tailored interaction strategies. This study investigates playful help-seeking as a strategy to encourage such bystander assistance. We compared our designed playful help-seeking concepts against two existing robot help-seeking strategies: verbal speech and emotional expression. To assess these strategies and their impact on bystanders' experience and attitudes towards urban robots, we conducted a virtual reality evaluation study with 24 participants. Playful help-seeking enhanced people's willingness to help robots, a tendency more pronounced in scenarios requiring greater physical effort. Verbal help-seeking was perceived less polite, raising stronger discomfort assessments. Emotional expression help-seeking elicited empathy while leading to lower cognitive trust. The triangulation of quantitative and qualitative results highlights considerations for robot help-seeking from bystanders.

Figures (7)

• Figure 1: The design process. Left: Initial sketches with bubble-blasting (top) and PacMan (bottom) style game elements for promoting helpful behaviours. Middle: Photo from the design workshop brainstorming session. Right: Example sketch generated during the workshop, which was sketched on printed a bird's-eye view illustration of the scenario including the robot, bystanders, and text descriptions.
• Figure 2: The design concepts. Left: Games that inspired the design concepts; Right: Illustration of the process of bystanders engaging in playful helping-seeking.
• Figure 3: Key frames of robot facial expression
• Figure 4: Study procedure
• Figure 5: Qualitative assessment of help-seeking strategy (Left), Box plot of participants' Willingness to Help the robot across each scenario (Right). M: Mean, SD: Standard Deviations, *: p < .05, **: p < .01, ***: p < .001