Abstract
While drones have traditionally served as industrial tools, a new class of "Companion Unmanned Aerial Vehicles" (UAVs) is emerging, designed for social interaction and emotional connection. This survey synthesizes nearly two decades of Human-Drone Interaction (HDI) research, introducing a "Perceptual Map" that balances technical autonomy with social intelligence. The authors identify critical gaps in current research and propose a roadmap for transitioning drones from simple tools to true social companions.
Introduction
The field of social robotics has long featured companions like the robotic seal Paro or the dog Aibo. Drones represent a new frontier in this domain, yet the field of Human-Drone Interaction (HDI) remains fragmented. Engineering research typically prioritizes flight stability and obstacle avoidance, while social science focuses on human perception of drone behavior. This survey addresses the lack of a unified framework by categorizing drones not just by their utility, but by their potential for companionship.
The Perceptual Map of UAVs
To organize the diverse landscape of HDI research, the authors propose a Perceptual Map defined by two primary axes: the degree of autonomy and the degree of sociability.
Most current research falls into two distinct silos. "Engineering efforts" focus on moving drones from remote-controlled to autonomous states (the vertical axis), while "Interaction studies" focus on moving them from tools to social agents (the horizontal axis). The "Companion UAV" exists in the top-right quadrant—a socially aware, fully autonomous agent. Currently, very few projects successfully bridge this gap.
Physical Classification and the Case for Blimps
UAVs can be broadly categorized by their physical characteristics, specifically mass and flight time. This classification is crucial for companionship, as the physical form dictates safety, endurance, and social acceptance.
While multirotors (quadcopters) are the industry standard, they often make poor companions due to high-pitched noise (>80dB) and safety risks from exposed blades. The authors argue that blimps or balloon-type UAVs are superior for long-term companionship. They are inherently safer to touch, quieter, and offer significantly longer flight times, making them better suited for persistent social presence in human environments.
Key Insights in Social Interaction
- Movement as Language: Given the noise constraints that hinder speech recognition, drones must communicate through "non-linguistic utterances." This includes LED light patterns or specific flight "dances" to signal intent.
- Holonomic Flight: Drones capable of moving sideways without tilting (holonomic flight) are perceived as more "natural" and less threatening than traditional drones that must pitch and roll to change direction.
- Social Perception: Research shows that shape matters; rounder, softer designs are generally more socially accepted than industrial, skeletal frames.
Future Directions
The survey identifies several "missing links" for future research:
- Accompanying Models: Developing standardized models for how a drone should physically accompany a human (e.g., side-by-side vs. following).
- Multi-modal Sensing: Integrating face, gesture, and voice recognition simultaneously for more robust social interaction.
- Long-term Studies: Moving beyond "one-off" experiments to understand how human-drone relationships evolve over weeks or months.
- Noise Reduction: Overcoming the "noise barrier" is essential for the widespread adoption of indoor companion drones.
Summary updated