At First Light: Expressive Lights in Support of Drone-Initiated Communication

Title of the Paper

At First Light: Expressive Lights in Support of Drone-Initiated Communication

Paper Information

• Research Domain: Human-Computer Interaction (HCI), Human-Drone Interaction (HDI), Nonverbal Communication
• Keywords: Human-Computer Interaction, Drones, Expressive Lights, Nonverbal Communication, Machine-Initiated Communication, Search and Rescue

Research Background and Problem Statement

• Challenges Identified by the Authors:

  1. Drones in search and rescue missions or other scenarios may need to initiate communication with bystanders or victims. However, current drone interactions primarily rely on human-initiated engagement, lacking best practices for drone-initiated communication.
  2. Nonverbal communication methods for drones are limited. Existing approaches, such as using flight trajectories and drone movements, can only convey simple intent information and lack the ability to express complex states.

• Significance:

  1. Drones hold great potential in emergency response and search and rescue missions, but existing communication methods may lead to misunderstandings or fear.
  2. Providing clear expressions of interaction intent can enhance the ability of drones to gain cooperation from target individuals, thereby improving task efficiency.

• Research Motivation and Related Work:

  1. The authors extended human interaction models (e.g., Kendon’s six-stage greeting model) and some best practices in robot interaction to propose an interaction model tailored for drones.
  2. Related studies show that light animations and drone flight trajectories are highly effective in conveying drone intent, but exploration in this area remains insufficient, particularly in the context of drone-initiated communication.

Proposed Solution

• Proposed Approach:

  1. Introduced a four-stage Drone-Initiated Engagement Model (DIEM), comprising Searching, Spotting, Approaching, and Ready to Interact stages.
  2. Developed an LED light animation system ("At First Light") to support DIEM. This system includes two lighting configurations (light strips and ring lights) and 26 light animations designed to suit different interaction stages.

• Innovations:

  1. Adapted traditional human greeting models to the domain of drone-initiated communication, simplifying interaction patterns suitable for drones.
  2. Combined light animations with drone movements to enhance nonverbal expression capabilities, thereby improving user experience and interaction effectiveness.

• Implementation Steps and Technology:

  1. Hardware Design: Integrated an LED lighting system into a commercial drone, including light strips (30 LEDs) and dual ring lights (24 LEDs), equipped with microcontrollers and communication modules for controlling light colors and animations.
  2. Animation Development: Designed 26 animations for the two lighting configurations, categorized into static and dynamic light groups, including brightness, tracking stripes, and gradient effects.
  3. Online Study: Conducted a survey with 156 participants using videos and static images to assess understanding of light animations and color preferences.
  4. Validation Experiment: Used a Wizard of Oz approach to simulate autonomous drone behavior, comparing drones equipped with the lighting system against baseline drones.

Research Outcomes

• Key Findings:

  1. Preliminary validation showed that the light animation-based drone communication system "At First Light" effectively conveyed interaction states across different stages.
  2. Identified the most suitable light animations and color standards for each communication stage:
     • Searching: Recommended animations include “BackForth” and “Chase,” with blue as the preferred color.
     • Spotting: Recommended animation is “Morse,” with green as the preferred color.
     • Approaching: Recommended animation is “OneDecent,” with orange as the preferred color.
     • Ready to Interact: Green with the “Blink” animation performed best.

• Advantages:

  1. Provided a new pathway for nonverbal drone communication, offering a more intuitive and efficient alternative to traditional flight behavior, which conveys limited intent information.
  2. The lighting configuration (especially ring lights) significantly enhanced the drone’s acceptability (friendliness, trustworthiness, and willingness to interact).

• Experimental and Evaluation Results:

  1. The online study revealed that participants could accurately interpret most states, particularly during the "Searching" and "Spotting" stages.
  2. Field experiments demonstrated that even in complex environments, light colors and animations effectively conveyed drone states.
  3. Compared to the baseline, drones equipped with the lighting system elicited more positive and acceptable responses from participants.

• Limitations and Future Directions:

  1. Did not test more complex light patterns or color combinations.
  2. Experimental settings were relatively idealized. Future studies should consider adverse weather conditions, long-distance perspectives, and dynamic scenarios to evaluate the impact on lighting effects.
  3. Suggested exploring the combined effects of lighting and drone flight paths, as well as incorporating multimodal interactions such as verbal feedback.
  4. Proposed investigating the influence of cultural differences on the interpretation of light colors and animations to enhance universality.

This study provides an important reference for designing nonverbal communication for drones and offers valuable insights for expanding drone interaction in the future.