"I am the follower, also the boss": Exploring Different Levels of Autonomy and Machine Forms of Guiding Robots for the Visually Impaired

Title of the Paper

"I am the follower, also the boss": Exploring Different Levels of Autonomy and Machine Forms of Guiding Robots for the Visually Impaired

Paper Information

• Research Domain: Human-Computer Interaction and Assistive Technologies
• Keywords: Guiding robots, visual impairment, navigation, autonomy levels, machine forms, control, trust, safety

Research Background and Problem Statement

• What problems or challenges did the authors identify?

  1. Lack of detailed studies on how guiding robot forms (e.g., cane or vehicle-like) impact navigation efficiency and user experience.
  2. Existing research has not explored whether partial autonomy is necessary for visually impaired individuals, nor the user perception of partial versus full autonomy in robots.

• Why is this problem important? Visually impaired individuals face significant challenges in independent mobility (over 2.2 billion people globally have vision problems). Efficient and safe navigation assistive technologies are crucial for improving their mobility and quality of life.

• Research Motivation and Related Work

  1. Current designs have experimented with various machine forms (e.g., cane-like, vehicle-like robots), but comparative studies are lacking.
  2. Some studies propose fully autonomous guiding devices, but whether users need the ability to switch autonomy levels remains unclear.
  3. This paper aims to fill these research gaps by exploring the actual needs of visually impaired individuals regarding different autonomy levels and machine forms.

Proposed Solution

• What methods or solutions did the authors propose? The authors designed two types of guiding robots, a cane-like form and a vehicle-like form, both equipped with autonomy switching functionality:

  1. Partial autonomy: Users control forward speed via push-pull actions, while the robot guides direction.
  2. Full autonomy: The robot fully controls forward movement and directional navigation.

• What is innovative about this solution?

  1. Combining the impact of machine forms and autonomy levels in one study, providing empirical data support.
  2. Implementing autonomy switching functionality, allowing users to flexibly choose based on situational needs.
  3. Integrating controlled and real-world scenarios in experiments for comprehensive evaluation.

• What are the implementation steps and key technologies used?

  1. Hardware Design:
     • The cane-like robot mimics the size of a traditional white cane, using fiber materials.
     • The vehicle-like robot is modeled after the size of a guide dog, incorporating shock absorption structures and wheel designs.
  2. Control Logic:
     • Based on finite state machines, users can switch between autonomy modes via buttons.
     • In partial autonomy mode, users control speed, with the robot enforcing emergency braking in dangerous situations.
  3. Experiment Design:
     • Controlled experiments (indoor environments comparing navigation efficiency and user experience across autonomy levels and machine forms).
     • Field experiments (real outdoor scenarios recording user preferences and behavior switching).

Research Findings

• What specific findings were obtained?

  1. Controlled Experiment Results:
     • The vehicle-like robot in full autonomy mode performed best in navigation efficiency (speed, path length, etc.) and user experience (low workload, high sense of safety).
     • While partial autonomy had weaker effects in controlled experiments, users still perceived it as valuable in complex scenarios.
  2. Field Experiment Results:
     • In real outdoor scenarios, most users preferred partial autonomy mode to increase their sense of control.
     • In dangerous or complex scenarios (rough terrain, downhill slopes, intersections, etc.), partial autonomy provided higher psychological safety.
  3. Comparison of Machine Forms:
     • Users favored the vehicle-like robot, considering it safer and more stable, even though the cane-like robot was lighter and more familiar.

• What advantages does this solution have compared to existing ones?

  1. Provides comprehensive data support by combining field and controlled experiments.
  2. Innovatively adopts a user-centered design approach for visually impaired individuals, surpassing previous studies focused on single forms or modes.
  3. Offers practical implementation of autonomy switching design, with potential applications in other assistive technologies.

• What were the experimental or evaluation results?

  1. SUS scores: Cane-like robot scored 81, vehicle-like robot scored 88, both indicating high usability.
  2. PIADS scale showed that fully autonomous robots significantly enhanced users' independence, safety, and trust.
  3. Partial autonomy usage rates were significantly higher in complex scenarios, correlating with psychological states.

• Limitations and Future Directions

  1. Limitations:
     • Small sample size, primarily urban participants, limiting generalizability.
     • Robot design is not fully optimized; speed control lacks naturalness, and the vehicle-like robot's bottom space comfort needs improvement.
  2. Future Directions:
     • Expand participant groups, including guide dog users and elderly individuals, to enhance applicability.
     • Optimize robot design, exploring integration of manual user control and automated robot control for speed logic.
     • Develop smarter shared control algorithms to balance user perception and robot performance improvement.

This research provides new design guidelines and empirical support for assistive technologies, contributing significantly to the advancement of human-computer interaction technologies.