Assessing Transparency of Robots, Exoskeletons, and Assistive Devices: A Systematic Review.

Area of Science:

• Robotics and Human-Computer Interaction
• Rehabilitation Engineering
• Assistive Technology

Background:

• Transparency is critical for safe and effective human-robot interaction in fields like rehabilitation and industry.
• Existing literature lacks a unified definition and measurement approach for transparency in robots, exoskeletons, and assistive devices (READs).

Purpose of the Study:

• To systematically review and analyze how transparency is conceptualized and evaluated in scientific studies on READs.
• To identify inconsistencies in definitions and methodologies for assessing transparency in READs.

Main Methods:

• Conducted a structured literature search across three major scientific databases.
• Screened 400 articles, with 20 selected for in-depth review.
• Analyzed definitions, evaluation metrics, and focus areas of transparency in the selected studies.

Main Results:

• Significant heterogeneity exists in the definition and evaluation of transparency across studies.
• Most studies define transparency mechanically or kinematically, often neglecting user impact and perception.
• Current evaluation metrics predominantly rely on device-centric kinematic and mechanical data, with limited exploration of user-centered or physiological assessments.

Conclusions:

• The inconsistent definition and evaluation of transparency hinder cross-study comparability and the development of standardized assessment frameworks.
• There is a pressing need for a shared taxonomy and a standardized framework for evaluating transparency in READs.
• Developing such standards will facilitate more reliable comparisons and advance the creation of effective, user-centered READs.