Improving Trust Estimation in Human-Robot Collaboration Using Beta Reputation at Fine-Grained Timescales
Resul Dagdanov, Milan Andrejevic, Dikai Liu, Chin-Teng Lin
AI summary
Problem
Existing trust models only update after task completion and require labor-intensive manual reward design, preventing robots from capturing continuous trust dynamics during collaboration.
Approach
The method uses maximum entropy optimization and human demonstrations to automatically construct a continuous reward function, which is then integrated with a beta reputation model to update trust estimates at every timestep.
Key results
- Automated continuous reward function generation via maximum entropy optimization
- Real-time trust estimation at fine-grained timesteps using beta reputation
- Elimination of manual performance indicator design
- Mathematical mapping of continuous rewards to probabilistic trust dynamics
Why it matters
Enables robots to adapt behavior in real-time based on accurate trust estimates, improving collaboration effectiveness while reducing manual engineering effort.
Abstract
When interacting with each other, humans adjust their behavior based on perceived trust. To achieve similar adaptability, robots must accurately estimate human trust at sufficiently granular timescales while collaborating with humans. Beta reputation is a popular way to formalize a mathematical estimation of human trust. However, it relies on binary per- formance, which updates trust estimations only after each task concludes. Additionally, manually crafting a reward function is the usual method of building a performance indicator, which is labor-intensive and time-consuming. These limitations prevent efficient capture of continuous trust changes at more granular timescales throughout the collaboration task. Therefore, this letter presents a new framework for the estimation of human trust using beta reputation at fine-grained timescales. To achieve granularity in beta reputation, we utilize continuous reward values to update trust estimates at each timestep of a task. We construct a continuous reward function using maximum entropy optimization to eliminate the need for the laborious specification of a performance indicator. The proposed framework improves trust estimations by increasing accuracy, eliminating the need to manually craft a reward function, and advancing toward the development of more intelligent robots. 1