Conference Paper

Human-AI Interaction in Industrial Robotics: Design and Empirical Evaluation of a User Interface for Explainable AI-Based Robot Program Optimization

We are happy to announce that our paper “Human-AI Interaction in Industrial Robotics: Design and Empirical Evaluation of a User Interface for Explainable AI-Based Robot Program Optimization” has been accepted at the 57th CIRP Conference on Manufacturing Systems (CMS) 2024!

The full text of the paper is accessible on ArXiv (Link).

In the paper, we presen an Explanation User Interface (XUI) designed for a deep learning-based robot program optimizer, aimed at making AI-based robotics more accessible in manufacturing settings. The interface features both “Guided” and “Expert” modes to accommodate users with different skill levels, along with explainable AI (XAI) features to help users understand the system’s decisions. The authors present a preliminary user study with 12 participants testing the interface on a robotic gearbox assembly task, followed by their proposal for a larger follow-up study. The study results suggest that while AI experts performed better in complex tasks like model training and parameter optimization, both novice and expert users could successfully complete basic tasks. The interface’s adaptive design and explainability features were generally well-received, though some novice users experienced higher stress levels and mental demands, indicating a need for additional guidance in complex tasks.

Overview of the proposed system: A user interface enables intuitive interaction of a human user with an AI system for robot program optimization.