A Resilient and Effective Task Scheduling Approach for Industrial Human-Robot Collaboration

Area of Science:

• Robotics
• Artificial Intelligence
• Industrial Automation

Background:

• Effective task scheduling is a major challenge in human-robot collaboration (HRC) due to uncertainties in shared industrial workspaces.
• Traditional offline scheduling strategies are insufficient for dynamic operational environments.
• The need for resilient and adaptive task scheduling in collaborative robotics is critical.

Purpose of the Study:

• To present a novel online framework for resilient and reliable task scheduling in HRC.
• To address operational uncertainties, human/robot errors, varying operator skills, and actor substitutions.
• To maintain efficient schedules by promoting parallel human-robot work.

Main Methods:

• Utilizing AND/OR graphs to represent collaborative jobs and potential deviations.
• Developing an adaptive architecture that selects optimal collaboration paths.
• Implementing local AND/OR graph adaptation to manage failures during operation.

Main Results:

• The proposed framework demonstrates resilience against operational deviations and errors.
• Efficient parallel human-robot work is promoted, enhancing collaboration.
• Successful validation in an industrial assembly scenario using a Franka Emika Panda robot.

Conclusions:

• The novel online framework provides a robust solution for dynamic HRC task scheduling.
• The approach enhances reliability and efficiency in industrial collaborative robotics.
• AND/OR graph adaptation is key to completing tasks despite unforeseen failures.