Modeling of human-robot collaboration for flexible assembly-a hidden semi-Markov-based simulation approach

Flexible assembly systems are difficult to fully automate, so humans are crucial to add flexibility to the system. Human-robot collaboration (HRC) enables the two parties to achieve a common objective. However, successful implementation of HRC requires addressing safety issues, assigning a clear role to each collaborating agent, and enabling robots to anticipate the actions and intentions of their counterparts. Furthermore, implementing alternatives of HRC in real manufacturing scenarios is risky, expensive, and time-consuming. This study proposes a simulation-based modeling approach for HRC scenarios testing and verification, consisting of the safety, task assignment, and operator risk perception modules. The risk perception module is embedded in a hidden semi-Markov model to enable the robot to act intelligently in an assembly scenario by monitoring the assembly rate of the human operator. The model links fast, medium, and low assembly rates to the operator's perceptions of high, medium, and low risk so that the robot's movements can be adjusted based on its prediction of operator behavior. The simulation model implemented for graphics card assembly shows a reduction in the assembly cycle and idle time of the robot while meeting safety requirements. The proposed simulation-based modeling approach helps improve the existing HRC modeling paradigms through its fast and flexible features.

Automated summary

Flexible assembly systems rely on human-robot collaboration to achieve common objectives. However, implementing this collaboration requires addressing safety concerns, defining clear roles for each agent, and enabling robots to anticipate human actions and intentions. This study proposes a simulation-based modeling approach for testing and verifying HRC scenarios, with modules for safety, task assignment, and operator risk perception. The model includes a hidden semi-Markov model to enable the robot to adjust its movements based on the human operator's assembly rate and risk perception.