Optimization of Multi-Agent Workflow for Human-Robot Collaboration in Assembly Manufacturing

Human-robot collaboration presents an opportunity to improve the efficiency of manufacturing and assembly processes, particularly for aerospace manufacturing where tight integration and variability in the build process make physical isolation of robotic-only work challenging. In this paper, we develop a robotic scheduling and control capability that adapts to the changing preferences of a human co-worker or supervisor while providing strong guarantees for synchronization and timing of activities. This formulation is then expanded to optimize the workflow of a team of robots according to a set of qualitative and quantitative spatial and temporal constraints and performance objectives. We describe the Adaptive Preferences Algorithm that computes the optimal flexible scheduling policy for task completion that meets hard temporal constraints. We use APA within a mixed integer multi-agent optimization algorithm that assigns a flexible schedule of agents to tasks. We show that execution of the Advanced Preferences Algorithm is fast, robust, and adaptable to changing preferences for workflow and that the multi-agent optimization, while slower, is practically useful for important applications in multi-robot assembly of large structures for aerospace manufacturing. We specifically demonstrate the capability for quick reoptimization of a plan in response to temporal disturbances in the schedule and changing high-level guidance from a human supervisor.