Dynamic Task Allocation for Cooperating, Heterogeneous Assembly Resources in LMAS

Abstract

AbstractFlexibility and changeability are key enablers for assembly systems to meet the challenges of progressing from mass to individualized production. The new paradigm of Line-less Mobile Assembly Systems (LMAS) provides the necessary changeability, by mobilizing multipurpose resources and realizing adaptive system behavior. A holistic allocation of tasks to resources is introduced by extending the allocation through a study of its spatial feasibility.To evaluate the state of the art and to assess the transferability of theoretical task allocation approaches to industrial assembly, key criteria are identified. The presented approach for dynamic task allocation for cooperating, heterogeneous assembly resources expands existing concepts by the consideration of spatial motion constraints and cooperative behavior. Through the identification of cooperatively solvable tasks, the solution space for allocation is expanded, and a reduction of non-value-adding waiting times is expected. The potential of reconfigurable resources is leveraged through incremental re-auctioning of tasks, enabling dynamic reaction to changes, such as process delays, altered task prioritization or downtimes of resources. Interdependencies between task allocation, trajectory and formation planning are included through verification of spatial feasibility. The verification contains the confirmation of an executable formation of allocated resources as well as collision-free trajectories to reach the planned formation.KeywordsLine-less Mobile Assembly Systems (LMAS)Multi-robot task allocationFormation planningMobile robots