A Planning-Based Architecture for a Reconfigurable Manufacturing System

Abstract

The paper describes a novel use of planning in Reconfigurable Manufacturing. Authors considered the nodes of a manufacturing plant as individual AI-based agents able to reason on continuously updated representation of their domain model, plan their own actions, and execute them. The paper aims at clarifying the role of planning, its connection with both a goal selection mechanism, and the agent's knowledge. It describes in detail how a planning system has been customized for the task of planning and execution and shows results of a realistic simulation on a manufacturing plant.

Towards an Effective Framework Combining Planning and Scheduling [Extended Abstract]

“…For our preliminary evaluation of our idea we considered production planning of Reconfigurable Machines that are becoming more widespread (Borgo et al 2016). Production activities require a specific (machine) configuration that can be changed by maintenance activities.…”

“…Note that classical planning domains from the International Planning Competition -Schedule or Woodworking -are examples of such "combined" tasks. Our idea is evaluated on tasks concerning scheduling activities on reconfigurable machines (Borgo et al 2016;Vahedi-Nouri et al 2022).…”

Toward intelligent continuous assistance

“…Such a higher level of problem solving can be achieved by integrating metareasoning mechanisms capable of analyzing and inferring the functional capabilities of external "agentive objects" of the environment. This can be achieved by leveraging formal representation like e.g., an ontology of functions as seen in (Borgo et al 2009(Borgo et al , 2016(Borgo et al , 2019. In this way, an assistive robot can dynamically infer which operation/function can be performed by which agent/object of the assistive context and adapt the (abstract) decision making process accordingly.…”

“…Fig. 6 shows the configuration process pipeline which generates a planning model description by leveraging mechanisms similar to those described in (Borgo et al 2019(Borgo et al , 2016 for a reconfigurable manufacturing system. The Configuration Detection step extracts the configuration of the environment from the KB in order to identify the set and types of sensor and their deployment as well as information about the capabilities of the GiraffPlus robot.…”