Automatic Planning of Manufacturing Processes using Spatial Construction Plan Analysis and Extensible Heuristic Search

Abstract

When automating small-batch manufacturing processes, the time spent for process planning and robot programming becomes more important. This paper proposes an automated process including construction plan analysis, process planning and execution to reduce the amount of manual work required. The process starts by analyzing the structure of the desired product and deriving required process step results, then uses heuristic search to find possible production steps and task assignments, and concludes by simulating or executing the resulting production plan. The approach is evaluated on a case study with a simulated robot automatically building different LEGO R DUPLO R structures starting from a 3D model defining the desired product.

Modular and Domain-guided Multi-robot Planning for Assembly Processes

“…This paper extends previous work of the authors (Nägele et al, 2015;Macho et al, 2016;Nägele et al, 2018) in the field of automatic assembly planning, using concepts from the aforementioned papers while making the previous solutions applicable to more complex assembly problems.…”

“…In recent work, the authors presented their idea of how to describe the spatial model of robots and products by Attributes and how this description can be derived from a construction plan (Nägele et al, 2018). In a nutshell, for the area of assembly these attributes are either a spatial position of one or a spatial relationship between several products and may also contain additional information about a specific type of relationship.…”

“…The authors introduced Analyzer Modules which convert a construction plan to a corresponding goal situation (Nägele et al, 2018). The construction plan is analyzed for contained products, and their related attributes are detected, e. g. by spatial investigation.…”