Key points are not available for this paper at this time.
SUMMARY CAD-directed assembly sequence planning is an essential component of automated robotic assembly task planning. Many research attempts have been made in the past to integrate computer-aided design (CAD) and robot task planning to increase the efficiency and effectiveness of robotic assembly automation. This paper presents a graph-based heuristic approach for automatic generation of assembly sequences from a feature-based data base. A feature-based representation is used to model product assembly. The automatic assembly sequence planning system utilizes four major stages to generate assembly sequences without any user intervention:(1) creates connective graphs based on the product feature representation; (2) decomposes an assembly into sub-groups using the connective graphs; (3) generates the disassembly sequence for each sub-group formed at stage 2; and (4) merges the disassembly sequences of the sub-groups into a complete disassembly sequence, and converts the disassembly sequence into the final assembly sequence. The assembly planning system associated with the feature-based product model has been implemented in Smalltalk-an object-oriented programming language. Several examples are included to illustrate the approach and the heuristic algorithms. The results show that the approach can be used to automatically generate assembly sequences for robot assembly task planning directly from the feature-based database.
Gu et al. (Wed,) studied this question.