A neural-network-based computational scheme for generating optimized robotic assembly sequences

Abstract This paper presents a neural-network-based computational scheme to generate optimized robotic assembly sequences for an assembly product. An assembly sequence is considered to be optimal when it meets a number of conditions: it must satisfy assembly constraints and minimize assembly cost. To generate such an optimal sequence, this paper proposes a scheme that utilizes both a neural network with functional link nets, and an expert system. Based upon the assembly constraints inferred and the assembly costs obtained from the expert system, the evolution equations of the network are derived, and an optimal assembly sequence is finally obtained from the evolution of the network. To illustrate the suitability of the proposed scheme, case studies are presented for industrial products such as an electrical relay and an automobile alternator. The performance of the neural network is analyzed by comparing the results with those of the expert system alone.