Objective metric study for DOE-based parameter optimization in robotic torque converter assembly

摘要

This paper presents the objective metric study on design of experiments (DOE)-based robotic force control parameter optimization in transmission torque converter assembly. Based on a real-world assembly production process, investigation and analysis are performed on the optimization metrics of assembly cycle time mean (MEAN), its mean plus three times of standard deviation (MEAN+3*STDEV), and first time through (FTT) rate. Simulations have been conducted to illustrate and explain the findings in the parameter optimization practice. Practical metric criteria have been proposed and discussed. An on-pendant robotic assembly parameter optimization tool with the objective metric concept is introduced. And automatic parameter optimization or online robot learning feature is also mentioned in terms of the objective metrics for the particular robot assembly parameter optimization tasks. Finally conclusions are drawn and discussion and further investigation is proposed.