Design specification of parts dimensional tolerance for robotic assembly

Abstract Success of robotic assembly can be limited by parts dimensional tolerances as well as robot repeatability. This paper examines the effects of these limitations on the success (or failure) of parts assembly. A Positioning Design Factor for Assembly ( pdfa ) is defined to represent the percentage of all cases where the combination of parts dimensional tolerances and parts positioning errors result in positive effective clearance, which is one of the conditions leading to successful assembly. Mathematical derivations have been applied to the classical peg-in-hole problem. The dimensional tolerances of the peg and hole as well as their relative position are considered to be randomly distributed variables. Robot repeatability was also modelled as a random variable within the robot specification. A Monte Carlo analysis was used to predict the probability of favourable positioning conditions for successful insertions. The simulation also incorporated several common assembly situations such as the effect of chamfers on the mating parts and the ansi Standard fit classes. The results have been condensed in this paper into a family of charts which relate the Positioning Design Factor for Assembly to parts size and different parts fit classes. A design chart which gives maximum allowable positioning error for the best pdfa versus parts size is presented. This chart can assist the designer in specifying and selecting both robots repeatability limits and parts dimensional tolerances for assembly applications.