A Study on Skill Acquisition based on Environment Information : Task Path Planning for Assembly Task Considering Uncertainty

Abstract

This paper describes a path planning method considering the data of assembly conditions, such as geometrical shape, displacement, uncertainty of the object position and so on. The uncertainty is an important factor for the success probability of the assembly task. Our proposed method represents the position error that comes from precision object shape and displacement error by Gaussian distribution, and estimates the success probability of an insert path by considering the relative position error distribution or the objects. Moreover we propose the uncertainty reduction motion(contact motion), and analyze the uncertainty that is derived after the contact. In this paper, as a simplified path planning of the assembly task, we define 2 strategies ; the insert without contact strategy and the insert with contact strategy and apply the proposed quantification method to determine the optimal path of each strategy for the planar peg-in-hole task.