Method of Generating Assembly Action Using Assembly Architecture and Task Skill

Abstract

This paper describes a method of generating an assembly action sequence from the task skill and the assembly architecture obtained from product design. The state space of executing assembly actions are defined in terms of qualitative landmarks which consist for example, of the state of objects and the state of a robot. We will also describe a method of controlling actions (fine motion planning) by monitoring the state of a force sensor and a position sensor, and show a modeling style of hierarchical intelligent control.