Failure recovery planning in assembly based on acquired experience: learning by analogy

Abstract

For complex tasks in flexible manufacturing as well as service applications, robots need to reason about the tasks and the environment in order to make decisions. This paper presents a method for recovering from execution failures based on analogies with previous failure recovery episodes. The basic principles that explain the success of a failure recovery strategy are extracted based on several deductive as well as inductive transformations. In recovery planning based on these learned principles, the inverse transformations are applied.