IEEERSJ International Conference on Intelligent Robots and Systems 1997 Experiments in Forc

A new discrete event controller synthesis methodology for the successful convergence of assembly tasks is presented. Discrete event control has been shown to be a very effective strategy to incorporate both the continuous and discrete natures of an assembl

Experiments in Force Controlled Assembly using a Discrete Event Framework David Austin david@faceng.anu.edu.au Brenan McCarragher brenan@faceng.anu.edu.au Department of Engineering Faculty of Engineering and Information Technology The Australian National University Canberra, Australia Fax: int+ 61 6 249 0506

IEEE/RSJ International Conference on Intelligent Robots and Systems 1997

Abstract

A new discrete event controller synthesis methodology for the successful convergence of assembly tasks is presented. Discrete event control has been shown to be a very e ective strategy to incorporate both the continuous and discrete natures of an assembly task. A discrete event controller synthesis methodology for forcecontrolled systems is highly desirable because force control is naturally compliant and compliance reduces sensitivity to positioning errors. In this paper we present a number of experimental results for the new controller synthesis method, including a complete assembly task. These experiments demonstrate the e ectiveness of the combination of discrete event control and force control for assembly tasks, successfully completing the assembly with positioning errors of up to 50mm and orientation errors of up to 30 .

1 Introduction

Positioning errors have been shown to be the primary cause of failure in robotic assembly 9]. Instead, force control is a natural paradigm for assembly tasks as it is fundamentally compliant, virtually eliminating any dependence on position. However, work in the eld of force control tends to concentrate on low-level details of the assembly problem, and as such has had limited success despite signi cant e ort. In contrast, we propose that work in assembly proceed at a more abstract level. In particular, this paper proposes a new methodology for discrete event controller synthesis for assembly tasks in which the continuous-time system is force controlled. The proposed controller synthesis methodology combines the advantages of discrete event control with the considerable body of work in the area of force control. Corresponding author

Automated assembly is a natural application of robotics and has been studied for many years. In particular, O'Connor et al 9] studied the identi cation and classi cation of errors in automated assembly processes. Assembly is a special type of constrained motion system and a great deal of research has been conducted in the eld of constrained motion systems. Hogan 4] developed impedance control for the manipulation of objects constrained by the environment. Mason 6] and Raibert and Craig 12] developed controllers which use both position and force control for the manipulation of constrained objects. The concept of using position and force control simultaneously has been extended by a number of researchers (e.g. 5] 13]) in search of a better control scheme for constrained motion systems. Unfortunately, assembly is still one of the most error-prone of robotic applications Hybrid dynamic modell

ing is an ideal tool for assembly as it provides a good framework for modelling and analysis of abstract concepts linked to continuous systems. A hybrid dynamic system consists of a discrete event system interacting with a continuous time system. Usually, the discrete event system is a decision-maker or supervisor and operates at an abstract level. A simple example is a furnace system in a typical home. The thermostat is an abstract, discrete event system with two states\too cold" and\warm enough", whereas the house and furnace are continuous time systems. The control algorithm is to turn the furnace on when in the\too cold" state and turn the furnace o otherwise. Even this simple example illustrates an important feature of hybrid dynamic models, which is to combine abstract concepts with continuous systems. There have been many wide-ranging applications of hybrid dynamic modelling, most notably in manufacturing systems and network protocols. To date, they have been used in detailed assembly tasks

第1页下一页