Robotics and Computer Integrated Manufacturing 18 (2002) 403–419 MD-SIR: a methodology for developing sensor— guided industry robots D.I. Kosmopoulos a, *, T.A. Varvarigou a , D.M. Emiris b , A.A. Kostas c a Department of Electrical Engineering, Division of Computer Science, School of Electrical and Computer Engineering, National Technical University of Athens, Athens GR157 73, Zografou, Greece b University of Piraeus, Department of Industrial Management, Deligiorgi 107, GR18534 Piraeus, Greece c University of the Aegean, Informatics and Communication Office, Demokratias Av. 1, GR 85100 Rhodes, Greece Received 1 June 2002; received in revised form 12 August 2002; accepted 25 August 2002 Abstract In this paper, we present a generic methodology for the synthesis of industrial robot applications with sensory feedback at the end-effector level. The presented methodology assumes an open controller architecture and leads to the creation of a library of modular and reusable entities, which can be used to build new systems based on the proposed architectural framework. The library facilitates the integration of new algorithms and it evolves as new modular applications are built. The library components belong to the control objects layer of the open controller architecture and implement functionality for sensor interfacing, sensor modeling, pattern recognition, state estimation and state regulation. The validity of the approach is verified by composing real industrial applications. The experimental results indicate the high quality of the developed systems. r 2002 Elsevier Science Ltd. All rights reserved. Keywords: Open robot controller architecture; Sensor integration; Control objects library; Application development 1. Introduction The current tendencies in the global economy and in the consuming behavior place new challenges in industry, concerning production line flexibility, produc- tion volume, quality and cost. The robots that employ sensory feedback at the end-effector level, are capable of adapting themselves to the environment or to manip- ulator uncertainties and they lend themselves as a feasible option for the fulfillment of those requirements. Despite the undeniable advantages of sensor-guided robots, such installations are still the exception in industry, due to technology constrains and market policies. For many years the sensor technology has posed many restrictions (e.g. high noise, low acquisition speed, lack of simple drivers); the robustness of robot guidance algorithms was in most cases inadequate; the hardware performance was low and the software development methods were immature. In addition to the above, the robot constructors adopted for many years closed proprietary controller architectures, which made prohibitive the integration of external sensory feedback. Recently, the above restrictions begun to be raised. The hardware platform of personal computers offers now big processing and networking capabilities, ease of use and maintenance and high communication capabil- ities at low cost. The modeling languages (e.g. UML), the integrated programming environments, the standard libraries (e.g. STL) and the open integration platforms (e.g. CORBA [6]) facilitate the development of robotic applications. Moreover, the sensors technology has abandoned many restrictions of the past and sensor employment is now possible with a relatively simple environment configuration [15]. Apart from the above, lately many robot constructors adopted a more open architecture for their controllers. Thus, it becomes obvious that the present political and technical environ- ment favors the development of a software methodology for sensor integration in robotic systems. The need for *Corresponding author. E-mail addresses: dkosmo@telecom.ntua.gr (D.I. Kosmopoulos), dora@telecom.ntua.gr (T.A. Varvarigou), emiris@unipi.gr (D.M. Emiris), akostas@aegean.rhodes.gr (A.A. Kostas). 0736-5845/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII:S0736-5845(02)00031-5