200 Int. J. Mechatronics and Automation, Vol. 4, No. 3, 2014 Copyright © 2014 Inderscience Enterprises Ltd. Developing adaptable joint admittance control in a virtual robotic testbed Eric Guiffo Kaigom* and Jürgen Roßmann Institute for Man-Machine Interaction, RWTH-Aachen University, Ahornstrasse 55, Aachen, Germany E-mail: kaigom@mmi.rwth-aachen.de E-mail: rossmann@mmi.rwth-aachen.de *Corresponding author Abstract: Over the past years, eRobotics has been proving itself to be a flexible and efficient approach to tackle highly complex issues in research and application areas of robotics and automation, including planetary exploration, orbital servicing, sustainable forest management and industrial manufacturing. One of the main objectives of eRobotics is to bring robotics know-how into the fields of virtual reality and simulation on the basis of a holistic simulation approach that swiftly adapts to new demands and lays the foundation for the development of new technologies. Following this line of ideas, we introduce in this paper a novel systematical and highly modular approach to endow an open-chain robot manipulator with adaptable joint admittance control on a robot simulator. A three-stage admittance, position and torque control is developed to provide a desired joint admittance when external torques are sensed. The joint admittance is adapted by inferring the human intension to accelerate and decelerate during interaction in an intuitive and simple fashion. Experiments carried out on a physical 7 DoF KUKA LWR 4+ reveal that the proposed approach substantially reduces the robot resistance felt by the operator, therefore enhancing considerably the ease of manipulation. Keywords: eRobotics; simulation; virtual testbed; robotics; admittance control. Reference to this paper should be made as follows: Kaigom, E.G., and Roßmann, J. (2014) ‘Developing adaptable joint admittance control in a virtual robotic testbed’, Int. J. Mechatronics and Automation, Vol. 4, No. 3, pp.200–211. Biographical notes: Eric Guiffo Kaigom studied Electrical Engineering at the RWTH Aachen University in Germany. Since 2010, he has been working as a Research Scientist towards his PhD at the Institute for Man-Machine Interaction at RWTH Aachen University, Germany. His research interests include: space-, industrial and human-centred robotics, as well as system simulation techniques. Jürgen Roßmann studied Electrical Engineering at the Universities of Dortmund and Bochum, Germany. After his studies, he worked as Researcher and Team Leader at the Institute of Robotics Research (IRF) in Dortmund. He received his Doctorate in 1993 and was appointed Visiting Assistant Professor for Robotics and Computer Graphics at the University of Southern California in 1998. He received his Habilitation degree in 2002 from the University of Dortmund and Managing Director of EFR-Systems GmbH in Dortmund from 2005 to 2006. Since 2006, he has been the Director of the Institute for Man-Machine Interaction and Full Professor at the RWTH Aachen University in Aachen, Germany. His research interests are projective virtual reality, multi-agent control and supervision, multi-sensor integration, system simulation and optimisation techniques, computer vision, real-time visualisation and man-machine interaction. This paper is a revised and expanded version of a paper entitled ‘A new eRobotics approach: simulation of adaptable joint admittance control’ presented at the Mechatronics and Automation (ICMA), 2013 IEEE International Conference on, Takamatsu, 4–7 August 2013. 1 Introduction 3D simulation technology is widely used in manufacturing today. Particularly in robot-assisted manufacturing, it is usually the method of choice to quickly pre-configure a work cell or to plan tasks. It has been adopted as a strategic approach to improve production processes by identifying inefficiencies and predicting the side effects of making changes (Holst, 2001). Nowadays, simulation-driven engineering is tightly incorporated in product lifecycle management and robotised automation in enterprises to substantially increase throughput, maximise benefits and stay competitive (Meike and Ribickis, 2011; Johari et al., 2007).