SC4: Biomechatronics and Biorobotic
Speaker:
Axel Schneider
Biomechatronics – A Definition:
Mechatronics is an engineering discipline which is concerned with spatial and
functional combination/consolidation of components from fields like Mechanical
Engineering, Electrical Engineering as well as Control Engineering and Computer Science. The above fields can be weighted differently to approach diverse mechatronical tasks. Biomechatronics expands the space of scientific and engineering tools by additional approaches and principles found in nature. Therefore, Biomechatronics builds bridges from technology to Biology, Biomechanics, Medical Sciences etc.
Content:
Most current robots use inelastic joint drives. The advantage of stiff actuation is high position accuracy. In contrast, biological systems – driven by muscles – are always elastic. The neurobiological control concepts which have been evolved together with muscles had to be attuned to muscle properties in evolution. Is it possible to transfer neurobiological approaches and nonlinearly elastic behaviours of muscles to technical actuation systems?
This course introduces how elastic joint actuators can be built and controlled. In
this context, different possibilities to generate elasticity in a joint will be discussed. The alternatives range from real, spring-like elements to virtual elasticity through control and from single to antagonistic actuation. Elastic effects vary from static to dynamically changeable elasticity. A special case is muscle-like elasticity and its biomimetic control. The course introduces underlying biological systems and explains the conceptual transfer. It will be discussed in how far bio-inspired approaches facilitate autonomous behaviour. The course also familiarises the audience with primary actuation systems and shows examples from robotics.
Disciplines:
Biology, robotics, neuroscience, mechatronics, drive technology, biomimetics,
autonomous behaviour
References:
- J. Schmitz, A. Schneider, M. Schilling and H. Cruse (2008), "No need for a
body model: Positive velocity feedback for the control of an 18-DOF robot
walker", Applied Bionics and Biomechanics, Special Issue on Biologically
Inspired Robots (vol. 5, no. 3, pp. 135-147) - A. Schneider, H. Cruse, J. Schmitz (2008), "Winching up heavy loads with a compliant arm: A new local joint controller", Biological Cybernetics
(vol. 98, no. 5, pp. 413-426) - A. Schneider, H. Cruse, J. Schmitz (2006), "Decentralized control of elastic
limbs in closed kinematic chains", The International Journal of Robotics
Research (vol. 25, no. 9, pp. 913-930) - Y. Bar-Cohen, C. Breazeal (2003), “Biologically inspired intelligent robots”,
SPIE press, Bellingham, Washington USA. - A. A. Biewener (2003), “Animal locomotion”, Oxford University Press, New
York USA.
CV:
Axel Schneider studied Electrical Engineering at the University of Applied Sciences in Bielefeld, Applied Computer Science in the Natural Sciences at the Faculty of Technology at Bielefeld University and conducted his PhD project in the Department of Biological Cybernetics, also at Bielefeld University. In his studies, he concentrated on topics covering the development of sensor systems, sensor fusion, mobile robotics and low-level control of drive systems. During his PhD project in the Biological Cybernetics lab of Holk Cruse and Josef Schmitz, he had his first encounter with bioinspired approaches for the decentralised control of legged robots in which the idea of embodiment played a major role. Since 2008 he is head of the junior research group Mechatronics of Biomimetic Actuators in the Faculty of Technology at Bielefeld University and one of the responsible investigators at CITEC, the Centre of Excellence Cognitive Interaction Technology. Here, his main goal is to bring together bio-inspired control aspects and sound technological drive development for robotics and prosthetics. His work also contains bottom-up approaches to facilitate cognitive and intelligent behaviour in robotics.