Professor Axel Gräser University of Bremen Short BIO: Prof. Dr. A. Gräser started his industrial career with Lippke/ Germany, at that time a leading manufacturer of process measure-ment and control systems for the pulp and paper industry. He was head of the control and software department and developed measurement and control algorithms espe-cially for cross profile control. From 1988 until 1994 he was Professor for Automation and Control with the University of Applied Sciences Koblenz. Presently he is head of the Institute of Automation at the University of Bremen. His main research interests are on • robotics and service robotics for the support of disabled and elderly peo-ple. • visual feedback, algorithms for semiautonomous control of mobile robots systems. • Research on Brain Computer Inter-faces for the control of devices by brain activity and application with rehabilitation robots. • feedback method in image process-ing for robustness enhancement. • . Prof. Graeser has been/is coordinator of the EU projects BrainRobot, Brain and Corbys. He has been appointed Honorary Professor at University Timisoara, Romania and For-eign Advisory Professor at the Korean Institute of Science and Technology (KAIST), Deajeon, South Korea. He is member of the steering committee of the International Conference of Rehabilita-tion Robots (ICORR). He is also a member of the ZKW- Zentrum für Kognitionswis-senschaften, Head of Automation in the Bremer Center for Mechatronics (BCM) and a Member of the Board of the Deutsche Forschungsvereinigung für Mess-Regel und Systemtechnik (DFMRS)

Control of rehabilitation robot FRIEND by BCI; Experiences, new developments and future applications

Brain Computer Interfaces are a promising Human Machine Interface for severely disabled people with limited communication capabilities. Before a widespread application of BCI with support systems like service robots become possible some important restrictions of present BCI’s have to be overcome. Limiting factors for all day use of BCI are encountered in the low information transfer rate, in the delay of the BCI signal recognition as well as in the  a limited accuracy of today’s BCI. In the talk  new solutions for multi modal BCI and the development of new adaptive signal processing  methods will be presented.  Experiences in the control of robots with the Bremen BCI will be discussed in detail. Also future research directions and applications will be presented.