Research interests

 

My research deals with the development and analysis of mathematical models for solving problems in Cognitive Science and Autonomous Robotics. The modeling is based, in general, on the theory of dynamical systems. More particularly, I am studying the formation of spatio-temporal patterns in dynamic fields (DFs) formalized by nonlinear integro-differential equations. The challenge is to investigate the parametric conditions, which guarantee the (co-) existence and stability of different types of patterns. Particularly important  for the applications is the analysis of the field dynamics in the presence of external perturbations of stable states.

In close cooperation with colleagues form neurobiology and cognitive psychology I develop dynamic field models to conceptualize and explain their neuronal and behavioral data.  Model predictions are then used as additional constraints for the design of new experiments.

In an analysis by synthesis approach,  the dynamic models are also tested as part of the control architecture of autonomous robots. The ultimate goal of these neuro-cognitive inspired architectures is to endow the robots with some cognitive capacities like decision making, memory, action understanding and anticipation. A concrete example is a DF-based control architecture for more natural human-robot interaction that we have developed in the context of the European project JAST (“Joint Action Science and Technology”). A video of the MInho robot in joint action with a human user can be found at http://cordis.europa.eu/ictresults/.

For a more detailed description of the dynamic field approach and its applications in Cognitive Sciences and Robotics see Erlhagen& Bicho, 2009.

 

Specific goals of the three closely linked lines of my research may be summarized as follows:

 

1)     Mathematical analysis of particular instantiations of DFs (Erlhagen & Horta, 2006)

• Existence and stability of stationary and travelling pulse solutions
• Multi-stability and transitions between different patterns
• Description of shape and temporal evolution as a function of model parameters
• Comparison with conceptually related spatially distributed systems (e.g., reaction-diffusion systems)
• Learning dynamics for establishing a coupling between  DFs

 

2)     Mathematical modeling of neuronal and behavioral data

• Neural position representation of moving objects (Jancke et al, 2004)
• Motion extrapolation and localization errors (Erlhagen & Jancke, 2004, Jancke & Erlhagen, 2009)
• Motor planning (Erlhagen & Schöner, 2002)
• Neural circuits underlying action understanding and imitation (Erlhagen, Mukovskiy & Bicho, 2006)

 

3)     Robot control architectures based on dynamical systems

• Imitation learning (Erlhagen et al., 2006)
• Action understanding, goal inference ( Erlhagen & Bicho, 2006; Erlhagen et al., 2007)
• Cooperation  in multi-agent systems ( Bicho et al, 2004)

 

My research is organized in several projects financed by national and international science foundations. It includes the co-supervision of multidisciplinary Ph.D projects.

 

 

 

Selected publications

(complete publication list)

 

·        D. Jancke and  W. Erlhagen, “Bridging the gap: A model of common neural mechanisms underlying the Fröhlich effect, the flash-lag effect , and the   representational momentum effect”, Chapter in  “Problems of Space and Time in Perception and Action”, R. Nijhawan and B. Khurana eds.,  Cambridge University Press, 2010, pp. 422-440

 

·        W, Erlhagen,  A.  Mukovskiy,  F.  Chersi and E.  Bicho, “On the Development of Intention Understanding for Joint Action Tasks”,  Proceedings of the 6th IEEE  International Conference on Development and Learning, Imperial College London, 11-13 July 2007 (pdf).

·        W, Erlhagen, E. Bicho, “The dynamic neural field approach to cognitive robotics”,  Journal of Neural Engineering 3:R36-R54, 2006 (http://www.iop.org/EJ/journal/JNE, pdf)

·        W. Erlhagen, A. Mukovsky, E. Bicho, “A dynamic model for action understanding and goal-directed imitation”,  Brain Research 1083:174-188, 2006 (http://dx.doi.org/10.1016/j.brainres.2006.01.114, pdf)

·        C. Horta, W. Erlhagen, “Robust persistent activity in neural fields with asymmetric connectivity”, Neurocomputing, 69:1141-1145, 2006 (pdf)

·        W. Erlhagen, A. Mukovskiy, E. Bicho, G. Panin, C. Kiss, A. Knoll, H. van Schie, H. Bekkering, “Goal-directed Imitation in Robots: a Bio-inspired Approach to Action Understanding and Skill learning”, Robotics and  Autonomous Systems 54(5): 353-360, 2006. (pdf)

·        W. Erlhagen, D.  Jancke, “The role of action plans and cognitive factors in motion extrapolation: A modelling study”, Visual Cognition 11: 315-341, 2004. (pdf)

·        D. Jancke, W. Erlhagen, G. Schöner, H. Dinse, “ Shorter latencies for motion trajectories than for flashes in population responses of cat primary visual cortex”, Journal of Physiology 556.3, 971-982, 2004. (pdf)

·        W. Erlhagen, “Internal Models for Visual Perception”, Biological Cybernetics 88:409-417, 2003. (pdf)

·        W. Erlhagen, G.  Schöner, “Dynamic field theory of motor preparation”, Psychological Review, 109:545-572, 2002. (pdf)

·        W. Erlhagen, A. Bastian, D. Jancke, A. Riehle, G.  Schöner, “The distribution of population activation (DPA) as a tool to study interaction and integration in cortical representations”, Journal of Neuroscience Methods, 94:53-66, 1999. (pdf)