Research Area "Neurocybernetics"

 
Home
Research topics
Projects
Tools
Publications
People
Links
Contact

Doctorate Programme
Institute of Cognitive Science
University of Osnabrück

Project: Bipedal Walker*

Applying an evolutionary algorithm, first the morphology of a simulated passive dynamic bipedal walking device without knees, able to walk down a shallow slope, is developed. Second, using such an evolved solution and adding minimal motor and sensory equipment to this morphology, a neural controller is evolved, enabling the walking device to walk on a flat surface with minimal energy consumption. The applied evolutionary algorithm fixes neither the size nor the structure of the neural controllers. Especially, it is able to generate recurrent networks small enough to be any analyzed with respect to their behaviour relevant inner dynamics. Such a controller realizes minimal energy consumption by using the dynamics of the morphology for most of the walking time.

For the development of the simulation and the optimization with an evolutionary algorithm our ISEE environment is used.

For future work the morphology of a walking device with knees is evolved and an appropriate neural network has to be evolved.

Videoclips and pictures

a straight legged passive dynamic walker (2.9 MB mpg-clip).
a passive dynamic walker with knees (2.0 MB mpg-clip).
the straight legged walking device with a neural controller (3.7 MB mpg-clip).


Contact: Frank Pasemann and Steffen Wischmann

* Work done at the former Intelligent Dynamics (INDY) group at the Fraunhofer Institute AIS (now FhI-IAIS).