In this talk we present geometric methods for robot modelling and control, we use as a mathematical framework the conformal geometric algebra for applications in robot vision, graphics engineering, learning, control and robotics. We will show that this mathematical system keeps our intuitions and insight of the geometry of the problem at hand and it helps us to reduce considerably the computational burden of the problems. Surprisingly as opposite to the standard projective geometry, in conformal geometric algebra we can deal simultaneously with incidence algebra operations (meet and join) and conformal transformations represented effectively using spinors (quaternions, dual quaternions). We present applications as interpolation, modeling and control problems, we reformulate Euler-Lagrange and Newton-Euler Dynamics and recursive Hamiltonians and design using screw theory controllers for robot manipulators. We illustrate these methods in robot vision for robot manipulators and humanoids. For control of manipulators and artificial hands, we have developed the quaternion spike neural network that is used in a decentralized control fashion.

Speaker(s): Bayro Corrochano ,

Location:
Poznan University of Technology
ul. Piotrowo 3A
Poznan, Wielkopolskie
60-965