Alergic Double-bill -- Wed 7th Nov, 4:30pm in ARUN-401
Two presentations from ECAL 2007 in Lisbon, brought all the way to Sussex.
Each about 20/30 mins plus time for questions.
==============================
(A)
New Models for Old Questions: Evolutionary Robotics and the 'A not B' error.
Rachel Wood, Ezequiel Di Paolo
Abstract. In psychology the 'A not B' error, whereby infants perse-
verate in reaching to the location where a toy was previously hidden
after it has been moved to a new location, has been the subject of fifty
years research since it was first identified by Piaget [1]. This paper de-
scribes a novel implementation of the 'A not B' error paradigm which
is used to test the notion that minimal systems evolutionary robotics
modelling can be used to explore developmental process and to generate
new hypotheses for test in natural experimental populations. The model
demonstrates that agents controlled by plastic continuous time recurrent
neural networks can perform the 'A not B' task and that homeostatic
mediation of plasticity can produce perseverative error patterns similar
to those observed in human infants. In addition, the model shows a
developmental trend for the production of perseverative errors to reduce
during development.
==============================
(B)
Evolution of Neural Networks for Active Control of Tethered Airfoils
Allister Furey, Inman Harvey
Abstract. Recent development in tethered airfoil i.e. kite technology allows the
possibility of exploitation of wind energy at higher altitudes than achievable
with traditional wind turbines, with greater efficiency and reduced costs. This
study describes the use of evolutionary robotics techniques to build
neurocontrollers that maximize energy recoverable from wind by kite control
systems in simulation. From initially randomized starting conditions,
neurocontrollers rapidly develop under evolutionary pressure to fly the kite in
figure eight trajectories that have previously been shown to be an optimal path
for power generation. Advantages of this approach are discussed and data is
presented which demonstrates the robustness of trajectory control to
environmental perturbation.
==============================