Automatic Design of Controllers for Miniature Vehicles through Automatic Modelling

Created by W.Langdon from gp-bibliography.bib Revision:1.3872

@PhdThesis{DeNardi2010PhD,
  author =       "Renzo {De Nardi}",
  title =        "Automatic Design of Controllers for Miniature Vehicles
                 through Automatic Modelling",
  school =       "School of Computer Science and Electronic Engineering,
                 University Of Essex",
  year =         "2010",
  address =      "UK",
  month =        sep,
  keywords =     "genetic algorithms, genetic programming, evolution,
                 quadrotor, helicopter",
  URL =          "http://www.cs.ucl.ac.uk/staff/R.DeNardi/DeNardi2010PhD.pdf",
  URL =          "http://discovery.ucl.ac.uk/1330831/",
  size =         "433 pages",
  abstract =     "This thesis investigates the problem of automatically
                 designing controllers for vehicles that can be
                 represented as a rigid body. The approach is based on
                 the idea of automatically obtaining a dynamic model of
                 the system of interest, and using it to design
                 controllers automatically. A novel aspect of our
                 approach is that of not requiring any form of platform
                 specific knowledge, and being as a consequence both
                 hands-off and very generic.

                 The acquisition of models is based on data logged when
                 a human pilot was controlling the vehicle, and is
                 carried out by an evolutionary algorithm based on
                 competitive coevolution. Models in the form of symbolic
                 expressions are coevolved along with the portions of
                 the training data that are used to compute their
                 fitness. This results in an effective and
                 computationally efficient way of constructing
                 models.

                 The modelling method is applied to a small toy car, a
                 full sized aeroplane and two different types of small
                 quadrotor helicopters. For comparison, models of the
                 same vehicles are also derived using standard modelling
                 techniques that exploit platform knowledge. The models
                 produced by our technique are shown to be as accurate
                 or better than those produced manually. Importantly
                 after a limited amount of rearrangement of terms, the
                 models also prove to be interpretable.

                 A method is presented for reproducing in the models the
                 noise and uncertainties that characterise real world
                 platforms. The evolved deterministic models produced
                 are augmented with a simple yet computationally
                 efficient Gaussian noise model, and a principled method
                 based on unscented Kalman filtering is used to estimate
                 the noise parameters. The augmented models are
                 demonstrated to reproduce most of the variability shown
                 by real vehicles.

                 The automatic design of controllers considers both
                 monolithic and modular structures based on recurrent
                 neural networks. Conventional steady state evolution is
                 used to evolve monolithic controllers, and cooperative
                 coevolution is applied to modular controllers. Manually
                 designed controllers are also developed for purposes of
                 comparison. Controllers are mainly evolved for
                 path-following tasks, but other tasks like imitating
                 game players' abilities are also considered.

                 In general monolithic controllers are shown to be very
                 effective in controlling the toy car, but have
                 limitations when applied to the helicopters. Modular
                 networks show a better ability to scale to more
                 demanding platforms, and in simulation reach levels of
                 performance comparable to or better than controllers
                 designed manually.

                 Tests show that for both the toy car and quadrotor
                 helicopters, the evolved controllers successfully
                 transfer to the real vehicles, although a certain
                 amount of mismatch exists between the performances
                 predicted in simulation and those on the real
                 platforms.",
  notes =        "examiners were Professor Alan Winfield (UWE) and
                 Professor Huosheng Hu.
                 http://www.essex.ac.uk/csee/department/news/newsletter/09_08_10.aspx",
  bibsource =    "OAI-PMH server at discovery.ucl.ac.uk",
  contributor =  "O. E. Holland",
  oai =          "oai:eprints.ucl.ac.uk.OAI2:1330831",
}

Genetic Programming entries for Renzo De Nardi

Citations