University of Sussex
Browse

File(s) not publicly available

Neuronal Plasticity and Temporal Adaptivity: GasNet Robot Control Networks

journal contribution
posted on 2023-06-08, 07:51 authored by Tony Smith, Phil HusbandsPhil Husbands, Andy PhilippidesAndy Philippides, Michael O'Shea
Designing controllers for autonomous robots is not an exact science, and there are few guiding principles on what properties of control systems are useful for what kinds of task. In this article we analyze the functional operation of robot controllers developed using evolutionary computation methods, to elucidate the strengths and weaknesses of the underlying control system class. By comparing and contrasting robot controllers based on two different classes of artificial neural network, the GasNet and NoGas networks, we show that the increased evolvability of the GasNet class on a visual shape discrimination task is due to the temporally adaptive nature of the GasNet, where neuronal plasticity mediated through the concentration of virtual neuromodulatory "gases" occurs over a wide range of time courses. We argue that the availability of mechanisms operating over a wide range of potential time courses is a crucial property for controllers used to generate adaptive behavior over time, and that the design process should easily be able to adapt those time courses to the natural time scales in the environment.

History

Publication status

  • Published

Journal

Adaptive Behavior

ISSN

1059-7123

Publisher

SAGE Publications

Issue

3-4

Volume

10

Page range

161-183

Pages

23.0

Department affiliated with

  • Informatics Publications

Notes

Originality: An investigation of time-scales and plasticity in the GasNet: a novel class of artificial neural network incorporating a diffusive modulatory signal. Rigour: Combines mathematical and behavioural analysis of neural networks, an innovative method of analysis in this field. Significance: Highlighted the importance of temporal-scales within ANNs used for robot control. Impact: Was part of the inspiration for a successful EPSRC grant. 12 citations (Google Scholar)

Full text available

  • No

Peer reviewed?

  • Yes

Legacy Posted Date

2012-02-06

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC