EPSRC logo

Details of Grant 

EPSRC Reference: GR/J42151/01
Principal Investigator: Denham, Professor M
Other Investigators:
Taylor, Professor J
Researcher Co-Investigators:
Project Partners:
Department: Computing
Organisation: University of Plymouth
Scheme: Standard Research (Pre-FEC)
Starts: 13 February 1994 Ends: 12 August 1997 Value (£): 96,098
EPSRC Research Topic Classifications:
EPSRC Industrial Sector Classifications:
Related Grants:
Panel History:  
Summary on Grant Application Form
1. To investigate a novel biologically-inspired architecture of coupled dynamic neural networks as a learning control system (LCS)2. To develop a mathematical model of the LCS and the neural network theories necessary to analyse its behaviour3. To demonstrate the novel capabilities of the LCS by simulation testing.Progress:Initial work on the project has involved:1. An investigation of the neural resistive grid technique as a basis for motor action planning and generation in the proposed control architecture. This technique has been applied both to the autonomous mobile robot control problem (in an associated University funded research project) and to the pole-cart problem, where it has been compared to the adaptive critic (ASE/ACE) method of Barto et al. These are the two benchmark control problems identified in the grant proposal for testing the novel control system. A paper on these results is in preparation. A paper on the resistive grid planning technique was published recently.2. An investigation of the learning, recognition and retrieval of temporal sensory-action sequences as a mechanism for building a goal-directed internal model of the problem space of the control system, which can be used to associate incoming sensory stimuli (measurements) with appropriate control actions. A paper on this work has been submitted to WCNN95.3. An investigation of the anatomy, function, theories and models of the prefrontal cortex, the area of the brain believed to be involved in goal-directed planning, temporal integration of behaviour, working memory and the integration of information on motivational and emotional significance in purposeful behaviour. The aim of this work is to provide the biological basis for appropriate artificial neural systems capable of displaying analagous behaviour in the proposed control system. An internal report on this work is available and a paper is in preparation. Associated work by PhD students has included the development of a model of declarative memory, in particular the role of the medial temporal lobe of the brain in long-term memory consolidation and an artificial neural network model of this process (thesis due Sept 1995), and an investigation of the reward system in reinforcement-based learning in control, based on a study of the anatomy, function and computational modelling of the amygdala, the part of the brain believed to be involved in assigning emotional significance to sensory stimuli (thesis due Sept 1996).
Key Findings
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Potential use in non-academic contexts
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Description This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Date Materialised
Sectors submitted by the Researcher
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Project URL:  
Further Information:  
Organisation Website: http://www.plym.ac.uk