ÃÛÌÒÖ±²¥

Mr Noel Khan

Job: PhD student

Faculty: Computing, Engineering and Media

School/department: School of Computer Science and Informatics

Address: ÃÛÌÒÖ±²¥, The Gateway, Leicester, LE1 9BH

T: N/A

E: P13167148@my365.dmu.ac.uk

 

Personal profile

Research group affiliations

Publications and outputs

F. Neri and N. Khan, "Two local search components that move along the axes for memetic computing frameworks," 2014 IEEE Symposium on Foundations of Computational Intelligence (FOCI), Orlando, FL, 2014, pp. 62-69
N. Khan, F. Neri and S. Ahmadi, "Adaptive Differential Evolution Applied to Point Matching 2D GIS Data," 2015 IEEE Symposium Series on Computational Intelligence, Cape Town, 2015, pp. 1719-1726
N. Khan, D. Elizondo, B. N. Passow and P. Hardaker, "Pose invariance through registration for hierarchical feature based pattern recognition systems," 2017 International Joint Conference on Neural Networks (IJCNN), Anchorage, AK, 2017, pp. 2902-2910

Research interests/expertise

Autonomy and bio-inspired optimization.

Qualifications

MSc. Intelligent Systems & Robotics, PMP

Conference attendance

2016 Hexagon Geospatial IGNITE Conference: 

PhD project

Title

Automatic Determination of Both Sensorimotor Primitives and Topology for Robots with Unknown or Dynamic Morphology

Abstract

However sophisticated a body, sensorimotor model, and controller, design-time coupling may stifle the development and emergence of general intelligence. Embodiment is required for an agent to acquire and ground knowledge through interactions with the environment. Disembodiment is required to expose those requirements necessary for an agent to become aware of its body, layout, and capabilities so that it can interact with the environment at inception and throughout its life. The ability to accommodate changes to body (better or worse), to manipulate tools, and adapt to a changing environment require greater self-awareness, modeling, and predictive capabilities. Disembodiment requirements include a mechanism for discovering the parts that comprise a body, the body’s topology (which may be disjoint but connected through communication), a mechanism for relating actions to sensory feedback, the ability to exploit sensorimotor relationships for planning, and communicable construct for executing a particular action. This review supports further research by defining key disembodiment capabilities, areas of agreement, or areas that warrant additional study.

Name of supervisor(s)

Dr David Elizondo, Dr Lipika Deka, Dr Miguel Ángel Molina Cabello

noel-khan