Acquisition of Sensor-Motor Competences in Mobile Robots


We investigate mechanisms that allow robots to acquire competences through interaction with the environment, in other words: to learn from experience.

A number of different mobile robots have been used for this research, and the robots successfully acquire competences such as obstacle avoidance, wall following, corridor following, dead end escape, box pushing, or phototaxis.


Relevant Publications

Learning in Multi-Robot Scenarios

  1. Ulrich Nehmzow, GA Learning in Multi-Robot Scenarios: The PEGA Algorithm Department of Computer Science, University of Essex. Technical Report Nr. CSM-367.

  2. Ulrich Nehmzow, Learning in Multi-Robot Scenarios through Physically Embedded Genetic Algorithms Proceedings Simulation of Adaptive Behavior, Edinburgh 2002. MIT Press.

Action Selection and Action Planning

  1. John Pisokas and Ulrich Nehmzow, Performance Comparison of Three Subsymbolic Action Planners for Mobile Robots, Int. J. Robotics and Autonomous Systems, 2005.

  2. John Pisokas and Ulrich Nehmzow, Experiments in Subsymbolic Action Planning with Mobile Robots, Proc. AISB convention 2003, Aberystwyth.

Robot Training through System Identification

  1. O. Akanyeti, T. Kyriacou, U. Nehmzow, R. Iglesias and S.A. Billings, Visual Task Identification and Characterisation Using Polynomial Models, accepted for publication in Int.J. Robotics and Autonomous Systems, 2007.

  2. Roberto Iglesias, Theocharis Kyriacou, Ulrich Nehmzow and Steve Billings, Task Identification and Characterisation in Mobile Robotics through Non-Linear Modelling, Int.J. Robotics and Autonomous Systems 55 (2007), 267-275.

  3. Ulrich Nehmzow, Roberto Iglesias, Theocharis Kyriacou and Steve Billings, Robot Learning Through Task Identification, Int.J. Robotics and Autonomous Systems, Vol 54, Issue 9, pp.766-778, Sep 06.

  4. Theocharis Kyriacou, Ulrich Nehmzow, Roberto Iglesias and Steve Billings, Task Characterisation and Cross-Platform Programming Through System Identification, Int.J. Advanced Robotic Systems, Vol 2, pp. 317-324, December 2005.

  5. Roberto Iglesias, Ulrich Nehmzow, Theocharis Kyriacou, Steve Billings, Training and Analysis of Mobile Robot Behaviour Through System Identification, in: Current Topics in Artificial Intelligence, LNAI 4177. Roque Marín, Eva Onaindía, Alberto Bugarin and José Santos (Eds.), ISBN: 3-540-45914-6. pp 470-479, Springer 2006.

  6. Ulrich Nehmzow, Otar Akanyeti, Cristoph Weinrich, Theocharis Kyriacou and Steve Billings, Learning by Observation through System Identification, in Proc. "Towards Autonomous Robotic Systems, Taros 2007

  7. O. Akanyeti , U. Nehmzow , C. Weinrich , T. Kyriacou and S.A. Billings, Programming Mobile Robots by Demonstration through System Identification, Proceedings European Conference on Mobile Robotics (ECMR), 2007.

  8. Roberto Iglesias, Theocharis Kyriacou, Ulrich Nehmzow and Steve Billings, Route training in mobile robotics through system identification, Proc. ICCI 2006, Enformatika Journal, Volume 15,October 2006, ISBN/ISSN 1305-5313.

  9. Roberto Iglesias, Theocharis Kyriacou, Ulrich Nehmzow and Steve Billings, Robot Programming Through a Combination of Manual Training and System Identification, Proc. European Conference on Mobile Robotics, Ancona 2005. ISBN 88-89177-187

  10. Theocharis Kyriacou, Ulrich Nehmzow, Roberto Iglesias and Steve Billings, Cross-Platform Programming Through System Identification, Proc. “Towards Autonomous Robotic Systems”, TAROS, London 2005. Shortlisted for Springer Best Paper prize. ISBN 0-905247-03-5




Complete list of publications


http://cswww.essex.ac.uk/staff/udfn/compacq.html
Last modified 26.6.2007