Current Research Projects

I am involved in a number of current research projects:

• The NERC-funded AFDA project aims to provide automatic on-board fault detection for Autosub 6000, a research AUV operated by the National Oceanography Center. The approach we are taking is a mixture of consistency-based diagnosis using Livingstone 2 and hybrid diagnosis algorithms based on particle filtering.

  Selected Publications:

  • Juhan Ernits, Richard Dearden, Miles Pebody, and James Guggenheim. 2010. Diagnosis of Autosub 6000 using Automatically Generated Software Models. In Proceedings of the 21st International Workshop on the Principles of Diagnosis (DX-10).
  • Richard Dearden. 2010. Bayesian Fault Diagnosis: Common Approaches and Challenges. In Proceedings of the 2nd International Workshop on Cognitive Information Processing, (CIP 2010).
  • Juhan Ernits, Richard Dearden and Miles Pebody. 2010. Automatic Fault Detection and Execution Monitoring for AUV Missions. In the IEEE Ocean Engineering Society Autonomous Underwater Vehicles 2010.
  • Minlue Wang and Richard Dearden. 2009. Detecting and Learning Unknown Fault States in Hybrid Diagnosis. In Proceedings of Diagnosis 2009 (DX-09).
• The EU FP7-funded CogX project is concerned with developing intelligent robotic assistants. The work I am doing in the project is on planning and decision-making, largely applying decision-theoretic planning techniques on a mobile robot to collect information about the world it operates in.

  Selected Publications:

  • Mohan Sridharan, Jeremy Wyatt and Richard Dearden. 2010. Planning to see: A hierarchical approach to planning visual actions on a robot using POMDPs. Artificial Intelligence Journal, 174(11), July 2010. pp. 704-725.
• My work on the EU FP7-funded GeRT project involves learning models of the effects of actions from hand-built robot programs, and then using the models to plan to solve novel tasks. Te project is designed to allow DLR's Justin robot to accomplish novel tasks with familiar objects or familiar tasks with novel objects.
• The EPSRC Bridging the Gap-funded Multiple Particle PEPT project involves tracking radioactive particles as they move around in a liquid to determine properties of the flow of the liquid. We hope to use the method to understand how impurities move in molten metal, among many other applications. The challenge is to move from techniques that can track one or perhaps two particles, to problems with tens of particles.