Take our PhD
survey for the
chance of
winning a
£50* Amazon
Voucher! Take Survey
* or equivalent in Euros or US Dollars
×
University of Warwick Featured PhD Courses

University of York Featured PhD Courses
University of Leeds Featured PhD Courses
University College London Featured PhD Courses
Study Options Featured PhD Courses

PhD Research Project

This project is no longer listed in the FindAPhD database
and may not be available.


Persistent Autonomy for Autonomous Underwater Vehicles

Institution:
PhD Supervisor:
Co-Supervisor:
Application Deadline:
No more applications being accepted
Funding Availability:
Funded PhD Project (Students Worldwide)

Ocean Systems Laboratory, School of Engineering and Physical Science, Heriot-Watt University, Edinburgh, Scotland, UK
http://www.oceansystemslab-heriotwatt.com/
http://persistentautonomy.com/

A PhD position is available immediately on a research programme aiming to enhance the cognition of autonomous underwater vehicles (AUVs).

The overall research theme is Persistent Autonomy – developing the underlying computational mechanisms that enable AUVs to achieve mission goals in unforeseen situations by adapting their actions based on observation and description of their environment, their condition and their own performance in executing tasks.

Applications are for inspection, cleaning and intervention tasks in the oilfield using hover capable AUVs, and for underwater archaeological survey with multiple vehicles.

Three technical areas are being emphasised, complementing a portfolio of work already underway. The PhD will be drawn from these areas according to the interest and aptitude of the successful applicant.

1. Matching and adapting semantic descriptions of the surrounding environment in ontologies using feature based localisation (SLAM) from forward looking sonar and video imagery. Initially, SLAM methods are used to localise a vehicle relative to a structure. However, beyond this we wish to match observed features to those expected based on a known geometry and associated semantic description. The objective is to first identify where the known geometry is incorrect, and subsequently consider building semantic descriptions (ontologies) of unknown objects. The semantic descriptions are then used as the basis for planning or plan repair to meet mission goals.

2. Detection and diagnosis of failure in task execution based on available sensor data and on the state of action execution in adaptive controllers and skill learning modules. Simple actions, such as move or hover, output from a mission planner will in general succeed at execution. However, more complex actions such as grasping and turning a valve or cleaning with a water jet may fail. The robot must detect this failure based on available information, where possible diagnose the cause, and then replan a further attempt. Key technology developments are required for these detection and diagnosis phases.

3. Optimising information exchange services between collaborating AUVs engaged in a joint mapping exercise using low bandwidth and intermittent acoustic communication. Vehicles should share mapping and other mission information as they acoustically and optically survey the seabed, automatically detecting objects of interest. Data exchange takes place within a service oriented architecture. Methods are required to ensure only timely information is exchanged and relayed appropriately, bearing in mind mission goals and the context of the data.

Applications are sought from interested individuals with MSc or 1st/2.1 honours degree (or equivalent) in computer science, signal processing or robotics, and with practical aptitudes/interests in real autonomous underwater vehicles and sensors. Field trials in Scotland and Europe are involved. The successful candidate will join an existing team of experienced researchers.

The position is of three years duration, starting immediately, with all fees covered and an annual stipend of up to £15k.

Funding Notes:


Open to international applicants, subject to visa

References:


David M. Lane, Francesco Maurelli, Tom Larkworthy, Darwin Caldwell, Joaquim Salvi, Maria Fox, Konsta
"PANDORA: Persistent Autonomy through Learning, Adaptation, Observation and Re-planning" Proceedings of IFAC NGCUV'12

C.C. Sotzing, D.M.Lane “Improving the Co-ordination Efficiency of Limited Communication Multi-AUV Operations Using a Multi- Agent Architecture” Journal of Field Robotics Vol 27 No 4 July 2010 pp412-429

K. Hamilton, D.M. Lane, K Brown, N. Taylor “An Integrated Diagnostic Architecture for Autonomous Underwater Vehicles” Journal of Field Robotics. Vol. 24, No 6, pp 497-526 (2007)

I. Tena Ruiz, S. de Raucourt, Y. Petillot and D. M. Lane “Concurrent Mapping and Localisation Using Side-Scan Sonar” IEEE Journal Oceanic Engineering, Vol. 29 (2): 442-456 (2004,)






Find A PhD

Copyright ©2011
All rights reserved

The Science Registry Ltd, Sellers Wheel, 151 Arundel Street, Sheffield, S1 2NU, United Kingdom. Tel +44 (0) 114 268 4940 Fax: +44 (0) 114 268 5766