Entity

Time filter

Source Type

Girona, Spain

Maurelli F.,Heriot - Watt University | Mallios A.,Office 016 PIV | Krupinski S.,Technopole de Chateau Gombert | Petillot Y.,Heriot - Watt University | Ridao P.,Office 016 PIV
IFAC Proceedings Volumes (IFAC-PapersOnline) | Year: 2010

An Autonomous Underwater Vehicle (AUV) needs to demonstrate a number of capabilities, in order to carry on autonomous missions with success. One of the key areas is autonomous localisation, i.e. the capability of the AUV to estimate correctly its position and orientation in the environment. However, most of the proposed approaches are \passive", with no robot motion control involved. The \active" localisation incorporates the control of the robot motion, as a way to improve the robustness of the AUV localisation, finding the best path to follow in order to reduce the uncertainty in the state estimation. Representing the vehicle's belief of the state with particles, the active localisation module is triggered when there is a clear grouping in the particles and produces in output the path to be followed in order to drop the uncertainty. Both results on simulated data and tank trials have shown the advantages of using this technique compared to the classical ones. Source


Petillot Y.,University of Edinburgh | Maurelli F.,University of Edinburgh | Valeyrie N.,University of Edinburgh | Mallios A.,Office 016 PIV | And 3 more authors.
Proceedings of the Institution of Mechanical Engineers Part M: Journal of Engineering for the Maritime Environment | Year: 2010

This paper presents two acoustic-based techniques to solve the localization problem for an autonomous underwater vehicle (AUV). After a brief description of the Bayesian filtering framework, the paper analyses two of the most common underwater scenarios. Each scenario corresponds to one of the two localization techniques. The scenarios are as follows: map-based localization, when the environment is sufficiently distinctive; transponder-based localization for navigation in large and not distinctive environments. An environment is said to be distinctive when there are possible reference points. For example, a completely flat featureless sea floor is not considered as distinctive. The proposed techniques were validated in simulation. The map-based localization technique was also validated in trials with the autonomous underwater vehicles Ictineu and Nessie IV. The trials with Ictineu took place in an abandoned marina with Ictineu travelling along a trajectory that was more than 600m long. The trials with Nessie IV took place in a rectangular pool and during an AUV competition. © Authors 2010. Source

Discover hidden collaborations