Daniel Toal

Neural network control of underwater vehicles

In this paper, recent research efforts in the field of the application of neural networks (NNs) for the control of (semi-)autonomous underwater vehicles are reviewed. Based on a literature review the authors propose a classification of approaches to control underwater vehicles using NNs and the presented articles are categorized according to the identified categories. Based on practical results as described in the discussed literature this paper presents a qualitative assessment regarding the performance of the control strategies. Per category, or control strategy, the major advantages and disadvantages are identified and discussed.

Experimental Investigations of Electromagnetic Wave Propagation in Seawater

It is often preferable to avoid using divers to undertake sub-sea activities, the alternatives being autonomous or remotely operated robotic vehicles and manipulators. This will only be achievable if robust communications can be established through seawater. Presently for such sub-sea activities it is necessary to use acoustic modems, which are only capable of operating with data rates of up to 50kbs-1. Optical sensors can also be used but these rely on clear water and in many sea conditions propagation beyond 10m is not possible. This paper presents new experimental results for electromagnetic wave propagation through seawater at MHz frequencies. These frequencies would enable the use of high speed data rates, suitable for a wide range of sub-sea activities

Precision navigation sensors facilitate full auto pilot control of Smart ROV for ocean energy applications

A Smart Remotely Operated Vehicle, ROV Latis designed as a prototype test bed for operation such as the challenging role of ocean engineering support in wave and tidal energy development is presented in this paper. With state of the art navigation sensors/instruments the vehicle can achieve precision navigation and positioning sub sea and this capability has been utilised within automatic control functionality and autopilot control systems developed and trialed on ROV Latis and not available in commercial ocean ROV technology. This paper describes the vehicles many novel design features: sensor and control systems, autopilot systems, station keeping, fully automatic way point navigation, rapid auto tuning when ROV configuration or payload is changed, fully automatic fault tolerant thruster control with redundancy. The paper also introduces the transparent ocean immersive augmented reality pilot control environment.

Implementation and application of a real-time sidescan sonar simulator

This paper describes the functional theory and design of a modular simulator to generate, in real-time, physically representative spatiotemporal sidescan sonar echo data. The seafloor topography is generated using fractal theory and the resulting terrain is tesselated into triangular facets. Propagation of the acoustic fan beam is determined by a ray theory solution of the governing wave equation while the contribution of each facet to the recorded intensity at the sonar transducer is resolved by applying Jacksons seafloor scattering model. The computational bottleneck inherent in the discovery of illuminated facets at each ping is significantly reduced by the implementation of an optimised mesh refinement scheme intended for interactive rendering of large-scale complex surfaces described by polygonal meshes. The resulting performance increases enabled the successful integration of the sonar simulator with an existing AUV simulator. Synthetic images and the performance measures of the sidescan simulator image generation are presented.

Design of an advanced AUV for deployment close to the seabed and other hazards

High-resolution (acoustic & video) seabed survey requires deployment of sensors on platforms close to the seabed. The unstructured, harsh and hazardous nature of the ocean environment and close proximity to the seabed and other hazards produces serious problems for autonomous underwater vehicle (AUV) during the mission. The Mobile & Marine Robotics Research Group at the University of Limerick is developing a highly manoeuvrable AUV/ROV platform to address the challenges of high-resolution seabed survey in both shallow and deep water. The overall system integrates state-of-the-art survey equipment, a multi-thruster open-frame AUV Tethra and an advanced control system. This paper gives a short overview of the overall system including simulation results.

Neural network augmented identification of underwater vehicle models

In this article the use of neural networks in the identification of models for underwater vehicles is discussed. Rather than using a neural network in parallel with the known model to account for unmodelled phenomena in a model wide fashion, knowledge regarding the various parts of the model is used to apply neural networks for those parts of the model that are most uncertain. As an example, the damping of an underwater vehicle is identified using neural networks. The performance of the neural network based model is demonstrated in simulations using the neural networks in a feed forward controller. The advantages of online learning are shown in case of noise impaired measurements and changing dynamics due to a change in toolskid.

A Flexible Multi-Mode of Operation Survey Platform for Surface and Underwater Operations

Abstract This paper describes a novel vehicle designed for operation flexibility in high-resolution near seabed survey from shallow inshore waters out to the continental shelf edge. The vehicle can be operated in surface tow or as a thrusted pontoon. With buoyancy module release the vehicle becomes neutrally buoyant and is operated as a survey class remotely operated vehicle (ROV) depth rated to 1,000m. Special features of the system include: deployment interoperability for small inshore boats and larger research vessel; fault tolerant thruster control; novel high frequency short range sonar; onboard computer control enabling real-time disturbance reaction; topside augmented reality system support etc.

Automated Optimisation of Simultaneous Multibeam and Sidescan Sonar Seabed Mapping

With the costs involved in offshore seabed mapping being very large and dominated principally by the expense of survey vessel time, any improvement in the planning and execution of seabed mapping, improvement in the quality of the resulting data, and especially reduction in survey costs are of significant interest to the marine and offshore community. This paper presents a terrain adaptive seabed mapping system, which simultaneously operates co-located high-resolution multibeam and sidescan sonar of similar frequency. The automated system is based on the interleaving of sonar transmission-reception cycles to avoid interference. The proposed system optimises the execution of the seabed mapping survey, improves the quality of the data and reduces survey costs.

Assistive tools for system integration, deployment, monitoring, and maintenance of ocean energy devices

This paper describes a set of assistive tools and technologies for system integration, deployment, monitoring, and maintenance of ocean energy devices. The flexible design of these tools enables their use as separate standalone modules, as well as their integration into a unique integrated system. A major component of the system is a smart remotely operated vehicle ROVLATIS — a novel, multi-mode of operation marine robotics vehicle designed for flexibility in near seabed operations from shallow inshore waters out to the continental shelf edge. Ocean energy technologies play an important part in meeting the Irish governments energy strategy, which targets 33 per cent of Irish electricity to be generated from renewable sources by 2020. The assistive tools, proposed in this paper, will help developers of ocean energy devices in meeting this target during different stages of design, deployment, and operation.

A low directivity ultrasonic sensor for collision avoidance and station keeping on inspection-class AUVs

This paper describes an ultrasonic sensor for integration on a UUV. The sensor is characterised by a wide-angle spatial transmission pattern, high range resolution, short minimum detection range and fast response time. For UUV operation in proximity to marine platforms, the seafloor or other hazards, sensors which have ranges below 2m with minimum detection ranges to a few centimetres are needed. Commercial sensors that adequately fulfil this role are as yet not available. The sensor presented has been designed to provide sensor coverage at these close ranges. It can be used to augment data obtained using conventional sonar and vision sensors in target applications such as station keeping, docking and close quarter’s collision avoidance.