"With the help of National Instruments hardware and software, the MMRRC has successfully developed a set of multi-purpose technologies for system integration, testing, and implementation of advanced control algorithms for unmanned underwater vehicles (UUVs)."
- Edin Omerdic, University of Limerick
The Challenge:
Creating a complete submersible remotely operated vehicle (ROV) capable of supporting the core survey suite including side scan, Inertial Navigation System (INS), Doppler Velocity Log (DVL) and pressure (depth) while accommodating subsea and wide-area scanning activities, as well as containing autotuning features and the ability to detect and isolate thruster faults while working under automation using way-point navigation.
The Solution:
Using an NI CompactRIO real-time controller with a variety of digital and analogue modules and the NI LabVIEW development suite to develop a set of multipurpose technologies for system integration, testing and implementation of advanced control algorithms for unmanned underwater vehicles (UUVs).
Author(s):
Edin Omerdic - University of Limerick
We used rapid control prototyping and hardware-in-the-loop (HIL) techniques, implemented these technologies during design and development stages of the first Irish-made smart remotely operated vehicle (ROVLATIS) and validated the system performance in simulation and real-world environments. LabVIEW increased our productivity by orders of magnitude. Built-in libraries and drivers for data acquisition, simulation, data analysis and presentation allowed us to focus on solving a real problem without worrying about details such as pointers and memory allocation. Programs that could have taken months to write using conventional languages were developed in hours and days. HIL setup with CompactRIO enabled easy integration with real-world signals. The quality, stability and performance of our final applications were aided by the efficiency of the LabVIEW development environment. Also, NI hardware and software played a crucial role in the design and development.
System Architecture
The complete ROV system is comprised of two core surface-side components supplying voltage and control communications to the ROV, and wet-side components including five wet bottles for housing different aspects of the system, eight thrusters, four cameras, six lights, multibeam, side scan, sound velocity probe, obstacle avoidance sensors, INS and a set of aiding sensors. We used each wet bottle differently within the ROV and the thruster bottle to control the onboard thrusters for the ROV from the control bottle, which holds all the control-related equipment to process navigation data. In addition, the umbilical bottle houses the signal and power junction units. We used CompactRIO as a real-time controller and I/O interface with thrusters, lights and leak detectors.
The power bottle houses the DC power supply units, the Ethernet-to-fibre-optic converters and the RS232/485 converters, which control the pan and tilt of the onboard cameras. The survey bottle contains a set of aiding sensors including DVL, depth sensor, ultra-short baseline (USBL) transponder and local GPS receiver for surface operations.
Mission Support Tools
We used a set of multipurpose platform technologies (MPPT) for subsea operations, including survey equipment integration, efficient planning and mission simulation, ROV pilot training, ROV fault-tolerant control, enhanced in-mission survey execution and offline analysis and replay of acquired data.
The MPPT ring represents the dual character of platform technologies: the inner and outer rings can be independently rotated and expanded, indicating that any technology or module can be transparently interchanged between the simulated and real-world environment. This duality of operation facilitates the application of modern control, modelling and simulation tools in marine technology development. It provides a framework for researchers to develop, implement and test advanced control algorithms in a simulated virtual environment under conditions very similar to the real-world environment.
Real-time models of ocean dynamics are very useful for the design, development, testing and validation of marine technology in simulation. To accommodate these models for real-time simulation of ocean dynamics, we have to achieve balance between complexity of the model and constraints imposed by the synchronisation requirements in real time. Models of waves and ocean currents implemented in LabVIEW as part of the simulation environment inside the MPPT ring are simple enough to fulfil real-time requirements and general enough to describe most phenomena of interest.
The main objective of the Mission Builder module is to transform the mission objective by gathering pilot inputs and measuring navigation data into the desired ROV trajectory to formulate the trajectory planning problem. Finally, the output control cluster is optionally blended with the Obstacle Avoidance Control Cluster to create the Winner Control Cluster, the ultimate “boss” with exclusive rights to control the actuators.
Testing and Deployment
We conducted offshore trials off the western portion of Ireland’s Connemara coastline in February 2009. The vehicle was mobilised using the ship RV Celtic Explorer. The cruise consisted of a day in Galway port integrating and testing ROV and ship systems and six days at sea. During the trials, we proved all of the ROV systems, demonstrated sensor interoperability and performed comprehensive vehicle diagnostics. In addition, we performed system identification on the ROV and successfully carried out vehicle controller tuning. We also conducted a series of preplanned survey missions.
These missions were used to trial the operation of the vehicle and the MPPT ring. The ROVs augmented reality topside control and visualisation, the Adaptive Multisonar Controller and the use of vision systems for near seabed navigation.
We implemented LabVIEW as the front panel display of the application running on CompactRIO. The augmented reality display and visualisation of thruster saturation bounds are provided by a LabVIEW application running on the visualisation PC.
We officially launched ROVLATIS in July 2009 and mobilised the vehicle using the Special Service Workboat Shannon I. We demonstrated excellent control performance on a surface and submerged through a series of complex manoeuvres, including full six-degree-of-freedom control in thruster fault-free and faulty cases, accurate speed and course following with independent control of attitude and heading, autotuning of heave and heading low-level controllers, accurate dynamic positioning and a host of other control sequences.
We successfully designed ROVLATIS to serve as the host platform for proving new technologies developed in the Mobile and Marine Robotics Research Centre. Using NI hardware and software, we successfully developed a set of multipurpose technologies for system integration, testing and implementation of advanced control algorithms for UUVs. These technologies have been implemented during design and development stages of our novel survey platform. We chose LabVIEW as the development tool for the versatile and intuitive GUI, easy learning, flexibility, stability, easy connection with real-world signals and other applications, easy data sharing and portability. Shared variable engine, field-programmable gate array (FPGA) support, simulation module and support for automatic multicore parallel loops execution were the most important features that assisted in developing an innovative and stable control platform for our ROV.
Author Information:
Edin Omerdic
University of Limerick
Ireland
Tel: 061 330316
edin.omerdic@ul.ie
Print
PDF
