Real-time modeling of cross-body flow for torpedo tube recovery of the Phoenix Autonomous Underwater Vehicle (AUV)

Authors
Byrne, Kevin Michael
Advisors
Brutzman, Donald
Second Readers
McGhee, Robert B.
Subjects
Virtual environment
simulation-based design
cross-body flow
autonomous underwater vehicle (AUV)
platform-independent simulation
Date of Issue
1998-03
Date
Publisher
Monterey, California. Naval Postgraduate School
Language
en_US
Abstract
A virtual world provides an exceptional resource for the testing and development of an Autonomous Underwater Vehicle (AUV). The difficulties associated with the underwater environment are numerous and complex. In order to properly verify vehicle results in the laboratory such a world must accurately model the physics associated with the vehicle, its submerged hydrodynamics characteristics, and interactions with the environment. Environmental effects such as wave motion, currents, and flow forces created by bodies moving through the water can cause unpredicted performance variations and failures in the ocean environment. The current Phoenix AUV virtual world includes steady state ocean currents, but does not take into account the environmental effects of waves and flow forces induced by adjacent vehicles (such as a moving submarine docking target). This work provides a thorough real time simulation of these complex factors using physically based models. The problem is broken down into wave motion effects, submarine induced flow fields, and virtual sensors to improve AUV motion control. Simulated testing is performed across a range of easy to worst case scenarios in order to justify assumptions. Extensive testing using virtual sensors is used to develop adequate control algorithms in the presence of turbulent cross body flow. The result of this research is an enhanced virtual world which more accurately depicts the ocean environment, along with the models and control algorithms required to design and operate an AUV during submarine launch and recovery. A platform independent approach to virtual environment simulation is presented through the use of the Virtual Reality Modeling Language (VRML) and Java. Finally, simulation test results provide strong evidence that AUV control with actual cross body flow sensors can enable stable navigation, first through a turbulent flow field and then for subsequent docking with a moving submarine
Type
Thesis
Description
Series/Report No
Department
Computer Science
Organization
Identifiers
NPS Report Number
Sponsors
Funding
Format
xvii, 209 p.: ill.;28 cm.
Citation
Distribution Statement
Approved for public release; distribution is unlimited.
Rights
Collections