No abstract

No abstract

No abstract

Autonomous underwater vehicles ( A UV's) are poised to become an indispensable tool for ocean exploration, underwater research and development, and military activities.One o,f the key elements f o r practical usage of these vehicles is dependability; there must be some form of fault recoveryLfault tolerance in the control system. I n this work, we use a n approach involving synthesis of motions a n degrees of freedom (DOF) that are unactuated, by combining motions in DOF that are actuated. The objective is to determine the usefulness of the method for a typical streamlined vehicle which is controlled by control surfaces rather than thrusters. W e find that the method has dificulties in its present ,form because 0.f the righting moment designed into the vehicle, but it is .functional o n a modi,fied model 0.f the vehicle. I-IntroductionAutonomous underwater vehicles ( AUV's) are becoming a valuable tool in searching out the mysteries of the ocean. In order to be even more effective in the future, we must have confidence that an AUV can deal with unexpected events, such as failures, in a predictable manner. If possible, we would prefer that the AUV continue its mission, and if not, we at least want the vehicle to make itself recoverable. In either event, there must be some form of fault recovery/fault tolerance in the control system.How does one control a highly non-linear system such as an AUV, that has significant uncertainty in the parameters, and is operating in an unpredictable, nonhomogeneous environment, in such a way as to be able to maintain control even in the event of component failure? Rodriguez and Dobeck [2] use an expert systems approach to keep the vehicle safe. Barnett et al. [l] also use a rule based method for handling faults, but their application is more sophisticated and allows completion of the mission where possible. Leonard [3] [4] synthesizes controls to achieve motion in directions not directly actuated. Payton et al. [6] have a novel approach that *Doug Perrault has been funded by a scholarship from the Natural Sciences and Engineering research Council of Canada.uses what is known to be possible rather than trying to determine the fault.The methodology used in our work is an adaptation of the theory developed in [3], which itself was implemented on a vehicle controlled exclusively by thrusters. The approach entails the use of differential geometry and averaging theory to develop control algorithms involving synthesis of small motions in a desired unactuated direction via open loop control of motions in actuated directions. The control theory is based on a kinematic model of the vehicle, a rigid body with six degrees of freedom.The focus of our work is to implement theoretical control algorithms on a modified dynamic model of a real vehicle: the ARCS vehicle built by International Submarine Engineering of Coquitlam, B. C. The ARCS is a streamlined, propeller-driven vehicle with six control planes to direct its motion through the water. We implement the algorithms and analyze the results t...

Modeling and simulation continues to be an important tool for determining the response of sea-going vessels to wind and waves. To provide appropriate forcing functions to the models, it is important to have environmental data of sufficient fidelity to facilitate an assessment of platform response, which is as accurate as possible within the practical constraints of time and resources. Fortunately, there are a variety of sources of good wave data, including the U.S. National Oceanic and Atmospheric Administration. This study examines the wave data in the context of simulation codes for assessing characteristics of ocean craft response. It also looks at some practical considerations to limit the scope of simulations. The work is strongly influenced by modeling and simulation of naval surface ships, looking for extreme behaviors, but many of the issues discussed are broadly applicable to other applications. 1. Background A variety of tools have been developed over the last few centuries to account for observable physical phenomena in the arts of vessel design, operation, and maintenance. With the advent of modern computing capability, numerical modeling and simulation has provided tools that are ubiquitous and indispensable. In particular, modeling and simulation continues to grow more important for determining the response of sea-going vessels to wind and waves. To provide appropriate forcing functions to the models, it is important to have environmental data of sufficient fidelity to facilitate an assessment of platform response, which is as accurate as possible within the practical constraints of time and resources. Fortunately, there are a variety of sources of good wave data, including the U.S. National Oceanic and Atmospheric Administration (NOAA). This study examines the wave data as input to simulation codes for assessing characteristics of ocean craft response. It also looks at some practical considerations to limit the scope of simulations. Although an attempt is made to be general in the scope of application, the work is strongly influenced by modeling and simulation of naval surface ships, looking for extreme behaviors. However, many of the issues discussed are broadly applicable to other ocean-going platforms.