ROVs and Undersea Cable Maintenance
   The number of Remotely Operated Vehicles (ROVs), as well as their potential users, have increased over the years since they were first conceptualized. As the number of installed undersea cables continue to increase, the potential for external aggression relating to these cables increases. Today, it is essential that each cable through its associated maintenance agreement have access to a remotely operated vehicle to ensure repairs on buried sections of cables. With SCARAB vehicles, it is possible to bury and rebury (post repair) these cables to minimize future faults and service disruption.

The Development of ROVs and Training Experts in the Field of Underwater Robotics in Far-Eastern State Technical University
   Since 1970 till the present day more than dozen of modifications of tethered, towed remotely operated vehicles and theirs combinations have been created in FESTU and in the “Dalnee” Designing Bureau (department of FEPI) as well. The successful marine and oceanic tests of some experimental models made it possible to carry on research and other works down to abyssal depth. The staff of the Institute of Marine Technology Problems (IPMT), Far-Eastern Branch of the Russian Academy of Sciences (FEB of RAS) has been engaged in the development of AUVs. The main features of the designed vehicles and construction principles of control systems and trends of their future development are considered in the report. The new ROV and AUV designing approaches of automatic control motion systems are given as a result. The interactive and intelligent control systems of IIIrd- generations robots are of the special interest. These robots are in a stage of mathematical modeling and half-natural tests. To create these systems following methods of automatic control have been used: theory of optimum systems, principles of fuzzy logic, identification methods of mathematical model of underwater robot dynamics as object of control, theory of systems with variable structures and theory of self-adapting systems. These methods were used to develop systems of supervisory control with typical commands that has allowed to considerably unload the operator. The use of a voice in dialogue systems permits to release hands of the operator for manipulators control and other underwater vehicles systems. Underwater robots applications are particularly important in finding and inspecting objects of possible ecological catastrophes such as sunken submarines with a nuclear reactor; nuclear, chemical and other weapon buried on the sea bottom; industrial waste, oil and gas fields' concentration, etc. The introduction problems of research and experimental design efforts in the process of experts (engineers and masters) training in the field of underwater robotics in FESTU are discussed in this report. The problems of Pacific Rim Countries cooperation in the field of underwater robotics and education programs development are offered to discussion.

A New Mini Modular Plastic AUV
   The reduction in size, cost, and power of the computers and electronics used in AUV sensing, navigation, and control systems has enabled the development of small sized vehicles with greatly reduced operational and development costs. This has resulted in many real world successful applications of AUVs as oceanographic sensor platforms. Despite all of the recent successes and advancements, however, UUVs remain for the most part customized, low volume, research experiments. AUVs are still very complex systems that are more expensive, difficult to operate, harder to maintain and are less reliable than other ocean sensor platforms such as towed systems. The relatively high cost of AUV systems means that the risk of loss is a significant factor governing operational procedures. Moreover, the software needed to control an autonomous vehicle in the hazardous ocean environment is typically quite complex and requires experts to operate and maintain. All of these factors provide a disincentive for mass production of UUVs and thereby keep the volume low and hence the costs high.
   Obviously, the key to more rapid proliferation of AUV technology and applications is to increase their cost effectiveness, reliability, and simplicity of operation. Two approaches may be used to address this problem. One is purpose built AUVs. In this approach a AUV is designed to execute a specific limited mission or narrowly constrained set of missions. The vehicle is then optimized for that mission including all the software and support systems. Only by limited the operational scenarios can the vehicle operation be simplified enough that expert operators would not be required. Moreover if enough units are produced the cost could be decreased such that the risk of loss is minimized thereby making the vehicles somewhat expendable. The major impediment to this approach is that a large advance investment is required to developed a low cost purpose built vehicle. This is antithetical to the way most small UUV technology is currently developed, that is, by university or oceanographic research institutions who must constantly push the envelope for UUV missions and applications.
   Another approach to increasing the cost effectiveness of AUV systems is to build modular multi-purpose vehicles. The key here is select the right levels and types of modularity to gain equivalent economies of scale but over a range of missions. That way the development resources of a number of groups and applications can be leveraged to gain economies of scale. This is the approach that the Advanced Marine Systems Lab at Florida Atlantic University has been pursuing over the past few years. The technical challenge in a multi-purpose vehicle is to minimize the overhead and down time associated with reconfiguring the vehicle for different missions. In addition, the vehicle must be easily upgradable to increase its life cycle so that the development costs can be amortized over a longer period.
   Traditionally there have been two main structural configurations used for small vehicles. One is a flooded fairing with internal pressure vessels and wet cabling between pressure vessels. This is the approach used by the Odyssey, Ocean Voyager, and Ocean Explorer. This has the advantage that the cabling layout is relatively unconstrained. Components can be inserted or removed with some rewiring of connections to the main control and power systems. Mounting brackets and internal foam must be redone for each configuration. Moreover sensors that need access to sea water have a variety of placement options. Lots of small inexpensive pressure vessels (with thin walls) are used and are easier to manufacture in house. The disadvantage is that there tends to be unusable space and extra weight for pressure vessels and wet cabling. In addition tight packing of components usually results in mechanical stress on the wet connectors and cabling resulting in failures and maintenance problems.
   The second structural configuration for small vehicles is to use the pressure vessels as the hull. This approach allows only one configuration a cylindrical "torpedo" shape vehicle. This is the approach uses by WHOI's "REMUS" and Penn. States ARL's small AUV. This has the advantage that there is less wasted space and less weight. Internal cabling is for the most part dry. For small or very small vehicles the hull pressure vessels are small diameter (< 10 inches.) and can be easily manufactured in house. AUVs require some wet sensors and therefore some of the sections on this type of vehicle must be flooded as well. Typically in the nose and tail. Cabling between wet and dry sections is through end cap mounted bulkhead penetrators. This type of vehicle is difficult to reconfigure since any communication between one part of the vehicle and another must pass through the intervening sections. Changes require rewiring. Typically spare wires and bulkhead connectors are built in to accommodate future additions but this still requires rewiring and is only reconfigurable to a point. Most importantly the order the sections are connected in is fixed they cannot be rearranged with out redoing all the bulkheads.
   FAU has developed a new third "hybrid" configuration that combines the advantages of the first two approaches. This is the approach used for our next generation mini AUV. In this approach a molded plastic hull pressure vessel is used as in the second approach described previously. However, the plastic pressure vessel has exterior faired channels for cables with bulkhead penetrators going through the sides of the pressure vessels and not the endcaps. Relative to cabling, this allows almost any section to be interchanged with any other section and home runs are supported. This minimizes the amount of wasted flooded space while providing protected cable ways and much more flexibility. The disadvantage is a rounded rectangular exterior cross section with some reduction in stability for yaw and pitch and more material is needed in the walls of the pressure vessel. The lightweight glass fiber reinforced resin used for the pressure vessel provides the needed strength without incurring to great a weight penalty.
   This paper describes the design and development of FAU's latest generation AUV.

The Development of Underwater Dredging System for Rocky Sea Bottom at Narrow Channel
    Kanmon channel located between Honshu and Kyusyu is a main steamship lane of Japan. On March 1991 the dredging plan to the depth of ń14m is decided but the part of Hayatomo-Strait is restricted by various conditions. Some of these conditions are as follows
   *the highest current comes to 9 knots
   *the width of the strait is 450m
   *700 vessels sails the strait per day.
    At present, dredging to the depth of ń13m, to the distance of 100m from the side end of the channel, we use a grab dredger equipped with spads unless using anchor lines for mooring. But we must restrict the navigation intolerably by the present method at the dredging of the central part of the channel. Therefore we are developing the underwater dredging system. As the sea bed soil is the weathered rock, we chose a load header for the excavating machine. The main subjects of this development are as follows:
   *the excavating machine design especially for its low height
   *the electric cable system for power supply to the excavating machine
   *the gathering and transporting system of excavated materials
   *he control system to operate the excavating machine on the condition of invisible underwater area.
    We are practicing some experiments to design the trial machine system from 1996 after two years of the examinations.

Software Design Techniques for the Man-Machine Interface to Autonomous Underwater Vehicles (AUV)
    This paper describes the management system used to the AUV integrated simulator, which we are developing. This management system operates using the SGI IRIX 6.2 operating system, X Windows system, IRIS Performer and OpenInventor graphical user interface. It runs on a SGI Indigo2 workstation connected via Ethernet and other function computers, and manages the simulation system while communicating with each other.
    The management system consists of four components as follows: the scene display window, the camera window, the global map window, and the information display window. The scene display window describes the
   simulation scene. The camera window displays the photos taken by the camera and transfers them to the computer that deals with them.
    This paper describes the predominant functions of the simulation management system, including:
   1. Network management. UNIX Socket based on TCP/IP protocol is used to transfer data. The system manages and coordinates the network.
   2. Simulation scene management. X Window, IRIS Performer and OpenInventor are used to describe the virtual world scene, AUV motion and coordination, operating, etc.
   3. Mission and path planning management.
   4. Data management. The system records the simulation data while running and analyses it after the
    simulation. It can replay the simulation using the data at any time.

Current Progress in the Development of a Solar Powered Autonomous Underwater Vehicle (AUV)
   For the past two years The Institute for Marine Technology Problems (IMTP) RAS, FEB and the Autonomous Undersea Systems Institute (AUSI) have undertaken a joint program to develop a solar powered Autonomous Underwater Vehicle. This paper will describe this effort, the vehicle design, and the results of this effort to date.
   The vehicle system relies on the availability of solar energy and its performance depends on the effective utilization of that energy. Various strategies and algorithms to manage the acquired energy were investigated in detail. The results of this analysis and the design of an energy system testbed will be described.
   The vehicle design is greatly impacted due to the necessity of integrating the solar array with the vehicle platform. This results in an unconventional vehicle shape that brings about some unique dynamics problems. These will be discussed along with the characteristics of the onboard control system.
   The prototype vehicle will be described and the results of in water field tests of the prototype system will be reported.

The Theseus Autonomous Underwater Vehicle - Two Successful Missions
   Over the past five years, International Submarine Engineering Research and the Esquimalt Defence Research Detachment of the Defence Research Establishment Atlantic have worked together to develop a large autonomous underwater vehicle, named Theseus, for laying optical fiber cables in ice-covered waters. In trials and missions conducted in 1996, this vehicle showed impressive capabilities. It was able to lay a fiber-optic cable in a completely autonomous mode for a distance of 200km under Arctic sea-ice and then return to the launch station for recovery. It demonstrated a navigational error of less than 0.5% of the distance traveled, and cross track error was reducible to 0.05%. It operates in either depth-keeping mode or bottom-following mode, was designed to operate at a maximum depth of 1000m, and has operated at a depths of up to 425m.
   The vehicle is equipped with an inertial navigation unit, a Doppler sonar speed sensor for autonomous navigation, a forward-looking obstacle avoidance sonar, an acoustic homing system and acoustic transponders for use with surface tracking stations, and an optical telemetry system is used for system monitoring while Theseus is laying optical fiber cable. All subsystems are controlled by an M68030 bases sensor integration and control computer.
   The vehicle is currently configured for cable laying. Other missions could be accommodated with minor changes to the payload section of the vehicle. Theseus' qualities of covertness, long endurance and precise navigation make possible such tasks as long baseline oceanographic data collection, remote route surveys, remote mine hunting, the rapid deployment of acoustic and non-acoustic surveillance systems, and even the towing of mobile sensor arrays. Longer missions could be accommodated by replacing the current silver-zinc batteries with fuel cells or higher energy/density batteries.
   The paper describes the vehicle and presents the results of the evaluation trials and its first cable laying mission.

System Design and Quasi-Sliding Mode Control of an AUV for Ocean Research and Monitoring

An Autonomous Underwater Vehicle as a Student Project
   Most autonomous underwater vehicles that are so far operational are extremely complex and costly systems. This project, carried out by five undergraduate students, takes a systems engineering approach to attempt to design and build a low-cost AUV starting from a functional requirement, by defining what the vehicle does, and a performance requirement, by defining how well it carries out the function. The requirements include the following: operation underwater to a depth of 150 metres, visible on the surface to a range of 500 metres in clear visibility, able to identify its location relative to a datum point to within 2 metres on any orthogonal axis, capable of detecting obstacles within five metres range, capable of aborting a mission should a function failure occur, power source lasting a minimum of two hours, data storage for two hours for post-mission diagnostics, manoeuvrable in pitch, roll and yaw axes to within 5%, capable of executing a defined mission profile, transportable in a small boat, moveable by two people, accessible to a computer via a data socket, and able to accommodate various sensors according to the mission objectives. The starting point was the acquisition of a hull having built-in thrusters. The rest of the system was designed from scratch by a team with no previous experience of AUVs.

On the Prediction of Maneuverability of the Autonomous Underwater Vehicle, AUV-HM1
   The maneuverability of an unmanned untethered submersible AUV-HM1 testbed developed at National Taiwan University has been investigated through two sets of stability indices, one is evaluated by empirically estimated values of stability derivatives, the other are those obtained by a series of captive model tests. The results presented in the previous papers have shown that the present configuration AUV-HM1 is stable in longitudinal mode, and unstable in lateral mode. In the present paper, a higher order panel together with vortex lattice method is developed and applied to the calculation of stability derivatives of AUV-HM1 testbed. Those results as applying this exact body shape close-fit method to the calculation would be discussed. With its efficiency and accurate enough for practical use, it would also act as a tool at the stage of preliminary design.

A Discussion on Distributed Control of Autonomous Underwater Vehicles
   Multiple autonomous underwater vehicles (AUVs) can be effective force multipliers for Navy missions and scientific experiments involving search and detection in large and/or cluttered environments. The application of multiple, low-cost vehicles can provide rapid, repeatable synoptic surveys with pay-offs in reduced run time, mixed sensor suite, finer data resolution and increased reliability.
   This paper will discuss the research problem of distributed control of multiple autonomous systems. Primary goals of this research initiative is to resolve the group dynamics and control necessary for successfully using "many" swimming and/or 'bottom crawling' vehicles. Critical issues are : control of spacing and dynamics of the group as a whole (i.e. formation), group communication, and group configuration (e.g. division of labor).
   A powerful and yet simple to implement control strategy for multiple vehicles is a random search strategy. There is little to no dependency on neighbors in determining next position. Given enough time an area can be completely covered much in the way a gas will fill a volume. The autonomous vehicles in this strategy use random changes in heading and random number of steps forward. This allows the vehicles to wander in and out of an area. The most pressing issues related to using random search strategy are the efficiency and focusing of a random search. Therefore and attempt is made to improve the efficiency of this simple search scheme. In the approach outline in this paper random search is enhanced by allowing exchanges of information between autonomous vehicles that will impact and improve the "next move" decision logic of the individual vehicle. This established a minimal level of connectivity between group members. This connectivity or "Encounter" is defined by two members coming in contact, recognizing each other and establishing a communication link long enough to exchange a pre-determined packet of information. Once the information is transmitted the connectivity is terminated. Various encounter strategies will be discussed with comparison to a purely random strategy. The encounter strategies are derived from observed social behaviors in animal groups during foraging expeditions.

A Plan of Utilization of AUVs in Shallow Water
    Recently many AUVs are planned or developed for various practical purposes, such as ocean water investigation, ocean ridge exploration, seabottom cable maintenance, military use, and so on. Several years ago, most of AUVs are just the test models, not for the practical use. This change has been occured by the recent progress of individual technology for AUV.
    The AUVs are aiming at the use in deep water. However, in another area where the "conventional" AUV is not expected, for example, in shallow water such as port area, the use of AUV can be considered. The inspection works in port are mostly carried out by divers. The problem is that the efficiency of human work is worse as the area where the port is constructed is getting deeper.
    The AUV for port area has several disadvantages. One is the affection of waves.Another one is that underwater structures are complicated in port area. The other one is water is usualy turbid in port area.
    On the other hand, the AUV for port area has several advantages. One is that ultrasonic information link is possible because the working distance of AUV is not so long. Another one is that radio wave transmission is possible when the AUV goes up on the surface. The other one is power source problem is not critical because the length of one mission is not so long. The objectives of AUVs of port area are mentioned below.
   1.Underneace piled pier inspection AUV The underneace of piled pier inspection is carried out by human work with a small boat. As it is narrow space and there are many piles, it is troublesome work. This AUV is going through the piles and observes the underneace condition from the surface of the water.
   2.Port facilities maintenance information AUV In a port, there are many underwater structures, like quay, sea walls, breakwaters and so on. This AUV is going around those facilities periodically and compare the inspected data with former data. If large change is detected, the AUV informs it to the port maintenance office.
   3.Hazard investigation AUV This AUV is usualy stored rigid and safe place to typhoon or earthquake. After the hazard, for example, in the case of typhoon, the wave height becomes smaller than certain value, AUV will start investigation work automatically. The damage of underwater structures is inspected, and using that information the recovery plan can be made quickly.
    An omnidirectional AUV suited for the use in the port area has been designed and experimental model is under construction. It has 4 vertical thrusters and 4 horizontal thrusters in 4 rectangle directions. At first stage, it will be controled through tethered cable. We made 6 year research plan using this model. And we have an idea of new control method of AUV among multi obstacles. We expect AUV wil be proved to be useful even in shallow water through our research work.

Autonomous Vehicle Obstacle Detection and Tracking
    Almost all autonomous underwater vehicles require some manner of acoustic obstacle detection and avoidance mechanism to prevent vehicle damage due to collisions with uncharted obstacles. Modern forward looking multibeam sonars use wide bandwidth pulses, fast sampling rates, and high bit count samples to form simultaneous beams having wide dynamic range and good target to clutter ratios. As a result, even moderate resolution sonars can generate sonar image data at rates of 10 Mbits per second or higher. This relatively large quantity of sonar data must be reduced to some sort of predicted physical environment ahead of the vehicle as part of the obstacle avoidance process. A data compression technique optimized for autonomous vehicles is described. A short video example of the algorithm working in real time on typical obstacle avoidance data is presented.

Towards Environmental Monitoring with the Autosub Autonomous Underwater Vehicle
    Oceanographers world-wide face the challenge of gathering biological and physical data from the ocean environment on the similar space and time scales. Researchers are beginning to learn that the coupling of physical and biological is of prime importance to the health of the ocean. They look to innovative and cost-effective solutions from developments in marine technology. One such development is the Autosub autonomous underwater vehicle project funded by the UK Natural Environment Research Council.
    The first vehicle, Autosub-1, built in 1995 has completed over 85 missions to data, encompassing engineering evaluation and scientific research. A cubic meter payload volume, or 100kg wet weight, allows an extensive sensor suite to be carried. Sensors flown to date include triplicate CDTs (from different manufacturers), two fluorometers, a 300kHz ADCP (for current profiles and acoustic estimates of zooplankton abundance and behaviour) and an experimental pH probe. These have been in addition to navigation sensors for position, depth, speed and altitude. Within the 7m long, 0.9m diameter vehicle sufficient energy can be carried in inexpensive lead-acid batteries to cover mission length of up to 70km at a speed of 7km/hr. Alternatively, by using low-cost manganese alkaline primary cells, ranges of up to 250km can be readily achieved at energy costs of about $15 per kilometre.
   In this paper we will outline the technology incorporated into Autosub-1 to achieve the performance required by ocean scientists, including : an advanced thick-section glass fibre reinforced plastic main pressure vessel, proof tested to 750m, whose high buoyancy to weight ratio ensures a high payload capacity; a novel, high efficiency direct drive propulsion motor that eliminates the need for a shaft seal; a distributed network architecture for all vehicle system. including mission management and an unique (D)GPS receiver abot to cope with interruptions due to wave washover. The paper will highlight how operational experience with the prototype vehicle has allowed to improve the engineering aspects. Missions that have already been completed will be used to illustrate various modes of operation of Autosub-1 including terrain following, leaving harbour autonomously, hydrographic and bio-physical profiling and contributions to studies of upper ocean process. The paper will conclude with a look to the future for large AUVs such as Autosub-1 and how they might cross the barrier between research community prototypes and operational commercially available vehicles.

Intelligent Control System for 6000m Deep Towed System
    As important tolls in the exploitation of inner spaces, various kinds of remotely controlled vehicles (ROVs) played an important tool in oceanographic research, offshore oil and gas exploration and undersea mining. The undersea mining, in particular, requires 6000m deep towed submersible to perform underwater observation, inspection and operation. Consequently, research on the movement control of such 6000m deep towed system is of significance. Since there are highly uncertain natures in the oceanic environment, the system would be subjected to stochastic disturbance and would be on non-linear and time-varying nature. It would be difficult to establish proper mathematical model without known hydrodynamic coefficients. Traditional control techniques, such as feedback controller, self-adaptive controller and variable structure controller, could solve such kind of problems to a certain extent. But they are depending on properly established mathematical model, they cannot meet the high performance requirements of calculating speed and accuracy. Based on the control theory and artificial neural network, an intelligent control technique is developed. The intelligent control technique is a mean of solving the complex non-linear problem and bypassing the difficulties and drawbacks of traditional control techniques.
    In this paper, intelligent control is employed to tackle the movement control problems of 6000m deep towed system. First, the dynamic response of the 6000m deep towed system is analyzed in detail. Then, an intelligent control system using the adaptive neural network is proposed and the learning-adaptive control algorithm is given. Finally, the simulation results in 6000m towed system are discussed. Compared with the traditional PID controller, the intelligent control system gives significant robustness to disturbances and uncertainties.

Results of the U. S. Naval Oceanographic Office's 1997 AUV Assessment Survey
    The U.S. Navy has been developing Autonomous Underwater Vehicles (AUVs) for operational scenarios for many years. The maturity of this technology has reached a point where it is both feasible and economical to transition AUVs to the collection of environmental data for the Navy. With this in mind, the Naval Oceanographic Office )(NAVOCEANO) has established an AUV program with the goal of matching different AUV capabilities to meet specific Navy requirements.
    During September 1997, NAVOCEANO conducted an assessment of AUVs developed and operated by the Woods Hole oceanographic Institute, the Massachusetts Institute of technology, and the Florida Atlantic University. Operating onboard the Texas A&M research vessel Grye, these teams collected a wide variety of environmental data parameters in coastal waters off the Mississippi Gulf Coast.
    This Paper looks at the AUV technology and the data collected by these institutions in comparison with data collected by traditional hydrographic survey techniques. The conclusion of this paper is that each AUV has characteristics that provide excellent quality survey data. The differences in capabilities )size, weight, mission duration, etc.) point out that no vehicle can be expected to meet all survey requirements. The best solution is a suite of vehicles, using common processing techniques, targeted to specific operational scenarios.

Design of a Semi-Autonomous Underwater Vehicle for Intervention Missions (SAUVIM)
   As the research in the autonomous underwater vehicle (AUV) field intensifies and the necessity of underwater robotic vehicles (URVs) increases, the requirements of an URV have expanded from fly-bymissions to more complex intervention missions. The AutonomousSystems Laboratory (ASL), Department of Mechanical Engineering,University of Hawaii is in the midst of designing and developinga Semi-Autonomous Underwater Vehicle for Intervention Missions (SAUVIM).The proposed AUV will possess a fully functional manipulator and allows human-intervention from a land-based computer system capableof vehicle path-planning and vehicle monitoring.

Autonomous Target Tracking by Underwater Robots Based on Vision
   Not fixed

Experience of the Use of Manned Submersibles in P. P. Shirshov Institute of Oceanology RAS
    P.P. Shirshov Institute of Oceanology RAS has one of the largest underwater scientific fleet: six manned submersibles with different operating depth. There are the submersibles "Angus" (600m) and lock-out type vehicle "Osmotr" (300m) - built by the engineers of P.P. Shirshov Institute and completed in 1975 and 1981 correspondingly: "Pisces-VII" and "Pisces-XI" (2000m) - built in Canada in 1974-76; "MIR-1" and "MIR-2" - built in Finland on joint Russian-Finnish technical design project in 1985-87.
    During twenty years over 100 expeditions with these vehicles were conducted to different sites of World ocean and inland waters of Russia, over 1500 dives of the submersibles were done.
    Big value of scientific data was obtained during the expeditions: special underwater operations on different objects on ocean floor were provided: deep ocean images for the films of IMAX "Titanica" and "Lightstorm entertainment inc." "Titanic" were done.
    During the expeditions new technologies and methods of the operations with manned submersibles were implemented.
    Short introductions of the experience of the operations with mentioned above vehicles are presented in this paper.

Experimental Verification for the Control System of Intelligent Underwater Vehicle
    This paper introduces an intelligent underwater vehicle, and the architecture of the control system and especially the mission execution system.
    The control system has eleven modules. They are divided into three layers, three columns, an overall database, and a man-machine interface. The three layers are the mission layer, the behavior layer and the execution layer respectively. The three columns are the sensing-fusion-understanding column, the thinking-prediction-dispatching column and the planning-control column respectively.
    The mission execution system which is in the execution layer uses artificial neural networks to filter sensors data, estimate the positions, azimuth and velocities of the underwater vehicle, and control its movements.
    The abilities on autonomous navigation and obstacle avoidance of this system are proved by the results of a series of tests in a lake. Especially, the system described in this paper has very good performance to the position and heading control of underwater vehicle.

On the Limit Cycle of the Underwater Vehicle Control System
   The underwater vehicle control systems usually experience limit cycles due to inherent nonlinear properties. Two major nonlinearities existing in such a system are the fluid drag and the thruster dynamics. However, most studies investigating control problems of the underwater vehicle system by neglecting influence of the thruster dynamics. Although some researches did put the nonlinear thruster behavior into consideration for controller design, effect of the thruster on the control system and characteristics of the limit cycle have not been well learned. Hence quantitative solution for the limit cycle of the underwater vehicle control system using the describing function method is presented in this paper. A case study will also be exhibited by simulation results.
   A typical thrust is composed of a propeller and a shroud. The thruster shroud has an enclosed volume and a cross sectional area. Based on energy balance in the controlled volume enclosed by the shroud, the dynamic equation of the thruster can be obtained and the thruster force generated by the propeller is equal to square of its angular velocity. The squared angular velocity contributes the thruster nonlinearity. A complete dynamic
   model of the underwater vehicle therefore should include two substantial nonlinearities. One is the drag force induced by the relative motion of the vehicle in the fluid and the other is the propulsion force of the thruster. In order to simplify the analysis work but without loss of generality, the vehicle is piloted by a typically proportional controller.
   The basic describing function technique cannot easily solve the limit cycle problem for a system with multiple nonlinearities. Therefore, a method for stability analysis of nonlinear multivariable systems is adopted to locate the limit cycle of the underwater vehicle control system. The characteristic equation can be partitioned into real and imaginary parts, after those two nonlinearities are replaced by their describing functions. As a result, both describing functions and the frequency are determined by incorporating an additional equation, which establishes the relationship between these two nonlinearities. Stability of the limit cycle can be verified by using the perturbation method. It can also be found that the oscillatory frequency of the limit cycle is bounded by a certain value.
   Although the limit cycle for the underwater vehicle control system can be predicted by employing the above approach, two nonlinear functions in the control loop increase difficulty of controller design. If one of them
   plays a less significant role and can be simplified or linearized, the succeeding design task will become much easier. Therefore, the limit cycle calculated from the above method is compared with the approximate solutions by linearizing one of those two nonlinearities in a case study.
   This investigation does not intend to produce a general conclusion for all underwater vehicles. Nevertheless, it provides a practical example for dynamic analysis of the underwater vehicle control system.

Motion Control and Obstacle Avoidance of Hyper-Redundant Manipulators
   To carry out inspection in clustered environments as wrecks or heat exchange pipelines used in power plants, we may mount a survey video camera at the tip of a flexible pole and "steer" it to the location of interests. An alternative is to replace the flexible pole with several short rigid bars that are jointed together with actuators. The linkage can be used as a platform to deliver cameras or tools to the service area. In robotic terms, this apparatus is called "hyper-redundant manipulator" for its abundant degree of freedom or "snake robot" for its motion. Typically, this type of manipulator has tens or more links, joints and actuators. In addition to its end-effector, the robot could also form a loop with it body to manipulate objects just as a snake does to its prey. Therefore different concerns are taken for hyper-redundant manipulators in design, control and application than that of its counter part used in assembly lines.
   Prior to applying hyper-redundant manipulators in real world situations, we have to dissolve some dilemmas coming along with hyper-redundancy. (1) To enhance the dexterity of the manipulator, we need to increase the number of joints/actuators. At the same time, rigidity of the structure decreases and inertia of the system increases. (2) Extra degrees of freedom bring complexity into control Computationally efficient algorithms must be developed. (3) If no priori information of the environment available, how a real-time map can be constructed with sensory data to assist navigation or manipulation.
   In order to study the feasibility of adapting hyper-redundant manipulators in inspection work, we develop a real-time simulator on a Pentium-166 personal computer. Different layouts of clustered environments are set to test hyper-redundant manipulators with different number of links. Preliminary results show that our motion control algorithm and map construction scheme work well for a 15-link planar manipulator operating in a work space with five obstacles. With this simulator as a testbed, we hope to know more about design criteria for building such a hyper-redundant manipulator.

A Rapid Reconfiguration Strategy for UUV Control
    Fault accommodation is a key aspect of UUV (Underwater Unmanned Vehicle) control, and control reconfiguration is an economic and effective way to realize it. However, in all of current methods, it must be after faults have been located accurately that the controller is reconfigured, this sometimes leads to inadmissible deviation from normal, especially when the UUV is executing a task requiring high control precision and working continuity. This paper deals with the above problem.
    First, a rapid reconfiguring strategy is proposed. Without the necessity to know what are fault accurately, in this strategy the controller is reconfigured in the first time so long as the faults are located into a suitable element set (called currently-suspected fault elements set, CSFES), and is reconfigured repeatedly whenever the diagnosis process advances and the CSFES changes. Thus, deviation from normal is prevented soon after the diagnosis process begins, and the performance of the fault UUV is recovered in stages with the diagnosis process advancing.
    The strategy is transformed into a robust model-following problem, and a mixed H2/Hinfinity method is provided to solve it, which optimized the H2 index under some Hinfinity constrains to guarantee that the closed-loop system is robust to the possible faults provided by the diagnosis process, and has consistent tynamic performance at all stages.
    Last, the above strategy and design method are applied to some UUV model, both the design and the simulation results show their effectiveness.

Control Performance of Fish Robot with Pectoral Fins in Horizontal Plane
    The existing underwater robots, of which motion is controlled by thrusters and wings, are poor not only in hovering and turning both in the vertical plane and in the horizontal plane over seabed terrain, but also in dexterous manipulation in the floating condition of underwater robots[1]. The mechanism of locomotion of the aquatic animals can provide us with new sights on maneuverability and stabilization of underwater robots. This paper focuses on biomimesis on maneuver performance of aquatic animals to create a new device for maneuvering of underwater robots.
    One of the authors developed an apparatus of pectoral fin motion for maneuver of underwater robots based on observation and experimental analysis of pectoral fin motion of Black Bass. The work revealed (1) that the combination of feathering and lead-lag motion of a pair of pectoral fins on both sides of fish dominantly generates the fish motions of forward swimming, backward swimming, hovering and turning, (2) that the apparatus making the feathering motion and the lead-lag motion of the pectoral fin generates thrust force in a certain range of phase difference between moth motions., and (3) that the fish robot consisting of fish rigid body and a pair of the apparatus on both sides of the body can perform forward swimming, backward swimming, and turning.
    This paper focuses on control performance of the fish robot in horizontal plane in terms of (1) rendezvous and docking with an underwater post in water currents, (2) hovering in water currents to seabed and (3) forward and backward swimming by use of flexible fins. The first issue is discussed to investigate the control preciseness that is of critical importance in docking of underwater robots with underwater station used as long-term underwater observatory. The second issue is a key performance in maneuver of working underwater robots over seabed. The third issue is investigated from the viewpoint of higher efficiency of the fin and increase of forward and backward swimming speed as a fundamental performance of working underwater robots. The use of fuzzy control enables the fish robot perform precise control of its attitude and position.

Intellectual Distribution Nets for Long-Term Deep Ocean Monitoring
    Major projects, as the World Ocean Circulation Experiment (WOSE) and Global Ocean Observation System (GOOS), are designed to provide data for global models, which will describe the state of the World Ocean and predict its future development to the limits of predictability. In these programs a prominent role in solving the problems of operative monitoring and of long-term routine collection of physical, chemical, biological and geophysical data is assigned to autonomous devices, in particular to unmanned systems, including acoustic remote sensing, autonomous unmanned vehicles and drifting and devices.
    It is apparent, that the wide use of autonomous devices puts enhanced demands as to their performance reliability under the conditions of a highly aggressive medium and to low energy consumption. Other crucial factors are the cost of a device, the simplicity of its repeated manufacture and the exploitation convenience.
    The experts of Institute of Oceanology have worked out a concept of construction of long-term monitoring systems for ecologically dangerous ocean region. The system is based on a new work of multi purpose autonomous bottom stations equip with microprocessor and means of telecommunications for remote riding of the results of measurements. An autonomous bottom station consists of original flouting gauges of bottom current intrusion detectors and other devices. The use of original engineering ideas made it possible to improve technical and economical characteristics such as energy consumption, size safety, cost and to simplify it's operation. The prospect of the long term monitoring systems for various tasks of the GOOS program are introduced in the report.

The Underwater Automatic Interfacing System
       Underwater interfacing system is one of the key equipment in submarine rescue, and besides this it can also be used to support demersal people in underwater living cabin . When underwater environment is in adversity, the interfacing task~{!*~}connecting the rescue chamber and the escape hatch together , perhaps is impossible to be finished through the submeriner's operation , maybe for the reason that underwater television is useless in turbidity water, and there are flowing stream which can disturb the rescue submarine violently . In order to perform rescue operation with less limitations on conditions , automatic underwater interfacing system is one of the method to be employed . We have already done some work on developing one kind of such system , and the experimental system will be finished in the end of next year.
   Principle diagram of interfacing system
   The diagram of the underwater automatic interfacing system catching the disabled submarine's escaping hatch, is illustrated in the left figure, the upper part is the rescue hatchway of the rescue submarine , and the lower part in the figure is the escaping hatch of the disabled submarine . Concerning the design features , the following topics about the  interfacing system will be discussed in the paper : 1. The overall description of the system ; 2. The consideration of the automatic interfacing catching area ; 3. the maximal disturbing forces which can be withstood by the system ; 4. The structure and function of the grippers' end-effector ; 5. The automatic interfacing strategy.

The Research on Maneuverability of Underwater Vehicle with Cycloidal Propellers
   Underwater vehicle with cycloid propellers has not been found in the world till now. This paper investigated that problem and pointed out that by setting three cycloid propellers on it, the vehicle can be controlled to move in six directions and it is easier for keeping vehicle's positions and azimuth, so its maneuverability is improved.
   First of all, a formula based on the experimental results of the cycloid propeller with 5 blades is given in the paper. This cycloid propeller's pitch ratio varies from 0.2p to 0.75p and steering angle from -90 degrees to +90 degrees. The experiment was done in Harbin Engineering University towing tank. Using the experimental results, a simulating calculation is also carried out and is compared with that of the vehicle which is controlled by other kind of propellers.

Subsea Applications of Continuum Robots
   This paper describes the design, development and future activities for the ongoing EU MAST II and III programmes AMADEUS I and II. AMADEUS is a dextrous subsea robot hand incorporating force and slip contact sensing, using fluid filled tentacles for fingers. Hydraulic pressure variations in each of three flexible tubes (bellows) in each finger create a bending moment, and consequent motion or increase in contact force during grasping. Such fingers have inherent passive compliance, no moving parts, and are naturally depth pressure-compensated, making them ideal for reliable use in the deep ocean. In addition to the mechanical design, development of the hand has also considered closed loop finger position and force control, co-ordinated finger motion for grasping, force and slip sensor development/signal processing, and reactive world modeling/planning for supervisory 'blind grasping'. Initially, the application focus is for marine science tasks, but broader roles in offshore oil and gas, salvage, and military use are foreseen. Phase I of the project is complete, with the construction of a first prototype. Phase II is now underway, to deploy the hand from an underwater robot arm, and carry out wet trials with users.
    Several mechanical designs evolved in fish are currently inspiring robotic devices for propulsion and manoeuvring purposes in underwater vehicles. Considering the potential benefits involved, this paper presents an overview of the swimming mechanisms employed by fish. The motivation is to provide a relevant and useful introduction to the existing literature for engineers with an interest in the emerging area of aquatic biomechanisms. The fish swimming types are presented following the well-established classification scheme and nomenclature originally proposed by Breder. Fish swim either by Body and/or Caudal Fin (BCF) movements or using Media and/or Paired Fin (MPF) propulsion. The latter is generally employed at slow speeds, offering greater manoeuvrability and better propulsive efficiency, while BCF modes can achieve greater thrust and accelerations. For both BCF and MPF locomotion specific swimming modes are identified, based on the propulsor and the type of movements (oscillatory or undulatory) employed for thrust generation. Along with general descriptions and kinematic data, the analytical approaches developed to study each swimming mode are also introduced. Particular reference is made to lunate tail propulsion, undulating fins and labriform (oscillatory pectoral fin) swimming mechanisms, identified as having the greatest potential for exploitation in artificial systems.

Autonomous Mobile Periscope System (AMPS)
   The worldwide proliferation of diesel electric submarines has impacted Undersea Warfare (USW) world wide. The threat of small, quiet diesel electric submarines plays an increasing roll in littoral warfare strategy of the United Sates Navy. Diesel electric submarines have the capability to operate nearly silently making their detection almost impossible. They do, however have a weakness. All diesel electric submarines require a periscope for navigation in shallow water and a snorkel to provide air to run the generator sets to recharge its batteries. Thus, while acoustically nearly invisible, the primary way of finding a diesel electric submarine is to locate its periscope or snorkel. Of the two the periscope is the most difficult to detect.
    Periscope detection capability for airborne, surface ship and land systems has become a high priority for the US Department of Defense and commercial agencies. Extensive testing is required as new systems are developed to detect the low radar cross section periscopes. As these new periscope detection tools come on line, training is required for the air sea and land based USW crews.
    The United States does not have a diesel electric submarine. The US nuclear submarine force is heavily tasked and not readily available to provide dedicated periscope target services for the RDT&E community and USW crews. As a result, a requirement for a less expensive readily available periscope detection target was defined.
    The Autonomous Mobile Periscope System (AMPS) is being developed by the Pacific Missile Range Facility (PMRF) and the Carderock Division of the Naval Surface warfare center (CDNSWC) formally the David Taylor Model Basin (DTMB), to meet the requirement for an inexpensive readily available periscope detection target. The AMPS will represent the above water portion of a submarine running at periscope depth. A simulated attack periscope and mast will be mounted to a free running, underwater vehicle. The vehicle will position and propel the periscope over a prescribed path and will have the capability to submerge the periscope at various intervals. The vehicle path can be radio controlled or preprogrammed. Other AMPS requirements include the ability to operate for five hours without battery recharge and be launched and recovered by PMRF's 85' range support boars with out a crane.
    Underwater technology has been advanced with the AMPS in the following ways: (1) Specific range support requirements (GPS and acoustic tracking, limited launch and recovery resources, and RF control) were satisfied with the integration commercial off the shelf equipment. (2) A system was developed to control vehicle's on range positioning as well as the vehicle's heave, pitch, and roll in shallow water.

Diesel Engine Integration into Autonomous Underwater Vehicles
    There are numerous technical limitations to underwater vehicle design in general, and the limited physical size and mission aspirations for the utilization of the Autonomous Underwater Vehicle (AUV) make the design of these vehicles even more problematic. However, most of the restrictions are not due to the lack of fundamental knowledge, since the building blocks currently exist in other engineering development areas - the knowledge simply has to be adapted to AUV design.
    The critical enabling technologies which have been identified to fully realize the potential of AUVs are : long endurance propulsion/energy systems; underwater communications; geodetic and relative navigation ; mission management/control; sensors and signal processing; and vehicle design. Analysis of the literature, and in particular the growth of underwater intervention system conferences, infers that progress is being made with continuing developments in the majority of the technology areas, especially microelectronics, micro mechanical sensors and artificial intelligence for mission planning, all of which will reduce the vehicle size, cost, power consumption and increase efficiency. However, perhaps the most critical technology for almost every AUV application, and often the operational limiting factor, is the availability of adequate onboard energy/power. Given the specialist nature of the AUV market, research and development into new AUV-specific power systems is inevitably limited by resources. Therefore the development focus is on near to medium term technology options, which will increase mission duration by a factor of three to five times that provided by the traditional secondary battery system.
    At the present, the relative merits and disadvantages of the competing Air-Independent Power Systems (AIPS) are fairly well known. However, the greatest need of advice is with the "total system" and its integration, i.e., how the AIPS is affected by, and affects the overall vehicle design. Hence, with the numerous design considerations of an AUVs, full knowledge and understanding of the total AIPS integration is essential, if a technically and operationally successful vehicle design is to be achieved.
    In the past, numerous attempts have been made at theoretically, qualitatively and quantitatively, evaluating the vehicle designs and associated power systems. However, the techniques are generally very subjective and open to wide applications and interpretations of design philosophies. What appears to be evident is that the complete design of the underwater vehicle to be used and the missions to be accomplished are essential elements in total vehicle design. Although the evaluations have usually addressed the vehicle aspect of the design, the power system evaluation and integration analysis are still severely limited in application. In addition, where the more advanced analyses have considered a vehicle with a variety of power systems for a particular mission, the power system has been assumed to operate with a fixed energy and power density. Hence the effects of the mission profile,and therefore the performance of the power system operation, have not been examined to evaluate the influence on total vehicle design and vessel operational performance.
    The aim of this paper is to examine the conceptual design of an AUV with specific emphasis on the integration of an Air-Independent Power System, thereby enabling the initial design of AUVs to be evaluated.

Low Power Stirling Engine for Underwater Vehicle Applications
    The selection and design of a power system for any form of underwater vehicle is an extremely complex and difficult task. The system must be capable of providing the vehicle with the required mission performance in terms of power and energy and also be volumetrically and gravimetrically compact.
    When the vehicle to be used in a newly designed U.S. Navy Diver Propulsion Vehicle (DPV), other power system constraints are highlighted. These constraints include limited vehicle diameter, high performance operation, low power requirements, safety and a non-magnetic signature. Of the many power systems available, very few can fulfill the design criteria for the DPV. One system that can is the hydrocarbon fuelled Stirling engine - a dynamic heat using an external combustion system.
    This paper describes the selection, design and development of a Stirling engine powered DPV. Details are given of the specialist vehicle requirements, engine selection and design and the development of a combustion gas recirculation system to enable pure gaseous oxygen to be used as the combustion oxidant. In addition, details are given of the restrictions imposed on component design and manufacture by the low vehicle power requirements.

Biologically-Inspired Dorsal Fin Inverse Control of a Nonminimum Phase Small Undersea Vehicle
   Fishes and Birds have always fascinated mankind since ages. They are far more superior than the most advanced man-made vehicles. Aquatic animals are agile and can do complex maneuvers very quietly. This swiftness and the intricate maneuvers can be attributed to the different fin configuration that they have including the dorsal and caudal fins. Thus, there is a great potential for maneuvering small agile vehicles with these kind of biologically-inspired dorsal and caudal fin-like control surfaces at low speed. Agile and quiet maneuvering of small underwater vehicles with these control surfaces still remains unexplored.
   This paper treats the inverse control of a small undersea vehicle using biologically-inspired dorsal fins. Although, this approach is applicable to yaw plane control, only dive plane maneuvering is considered. It is assumed that the vehicle is under the action of a free surface wave. The transfer function which relates the input force produced by the dorsal fins and the output variable(depth) is nonminimum phase. Since inverse controller cannot control nonminimum phase systems, a new controlled output variable and a new system representation are derived for the design of control systems. Then inverse control law are derived for reference trajectory tracking and it is shown that in the closed loop system, the tracking error is uniformly ultimately bounded. Simulation results are presented to show that the inverse controller accomplish good depth control and regulation of pitch angle inspite of the presence of surface waves and uncertainty in system parameters.
   Key references:
   1. T.Y.T. Wu, C.J. Brokaw, and C. Brennen,eds., Swimming and Flying in Nature, vols. 1 and 2, Plenum Press, New York, 1975.
   2. A. Azuma, The Bio-Kinetics of Flying and Swimming, Springer-Verlag, New York, 1992.
   3. S.N. Singh and P.R. Bandyopadhyay, A Theoretical Control Study of the Biologi cally-Inspired Maneuvering of a Small Vehicle Under a Free Surface Wave, NUWC-NPT Technical Report 10,816, Naval Undersea Warefare Center Division, Newport, RI, 1997.
   4. P.R. Bandyopadhyay, J.M. Castano, J.Q. Rice,R.B. Philips, W.H. Nedderman and W.K. Macy, Low-Speed Maneuvering Hydrodynamics of Fish and Small Underwater Vehicles, ASME Jl. Fluids Engg., vol. 119, 1997, pp. 136-144.