Forng-Chen Chiu

Design of a sliding mode fuzzy controller for the guidance and control of an autonomous underwater vehicle

Abstract This work demonstrates the feasibility of applying a sliding mode fuzzy controller to motion control and line of sight guidance of an autonomous underwater vehicle. The design method of the sliding mode fuzzy controller offers a systematical means of constructing a set of shrinking-span and dilating-span membership functions for the controller. Stability and robustness of the control system are guaranteed by properly selecting the shrinking and dilating factors of the fuzzy membership functions. Control parameters selected for a testbed vehicle, AUV-HM1, are evaluated through tank and field experiments. Experimental results indicate the effectiveness of the proposed controller in dealing with model uncertainties, non-linearities of the vehicle dynamics, and environmental disturbances caused by ocean currents and waves.

Dynamic characteristic of a biomimetic underwater vehicle

For predicting the dynamic characteristics of a fishlike AUV testbed, which is under development at National Taiwan University, we developed a practical method to simulate the undulatory locomotion of a flexible slender body. When a flexible slender body, which is divided into a number of segments undulates, the wave passes from the nose to the tail. Reaction forces due to momentum change, friction, as well as cross flow drag acting on each segment are taken into account. Equations of motions described by the body-fixed coordinate are obtained by taking the summation of the longitudinal force, lateral force and yaw moment acting on all the segments. Equations of motions are solved step by step in time axis and the velocity is transferred to space-fixed coordinates for integrating the trajectory. In numerical simulation, a digital filtering technique is applied to avoid drifting in sway and yaw motions and the steady state solutions can be obtained. Based on the results of a series of simulation calculations, the dynamic characteristics of the BAUV testbed are shown and discussed in this paper.

Determining the bodily motion of a biomimetic underwater vehicle under oscillating propulsion

The paper describes a biomimetic autonomous underwater vehicle (BAUV) that mimics the shape and behavior of fish. The swimming motion of the BAUV is achieved using an oscillating body. The body spline is specified by a set of parameters, which are utilized using genetic algorithm (Gas) by evaluating a fitness function of the optimization which is defined as the ratio of the forward velocity to the required driving power of the joint motors. The resulting body spline is found to be better than all other body splines at all tail-beating frequencies. Each body spline has an optimal tail-beating frequency.

Adaptive control of an autonomous underwater vehicle testbed using neural networks

The control of autonomous underwater vehicles has been a challenge to control engineers due to combined nonlinear nature of both the vehicle itself and the environment in which they operate. This paper presents an implementation research on the adaptive controller of an autonomous underwater vehicle testbed in which the controller architecture is made using multilayered neural networks. The problem considered is that of designing a controller for an autonomous underwater vehicle to provide directional control. A flux gate compass is used to measure the yaw angle and yaw rate. Directional control is performed by two thrusters in the horizontal plane. Weight adaptation of the neural network is achieved by minimizing an objective function that is weighted sum of tracking errors and control input rates. According to the experimental tests on various command trajectories, we show that when the learning process is kept active through the control operation, the neural network adapts to time-varying plant dynamics as well as disturbance upsets.

A maximum entropy method for multi-AUV grouping

A vector-quantization formulation is used to define a grouping problem for multiple AUVs in a sampled environment. The objective is to minimize a quantization error function. The self-organizing network structure developed by Kohonen is a famous quantization model. Difficulties of applying the Kohonens network is that the convergence property is not guaranteed. In addition, learning gains in the Kohonens network have to be manually adjusted. This paper proposes a control method for the grouping of multiple AUVs under the structure of the Kohonens network. To solve the difficulties encountered in the framework of Kohonens network, we incorporate a Lyapunov function of a thermal statistical model to solve the problem of convergence. The position of each AUV is treated as a probability distribution function under thermal equilibrium. The learning gains are determined using the condition of asymptotically stability of the network. The minimization problem is formulated in a Lagrange optimal form with the constraint of maximum entropy. The intervehicle distance is controlled by the optimal distribution of the entropy. We prove that the global-minimum-error of the cost function can be achieved for the grouping

Maneuverability of a flat-streamlined underwater vehicle

Maneuverability describes a vehicles ability to change course or turn. Maneuverability of conventional underwater vehicles, such as torpedoes, can be determined by altering the position and length of control fins. To perform large-area surveying tasks, autonomous underwater vehicles (AUV) generally require different maneuverability characteristics in their vertical and horizontal planes of motion. Furthermore, AUV are significantly slower than torpedoes, and control fins are relatively ineffective at slow speeds. While relying solely on control fins to determine the maneuverability, this study investigates the maneuverability characteristics of a flat-streamlined underwater vehicle. A planar motion mechanism (PMM) testing system is adopted to conduct a series of captive model tests in order to measure the stability derivatives of the vehicle, AUV-HM1. Stability and maneuvering indices are then derived from the measured data of stability derivatives. Finally, maneuvering criteria of a flat, baseline vehicle are evaluated using a prediction method.

A practical method for simulating pectoral fin locomotion of a biomimetic autonomous underwater vehicle

In this paper, a practical method to simulate pectoral fin locomotion of a fishlike AUV testbed is presented. Basing on a blade element synthesis scheme, the authors developed a simple mathematical model to evaluate the hydrodynamic forces acting on a pectoral fin in feathering motion and lead-lag motion. The pectoral fin is treated as a number of moving blade element, the lift, cross flow drag and added inertia acting on each blade element are evaluated as a two dimensional oscillating thin foil and they are described in the fin-fixed coordinate system. These forces of blade element are transferred to the fuselage-fixed coordinate system and then integrated to obtain the total forces acting on a pectoral fin. Katos data from model test and calculation on a pectoral fin of a bass are cited to compare with the corresponding calculation results by the present simple mathematical model. The quite satisfactory agreement confirmed the validity of the simple model for evaluating the hydrodynamic forces of a pectoral fin. On this base, the previously developed computer program for simulating body-tail undulatory locomotion is then extended to simulate the pectoral fin locomotion of a biomimetic autonomous underwater vehicle

Motion control and way-point tracking of a biomimetic underwater vehicle

AUVs with rigid hulls and powered by rotary propellers have problems such as low propulsion efficiency, difficulties in positioning, agile turning, and precise hovering. Research and development into biomimetic AUVs (BAUVs) is growing in order to overcome the problems of AUVs. A BAUV testbed is being developed. Our BAUV testbed is composed of several links and joints. We develop a local control law that coordinates body angles by periodically alternating the position of the center-of-mass in the local coordinate. We have found that three parameters are sufficient to coordinate joint angles to perform the carangiform swimming. We then develop a global control law for the way-point tracking problem. The effectiveness of this method is evaluated by computer simulation. Control performance of the BAUV system with model uncertainties to track way-points under the influence of disturbances is also discussed.

Design of an Underwater Glider with Fore and Aft Buoyancy Engines

In this paper, we discuss design issues in applying buoyancy engines as the device to vary net buoyancy and to alternate the position of the center of gravity of a glider. The buoyancy engines arrangement contains two tanks located at the fore and end aft part of the hull. Buoyancy engines considered here are those of piston-type. Forces equations which model buoyancy, gravity, and hydrodynamic forces in gliding are derived. Performances of different sizes of buoyancy engines are compared. Operational constrains considering the power consumption of buoyancy engines are also specified. Gliders with rectangular wings of various shape and wing location are then examined in terms of the energy cost for gliding controlled by buoyancy engines.

A recursive neural networks model for ship maneuverability prediction

In this paper, a recursive neural networks model is developed and applied to simulate the maneuvers of a tanker, which is full and may have inherent poor coursekeeping ability. In the present model, component force modules is developed to calculate five component forces as inputs to the networks. It consists of the net thrust and lateral force due to propeller revolution and rudder angle, approximate Munk moment, longitudinal component and lateral component of centrifugal force acting on ship hull. These forces are related to the input control variables such as ruder angle, propeller revolution and the output state variables such as motion velocities by very simplified functions without any undetermined hydrodynamic coefficients or empirical factors. The present recursive neural network is constructed with one input layer, one output layer and two hidden layers. Not only the above-stated forces, but also the outputs of longitudinal velocity, lateral velocity and yaw rate are fed back to the input layer of the network. In this study, an existing ship maneuvering simulation program, which has been developed basing on Japan MMG hydrodynamic model, is used for generating all the sample data of maneuvers for training and validating the recursive neural networks. The ship maneuvering motions are investigate d using the recursive neural networks which has been trained on limited maneuvers including turning, zigzag, spiral as well as accelerating maneuvers, and its validity to predict the maneuverability of a full ship with poor course stability is also discussed