Intelligent control systems

novel approach to extend the decision-making capabilities of unmanned surface vehicles
(USVs) is presented in this work. A multi-objective framework is described where separate
controllers command different behaviors according to a desired trajectory. Three behaviors
are examined – transiting, station-keeping and reversing. Given the desired trajectory, the
vehicle is able to autonomously recognize which behavior best suits a portion of the
trajectory. The USV uses a combination of a supervisory switching control structure and a
reinforcement learning algorithm to create a hybrid deliberative and reactive approach to
switch between controllers and actions. Reinforcement learning provides a deliberative
method to create a controller switching policy, while supervisory switching control acts
reactively to instantaneous changes in the environment. Each action is restricted to one
controller. Due to the nonlinear effects in these behaviors, two underactuated backstepping
controllers and a fully-actuated backstepping controller are proposed for each transiting, reversing and station-keeping behavior, respectively, restricted to three degrees of freedom.
Field experiments are presented to validate this system on the water with a physical USV
platform under Sea State 1 conditions. Main outcomes of this work are that the proposed
system provides better performance than a comparable gain-scheduled nonlinear controller
in terms of an Integral of Absolute Error metric. Additionally, the deliberative component
allows the system to identify dynamically infeasible trajectories and properly
accommodate them.

Options for tracking dynamic underwater targets using optical methods is currently limited. This thesis examines optical reflectance intensities utilizing Lambert’s Reflection Model and based on a proposed underwater laser tracking system. Numerical analysis is performed through simulation to determine the detectable light intensities based on relationships between varying inputs such as angle of illumination and target position. Attenuation, noise, and laser beam spreading are included in the analysis. Simulation results suggest optical tracking exhibits complex relationships based on target location and illumination angle. Signal to Noise Ratios are a better indicator of system capabilities than received intensities. Signal reception does not necessarily confirm target capture in a multi-sensor network.

The design and validation of a low-level backstepping controller for speed and
heading that is adaptive in speed for a twin-hulled underactuated unmanned surface
vessel is presented. Consideration is given to the autonomous launch and recovery of an
underwater vehicle in the decision to pursue an adaptive control approach. Basic system
identification is conducted and numerical simulation of the vessel is developed and
validated. A speed and heading controller derived using the backstepping method and a
model reference adaptive controller are developed and ultimately compared through
experimental testing against a previously developed control law. Experimental tests show
that the adaptive speed control law outperforms the non-adaptive alternatives by as much
as 98% in some cases; however heading control is slightly sacrificed when using the
adaptive speed approach. It is found that the adaptive control law is the best alternative
when drag and mass properties of the vessel are time-varying and uncertain.

Multi-agent control is a very promising area of robotics. In applications for which it is difficult or impossible for humans to intervene, the utilization of multi-agent, autonomous robot groups is indispensable. This thesis presents a novel approach to reactive multi-agent control that is practical and elegant in its simplicity. The basic idea upon which this approach is based is that a group of robots can cooperate to determine the shortest path through a previously unmapped environment by virtue of redundant sharing of simple data between multiple agents. The idea was implemented with two robots. In simulation, it was tested with over sixty agents. The results clearly show that the shortest path through various environments emerges as a result of redundant sharing of information between agents. In addition, this approach exhibits safeguarding techniques that reduce the risk to robot agents working in unknown and possibly hazardous environments. Further, the simplicity of this approach makes implementation very practical and easily expandable to reliably control a group comprised of many agents.

Designing a dependable network for a highly sustainable system gives a challenging network design problem. The network must be highly adaptive to the changes in the network environment. It should also sustain any damages occurring in the network and recover itself quickly and efficiently. This thesis ultimately maps a real network to simulated network by developing a concept of generic nodes and experimentally investigates different parameters that affects the reliability of the system. The work includes designing a simulation for generation of network traffic in a simulated network and studying the behavior of the network with different parameters. The experiment helped us in determining the optimum values of these parameters. For the selected set of experiments and further implies that simulation can determine the nodes different parameter in a control network and will result in a Dependable system.

This dissertation presents the design, implementation and application of soft computing methodologies to Reverse Osmosis (RO) desalination technology. A novel intelligent control scheme based on the integration of Neural Network (NN) and Fuzzy Logic (FL) is presented to optimize plants' performance. In the first part of the research work, two optimal NN predictive models, based on backpropagation and Radial Basis Function Networks (RBFN), were developed for three types of RO feed intakes. The predictive models utilized actual operating data for the three RO plants in order to predict system recovery, total dissolved solids and ion product concentration in brine stream A predictive model is proposed based on redistributed receptive fields of RBFN. The proposed algorithm utilizes integration of supervised learning of centers and unsupervised learning of output layer weights. Extensive simulations are presented to demonstrate the effectiveness of the proposed method for generalization on prediction of nonlinear input-output mappings. In the second part of the study, the design of FL control strategy for direct seawater RO system is carried out. The real-time controller design is based on integration of sensory information, predicted outputs, mathematical calculations, and expert knowledge of the process to yield a constant recovery, constant salt rejection and minimum scaling under variable operating conditions. To implement the designed methodology, a 250/800 Gallon per Day (GPD) prototype RO plant with direct Atlantic Ocean intake is constructed at FAU Gumbo Limbo research laboratory. Two types of membrane modules were used for this study: Spiral Wound (SW) and Hollow Fine Fiber (HFF). The prototype plant indeed demonstrated the effectiveness and optimum performance of the proposed design under variable operating conditions. The system achieved a constant recovery of 30% and salt passage of 1.026% while ion product concentration for six major salts were kept below their solubility limits at all time. The implementation of the proposed intelligent control methodology achieved a 4% increase in availability and a 50% reduction in manpower requirements, as well as reduction in overall chemical consumption of the plant. Therefore, it is expected that the cost of producing fresh water from seawater desalination will be decreased using the proposed intelligent control strategy.

Design and Tuning a fuzzy logic controller (FLCs) are usually done in two stages. In the first stage, the structure of a FLC is determined based on physical characteristics of the system. In the second stage, the parameters of the FLC are selected to optimize the performance of the system. The task of tuning FLCs can be performed by a number of methods such as adjusting control gains, changing membership functions, modifying control rules and varying control surfaces. A method for the design and tuning of FLCs through modifying their control surfaces is presented in this dissertation. The method can be summarized as follows. First, fuzzy control surfaces are modeled with Bezier functions. Shapes of the control surface are then adjusted through varying Bezier parameters. A Genetic Algorithm (GA) is used to search for the optimal set of parameters based on the control performance criteria. Then, tuned control surfaces are sampled to create rule-based FLCs. To further improve the system performance, continuity constraints of the curves are imposed. Under the continuity constraints with the same number of tunable parameters, one can obtain more flexible curves that have the potential to improve the overall system performance. An important issue is to develop a new method to self-tune a fuzzy PD controller. The method is based on two building blocks: (I) Bezier functions used to model the control surfaces of the fuzzy PD controller; and, shapes of control surfaces are then adjusted by varying Bezier parameters. (II) The next step involves using a gradient-based optimization algorithm with which the input scaling factors and Bezier parameters are on-line tuned until the controller drives the output of the process as close as possible to the reference position. To protect vendors and consumers from being victimized, various trust models have been used in e-commerce practices. However, a strict verification and authentication process may pose unnecessary heavy cost to the vendor. As an application of the control strategy proposed, this dissertation presents a solution to the reduction of costs of a vendor. With two fuzzy variables (price, credit-history), a trust-surface can be tuned to achieve an optimal solution in terms of profit margin of the vendor. With this new approach, more realistic trust decisions can be reached.

Design of intelligent systems that can learn from the environment and adapt to the change in the environment has been pursued by many researchers in this age of information technology. The Group Method of Data Handling (GMDH) algorithm to be implemented is a multilayered neural network. Neural network consists of neurons which use information acquired in training to deduce relationships in order to predict future responses. Most software tool during the simulation of the neural network based algorithms in a sequential, single processor machine like Pascal, C or C++ takes several hours or even days. But in this thesis, the GMDH algorithm was modified and implemented into a software tool written in Verilog HDL and tested with specific application (XOR) to make the simulation faster. The purpose of the development of this tool is also to keep it general enough so that it can have a wide range of uses, but robust enough that it can give accurate results for all of those uses. Most of the applications of neural networks are basically software simulations of the algorithms only but in this thesis the hardware design is also developed of the algorithm so that it can be easily implemented on hardware using Field Programmable Gate Array (FPGA) type devices. The design is small enough to require a minimum amount of memory, circuit space, and propagation delay.

Several technologies are being made available for the blind and the visually impaired with the use of infrared and sonar sensors, Radio Frequency Identification, GPS, Wi-Fi among others. Current technologies utilizing microprocessors increase the device's power consumption. In this project, a Verilog Hardware Language (VHDL) designed handheld device that autonomously guides a visually impaired user through an obstacle free path is proposed. The goal is to minimize power consumption by not using the usual microcontroller and replacing it with components that can increase its speed. Utilizing six infrared sensors, the handheld device is modeled after current technologies which use IR and sonar sensors which are reviewed in this project. By using behavioral modeling, an algorithm for obstacle avoidance and the generation of the obstacle free path is reduced using a K-map and implemented using a multiplexer.