Digital Commons Network^™

Using Discovery-Based Learning To Prove The Behavior Of An Autonomous Agent, David N. Mezera

Theses and Dissertations

Computer-generated autonomous agents in simulation often behave predictably and unrealistically. These characteristics make them easy to spot and exploit by human participants in the simulation, when we would prefer the behavior of the agent to be indistinguishable from human behavior. An improvement in behavior might be possible by enlarging the library of responses, giving the agent a richer assortment of tactics to employ during a combat scenario. Machine learning offers an exciting alternative to constructing additional responses by hand by instead allowing the system to improve its own performance with experience. This thesis presents NOSTRUM, a discovery-based learning DBL system …

Design And Construction Of The Aerobot Robotic Manipulator (Arm), William L. Cochran

Theses and Dissertations

This thesis designed, constructed, and tested a robotic arm for the Aerobot Aerial Robot. The main purpose of the ARM is to enable the Aerobot to retrieve objects for use in an annual robotics competition. Design of the ARM involved synthesizing the characteristics of simplicity, weight, strength, and size. The result was a three-degree-of-freedom manipulator that uses electric motors, cable linkages, and telescoping tubes to access a work space below the Aerobot. Forward and inverse kinematics were investigated to enable automation of the ARM. Data was collected from infrared sensors to validate the model. Manipulation of the ARM is presently …

Discovery Learning In Autonomous Agents Using Genetic Algorithms, Edward O. Gordon

Theses and Dissertations

As the new Distributed Interactive Simulation (DIS) draft standard evolves into a useful document and distributed simulations begin to emerge that implement parts of the standard, there is renewed interest in available methods to effectively control autonomous aircraft agents in such a simulated environment. This investigation examines the use of a genetics-based classifier system for agent control. These are robust learning systems that use the adaptive search mechanisms of genetic algorithms to guide the learning system in forming new concepts (decision rules) about its environment. By allowing the rule base to evolve, it adapts agent behavior to environmental changes. Addressed …

A Study Of Automated Estimating Of Software Cost, Steven Shouli Wang

Masters Theses

The objectives of this thesis are to survey software cost estimation methods and to discuss how the methods can be applied as conceptual knowledge in a software cost estimation expert database system. The various applications of expert database systems are discussed. Consequently, a new method to approach the software cost estimation has been proposed.

This proposed method is based upon a dozen years of analyzing software cost models which have been done by some dedicated scientists in universities, research organizations, and industry. A CASE tool called IASCE has been designed to assist the project manager to estimate a proposed project …

Real Time Texture Analysis From The Parallel Computation Of Fractal Dimension, Halford I. Hayes Jr.

Computer Science Theses & Dissertations

The discrimination of texture features in an image has many important applications: from detection of man-made objects from a surrounding natural background to identification of cancerous from healthy tissue in X-ray imagery. The fractal structure in an image has been used with success to identify these features but requires unacceptable processing time if executed sequentially.

The paradigm of data parallelism is presented as the best method for applying massively parallel processing to the computation of fractal dimension of an image. With this methodology, and sufficient numbers of processors, this computation can reach real time speeds necessary for many applications. A …

Recognition Of Quadric Surfaces From Range Data: An Analytical Approach, Ivan X. D. D'Cunha

Electrical & Computer Engineering Theses & Dissertations

In this dissertation, a new technique based on analytic geometry for the recognition and description of three-dimensional quadric surfaces from range images is presented. Beginning with the explicit representation of quadrics, a set of ten coefficients are determined for various three-dimensional surfaces. For each quadric surface, a unique set of two-dimensional curves which serve as a feature set is obtained from the various angles at which the object is intersected with a plane. Based on a discriminant method, each of the curves is classified as a parabola, circle, ellipse, hyperbola, or a line. Each quadric surface is shown to be …

Graphical Evolution Experiments In Artificial Life, Gary R. Greenfield

Department of Math & Statistics Technical Report Series

Larry Yaeger's alife simulation running on a Silicon Graphics Iris Workstation is called Poly World. Our description of PolyWorld is based on notes taken during an oral presentation and video demonstration given in the Artificial Life Panel Session of SIGGRAPH '92: In PolyWorld the visual organisms roam on a bounded two dimensional grid. The organisms "brains" are small neural nets enabling the organisms to control their external visual appearance and to perceive the external world by processing pixmaps. The simulation controls for total energy while striving to explore competition and self-organization. Genes present are for size, strength, maximum speed, mutation …

Robust Reinforcement Learning In Motion Planning, Satinder P. Singh, Andrew G. Barto, Roderic Grupen, Christopher Connolly

Roderic Grupen

While exploring to find better solutions, an agent performing on-line reinforcement learning (RL) can perform worse than is acceptable. In some cases, exploration might have unsafe, or even catastrophic, results, often modeled in terms of reaching 'failure' states of the agent's environment. This paper presents a method that uses domain knowledge to reduce the number of failures during exploration. This method formulates the set of actions from which the RL agent composes a control policy to ensure that exploration is conducted in a policy space that excludes most of the unacceptable policies. The resulting action set has a more abstract …