Elicit: Expertise Learner and Intelligent Causal Inference Tool

Abstract

Knowledge-based systems are tools for solving difficult and complex problems for which standard algorithmic methods of solution are inadequate. The published literature describes very few knowledge-based systems developed to aid in solving complex reasoning tasks encountered in engineering. Planning, monitoring, design, and control tasks are considered to be complex reasoning tasks. Model-based reasoning about physical systems is required, involving time, space, and causality. ELICIT, Expertise Learner and Intelligent Causal Inference Tool, is a knowledge-based system developed to effectively acquire a domain-dependent knowledge base for an Intelligent Simulation Training System (ISTS). An ISTS consists of a graphic computer simulation, an expert system, and an interface. An ISTS is useful for training students in the monitoring and control of simulation objects. To this end, ELICIT accomplishes three goals: 1. achieves an adequate reasoning capability for performing instruction, monitoring, and control related to the graphic computer simulation of a physical system, 2. provides for the automatic acquisition of a knowledge base which is suitable for such reasoning tasks, and 3. maintains a consistent and complete knowledge base. ELICIT acquires an adequate knowledge of the domain to allow situation-dependent reasoning about a specific task. ELICIT's reasoning layer is designed to determi ne appropriate actions to take in the performance of expert behavior in controlling simulation objects. The representation allows the interleaving of causal spatial reasoning with temporal reasoning about sequences of situations. ELICIT accomplishes expert decision-making capability by dynamically modifying predictions of future events to determine what simulation data is needed and when that data should be acquired by the reasoner. A mobile robot simulation was developed to evaluate ELICIT. The robot's goal was to meet with a mobile recharger in an enclosed area and exit through a gate without depleting its energy supply. A robot which accesses positional data from the simulation at regular time intervals was compared to a robot which acquires dynamic spatial simulation information based on expectations of future events. The robot with dynamic prediction capability proved to be more successful at its task. ELICIT represents a significant development in the areas of automated reasoning and knowledge acquisition as related to engineering problem-solving t asks . Specifically, ELICIT provides a representation and acquisition strategy for the knowledge needed for monitoring, control, and instruction in an ISTS.

Notes

This item is only available in print in the UCF Libraries. If this is your thesis or dissertation, you can help us make it available online for use by researchers around the world by STARS for more information.

Graduation Date

1990

Semester

Summer

Advisor

Biegel, John E.

Degree

Doctor of Philosophy (Ph.D.)

College

College of Engineering

Department

Industrial Engineering and Management Systems

Degree Program

Industrial Engineering

Format

PDF

Pages

199. p.

Language

English

Length of Campus-only Access

None

Access Status

Doctoral Dissertation (Open Access)

Identifier

DP0027325

Subjects

Dissertations, Academic -- Engineering; Engineering -- Dissertations, Academic

Accessibility Status

Searchable text

This document is currently not available here.

Share

COinS