Navigation agents and traffic simulation

<p>This thesis addresses the development of intelligent navigation agents for traffic simulation. Discussions are made on issues considering the underlying data structure, the algorithms used for the creation of the data structures, the specification of various agents, and their performance. It also includes descriptions of three different types of agents and the analysis of their performance. One agent has full knowledge of the environment and takes the most efficient path in terms of total path distance. This is what the other agents are compared to. The other two agents make navigation decisions based on taking minimum hops or random choice. Performance analysis is given by comparing those types of agent in two cases: single agent and multiple agents. Issues encountered during this study and further extensions are discussed. The conclusion is that the simulator created here is a good start on a tool for evaluation of independent agents for navigation.</p>

Name thesis:degree_name

Master of Science (MS) in Computer Science

Level thesis:degree_level

Thesis: EWU Only

Discipline thesis:degree_discipline

Computer Science

Year

2005

Author dc:creator

• Hunking, Bart

• Artificial Intelligence and Robotics
• Databases and Information Systems

Statement dc:rights

• Access perpetually restricted to EWU users with an active EWU NetID

Repository record dc:identifier

https://dc.ewu.edu/theses/800

OAI identifier oai:identifier

oai:dc.ewu.edu:theses-1798