Empirical Comparison of Predictive Models for Mobile Agents

Published

December 31, 2004

Author(s)

A E. Henninger, Rajmohan (. Madhavan

Abstract

The need to predict an agent's intents or future actions has been well documented in multi-agent system's literature and has been motivated by both systematically-practical and psychologically-principled concerns. However, little effort has focused on the comparison of predictive modeling techniques. This paper compares the performance of three predictive models all developed for the same, well-defined modeling task. Specifically, this paper compares the performance of an extended Kalman filter based model, a neural network based model and a Newtonian based dead-reckoning model, all used to predict an agent's trajectory and position. After introducing the background and motivation for the research, this paper reviews the form of the algorithms, the integration of the models into a large-scale simulation environment, and the means by which the performance measures are generated. Performance measures are presented over increasing levels of error tolerance.

Citation

American Association of Artificial Intelligence

Pub Type

Journals

Download Paper

Local Download

Keywords

dead-reckoning, error tolerance, extended Kalman filter, neural networks, performance measures

Citation

Henninger, A. and Madhavan, R. (2004), Empirical Comparison of Predictive Models for Mobile Agents, American Association of Artificial Intelligence, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=823488 (Accessed October 31, 2024)