Modeling human behaviors and reactions under dangerous environment

Kang, J., Wright, David, Qin, Sheng-feng and Zhao, Y. (2005) Modeling human behaviors and reactions under dangerous environment. Biomedical Sciences Instrumentation, 41. pp. 265-270. ISSN 0067-8856

Full text not available from this repository. (Request a copy)

Abstract

This paper describes the framework of a real-time simulation system to model human behavior and reactions in dangerous environments. The system utilizes the latest 3D computer animation techniques, combined with artificial intelligence, robotics and psychology, to model human behavior, reactions and decision making under expected/unexpected dangers in real-time in virtual environments. The development of the system includes: classification on the conscious/subconscious behaviors and reactions of different people; capturing different motion postures by the Eagle Digital System; establishing 3D character animation models; establishing 3D models for the scene; planning the scenario and the contents; and programming within Virtools Dev. Programming within Virtools Dev is subdivided into modeling dangerous events, modeling character's perceptions, modeling character's decision making, modeling character's movements, modeling character's interaction with environment and setting up the virtual cameras. The real-time simulation of human reactions in hazardous environments is invaluable in military defense, fire escape, rescue operation planning, traffic safety studies, and safety planning in chemical factories, the design of buildings, airplanes, ships and trains. Currently, human motion modeling can be realized through established technology, whereas to integrate perception and intelligence into virtual human's motion is still a huge undertaking. The challenges here are the synchronization of motion and intelligence, the accurate modeling of human's vision, smell, touch and hearing, the diversity and effects of emotion and personality in decision making. There are three types of software platforms which could be employed to realize the motion and intelligence within one system, and their advantages and disadvantages are discussed.

Item Type:	Article
Additional Information:	This paper was presented at the 42nd Annual Rocky Mountain Bioengineering Symposium at Copper Mountain, Colorado, 8-10 April 2005.
Subjects:	G900 Others in Mathematical and Computing Sciences
Department:	Faculties > Arts, Design and Social Sciences > Design
Depositing User:	Paul Burns
Date Deposited:	20 Mar 2014 10:27
Last Modified:	12 Oct 2019 19:34
URI:	http://nrl.northumbria.ac.uk/id/eprint/15881

Downloads