曝光臺 注意防騙 網曝天貓店富美金盛家居專營店坑蒙拐騙欺詐消費者

Representation of Human Behavior
Although the development of computer models that provide a formal representation of human behavior has been the subject of research efforts for many years, the majority of these models have been primarily intended to demonstrate the ability to replicate certain aspects of human behavior in a simplified environment, rather than to model the full range of human behavior as a component of a larger system.  However, more capable models are becoming available that have been designed to be adaptable to a range of applications.  One such model, and the one adopted for the current project, is the Man-machine Integration Design and Analysis System (MIDAS).
MIDAS
MIDAS is based on “first-principles” of human performance (Laughery and Corker, 1997), that is, on the mechanisms that underlie and cause human behavior.  This first-principles characterization has also been termed emergent behavioral representations.  MIDAS is structured as shown in Figure II-1, where its main components comprise the simulated representation of the real world within which the agent modeled by MIDAS exists, and a symbolic operator model (SOM) that represents the perceptual and cognitive activities of the agent.  A key element of the SOM is the Updateable World Representation (UWR), which represents the agent’s knowledge and perception of the world within which it functions.  This is modified through the SOM’s perceptual processes, which take information from the simulated world and update the UWR.  In turn, the SOM selects activities to perform, some of which interact with the representation of equipment in the simulated world and change the behavior of the relevant part of the system.  The activities that are contained within MIDAS are procedures of operator actions.  The environment triggers the activities within the agent.  This series of actions and interactions among the structures within the HPM software is key when attempting to model contextual effects on an agent's performance.
Integration of Human Performance and Air Traffic Simulation Models
The integration of the human performance and air traffic simulation models in the current project was achieved through the use of High-Level Architecture (HLA) networking capabilities implemented in both RFS and MIDAS (DMSO, 2000).  Communication between agents represented in RFS and the corresponding representation in MIDAS is achieved through the design of a networking agent within RFS.  This agent keeps track of whether a particular agent in the simulation is represented in MIDAS and if so directs communications from other RFS agents to MIDAS through the networking agent.
 
Source:  Corker, 2000.

Figure II-1  MIDAS Architecture
Issues in Systemwide Modeling
The application of fast-time simulation to systemwide modeling raises a number of issues that need to be considered in both the design of the simulation software, as well as the specification of simulation experiments to be run.  The most obvious is the need to balance the fidelity with which the system is represented with considerations of computational efficiency.  While it may be desirable to explicitly model the decisions and actions of as many of the human agents as possible using a human performance simulation model, the computational burden involved will limit the number of such agents that can be reasonably modeled.  Therefore the design of the experiments will need to carefully consider how many human agents must be represented in detail in order to obtain reasonable results.  In the experiments performed during the current phase of the research, the simulation modeled activity in a single sector, and the air traffic controller actions and decisions for that sector were represented in detail.  Surrounding sectors were modeled using a much simpler representation of the sector controller, which only served to generate and accept traffic crossing into and out of the sector.
　
中國航空網 www.k6050.com
航空翻譯 www.aviation.cn
本文鏈接地址：DEVELOPMENT OF FAST-TIME SIMULATION TECHNIQUES(16)