Intelligent Monitoring and Control

Intelligent monitoring and control involves observing and guiding the behavior of a physical system toward some objective, with real-time constraints on the utility of particular actions. Generic functional requirements for this task include: integration of perception, reasoning , and action; integration of multiple reasoning activities; reasoning about complex, time-varying systems; coordination of multiple response modes; dynamic allocation of limited computational resources. We illustrate these requirements in the domain of patient monitoring in a surgical intensive care unit (SICU). We propose a generic architecture, designed and implemented in layers: top-level system organization ; reasoning architecture; generic reasoning skills and knowledge representation; first-principles knowledge of physical systems; domain knowledge. We illustrate the architecture in the "Guardian" system for SICU monitoring and describe Guardian's performance on an illustrative scenario. Finally, we discuss the generality and limitations of the proposed architecture. 1 The Problem Intelligent monitoring and control involves observing and guiding the behavior of a physical system toward some objective, with real-time constraints on the utility of particular actions. Control theory ([Bollinger and Duffie, 1988],[Hale, 1973]) is useful for tasks that permit a straightforward mapping between sensed data values and appropriate control actions. By contrast, we are con-participation in the project. We gratefully acknowledge contributions by Reed Hastings and Nicholas Parlante to an early version of Guardian and to Adnan Darwiche, who recently joined the project. We have benefitted from discussions with Lawrence Fagan and his students, who are working on a related 1CU monitoring system, called 'QQ.' We thank Edward Feigenbaum for sponsoring the work at the Knowledge Systems Laboratory. cerned with tasks that require a more "intelligent" approach , including: interpretation and prediction of system behavior, diagnosis of exceptional events, explanation of causal mechanisms, reactive response to urgent events, reasoned response to other events, and planning of longer-term courses of action. Such tasks occur in a variety of domains, for example: process control, cri-Our goal is to develop a generic AI architecture for intelligent monitoring and control, suitable for application in multiple domains. In addition to its potential practical utility, this research addresses fundamental AI issues, including: reasoning methods, knowledge representation , resource allocation, distributed intelligence, and real-time performance. In this paper, we propose an architecture and illustrate it in the domain of patient monitoring in a surgical intensive care unit (SICU). Section 2 identifies important generic requirements for intelligent monitoring and control and illustrates them for the SICU monitoring task. …