A Granular Multi-Sensor Data Fusion Method for Life Support Systems that Enhances Situation Awareness
Abstract
Slow-changing characteristics of controlled environmental systems and the increasing availability of data from sensors and measurements o er opportunities for the development of computational methods that enhance situation observability, decrease human workload, and support real-time decision making. Some of these methods are known as multi-sensor data fusion; they combine measurements from multiple sources to produce a more concise representation of the information contained therein. Such information can be used to design better user-centered interfaces, allowing human operators to maintain situation awareness. Situation observability enables humans to perceive and comprehend the state the system at a given instant of time, and helps human operators in deciding what actions to take at any given time that may a ect the projection of such state into the near future. This paper presents a multi-sensor data fusion method that makes use of a collection of discrete human-inputs and measurements to generate a granular perception function that supports situation awareness. These human-inputs are situation- rich, meaning that they combine measurements defining the operational condition of the system with a subjective assessment of its situation. As a result, the perception function produces situation-rich signals that may be used in ecological human-interfaces or as a switching mechanism in automation strategies and fail-safe/fail-op mechanisms. The granular perception function is a fuzzy associative memory composed of a number of granules equal to the number situations that may be detected by human observers; its development is based in the interaction of human operators with the system. The human-input data sets are transformed into a fuzzy associative memory by an adaptive method based on particle swarms. The paper describes the multi-sensor data fusion method proposed and its application to a ground-based aquatic habitat working as a small-scale environmental system. Results show how this approach helps to generate signals that enhance the situation observability of the aquatic habitat.