Sensors and Learning Maps

Research output: Contribution to conferenceAbstract

Abstract

Recent disasters such as the Fukushima nuclear accident have shown that (informal) sensor data are an important source of information for citizens and professional decision makers. Typically, the information content of sensor data changes with time while it depends on the measured values, which are in turn dependent on the locations where observations are made. Decisions about where to measure require timely integration of available data and prompt feedback as soon as new information becomes available. The expected value of information is a measure of the relevance of future observations within the context of decision making and it is proposed as a tool for automated selection of new measurement locations. On the other hand, human decision making about where to measure and which places to avoid is expected to benefit from real-time mapping using geostatistical methods and live feedback of up-to-date information. In this session we will look at simulated examples of automated mobile sensors exploring a contaminated environment and students mapping (1) an invasive species in a natural park and (2) a fictive dynamic toxic plume over Wageningen campus. The focus is on server side processing of sensor data and decision making.
Original languageEnglish
Publication statusPublished - 2012
EventGeospatial World Forum, Amsterdam, The Netherlands -
Duration: 23 Apr 201227 Apr 2012

Conference

ConferenceGeospatial World Forum, Amsterdam, The Netherlands
Period23/04/1227/04/12

Fingerprint Dive into the research topics of 'Sensors and Learning Maps'. Together they form a unique fingerprint.

  • Cite this

    de Bruin, S. (2012). Sensors and Learning Maps. Abstract from Geospatial World Forum, Amsterdam, The Netherlands, .