Unmanned aircraft systems for airborne reconnaissance in emergency and disaster management.

The SkyObserver project focuses on the development of a concept for the use of unmanned aerial vehicles in identifying and predicting possible dangers on the ground.

Where dangerous situations occur, for example the uncontrolled release of liquid or gaseous toxic substances, it is imperative that detection is reliable and quick so that possible effects can be predicted. It is also important that people fleeing from the danger and already exposed to the toxic substances are found as quickly as possible.

The aim of SkyObserver was therefore to research new methods for detecting dangerous or harmful situations more quickly and comprehensively and hence to reduce the danger to operational forces while at the same time cutting costs. A chemical accident involving the release of toxic gases was selected as the reference scenario, as this situation covers all the aspects under consideration in the project.

The main tasks in the SkyObserver project were to measure the toxic substances, provide an overview of the situation and detect people using unmanned aerial vehicles:

  • Measurement of toxic substances: In order to predict the spread of the toxic gas cloud the dissemination model needs to incorporate the type and rate of release. Ongoing measurements of the concentration of the toxic substance enable realistic forecasts of the spread of the gas cloud. The sooner these measurements are available, the more quickly mission control can obtain a reliable forecast. A swarm of autonomous air vehicles equipped with toxic substance sensors is used for this purpose. The vehicles fly autonomously to the area affected and transmit the measured toxic substance concentration to the ground station.
  • Overview of situation: A swarm of autonomous vehicles is also used to give the operational forces an overview of the situation as a whole and enable assessment of whether the access routes are driveable. The vehicles are equipped with wide-angle cameras that generate images at a resolution that ensures that people’s faces are not recognisable, thus avoiding any possible data privacy issues. When the air vehicles return, the images are transferred to the ground station and compiled into a high-resolution overview image.
  • Person detection: To enable the operational forces to help those affected as quickly as possible, a swarm of autonomous air vehicles is equipped with an embedded system for the detection of moving people. Once again there is scope for optimising effectiveness by varying the number of air vehicles used.

The aircraft used are known as ‘micro aerial vehicles’, which have a maximum take-off weight of about two kilograms. These air vehicles are highly robust in handling and, thanks to their low weight, can be launched by a single person. Given the three different tasks, the project involves a minimum of three air vehicles with different equipment configurations. Time pressures or the size of the affected area may make it necessary to deploy more air vehicles. The task-specific sensors of the air vehicles are designed in a modular fashion so that the operational commander on site can decide which and how many vehicles to use for which tasks.

Facts:

  • Project start: January 2009
  • Project duration: 3 years
  • Funding: the project is funded within the security research support program KIRAS by the Bundesministerium für Verkehr, Innovation und Technologie (federal ministry for transport, innovation and technology)
  • Coordination: AIT Austrian Institute of Technology GmbH
  • Partners: Aerospy Sense & Avoid Technology GmbH, ZAMG – Central Institute for Meteorology and Geodynamics, Johannes Kepler University Linz, SG concepts gmbh, NOVOTECH Elektronik GmbH, market Marktforschungs-Ges.m.b.H. & Co.KG, Federal Ministry of Defence and Sports, Fire Brigade of the City of Linz, Regional Fire Brigade Association for Upper Austria, Disaster Control Centre of the Regional Government of Upper Austria