KIRAS Security Research

Call results by calendar year

Call results by funding instruments

Projects of the joint German-Austrian call

Cooperative Research and Innovation Projects >Call 2019 >

UAV sensor platform for AI-based support during search & rescue missions

Interactive 3D situational awareness map for search and rescue missions in/around collapsed buildings. Detection and localization of victims, qualitative and quantitative analysis of debris fields, and risk assessment with respect to remaining structural stability and gas leaks.

In numerous catastrophes such as gas explosions, terrorist attacks or natural events, buildings might collapse (in part), rendering the search for missing persons a dangerous and difficult mission. It is important to rescue any victims as quickly as possible, since the chances of survival are the highest within the “golden hour” after the event. However, the safety of the rescue workers is equally important: They must not risk their own lives during the rescue measures.

UAV-Rescue will support the emergency services in being able to assess circumstances and make decisions more quickly. For this purpose, an experimental UAV search-and-rescue system is being developed. It provides the emergency services with a complete real time 3D map of the surroundings without requiring them to step onto the debris fields. The main advantages are that the situation overview (i) maps even hard-to-see areas in detail, (ii) detects and analyzes building damages or debris, (iii) locates missing persons and (iv) informs the emergency services about possible dangers.

The German project partners will leverage AI to develop an UAV system capable of semiautonomous indoor exploration featuring 3D mapping and radar-based person detection. The Austrian consortium explores the outside area and merges indoor and outdoor scans into a common 3D map. Demanding flight maneuvers near structures are carried out semiautomatically. Building damages or debris accumulations are analyzed using neural nets. Automatically optimized camera positioning assists operators when zooming into the 3D model to avoid occlusion of points of interest by other objects or walls. A for automated structural risk assessment is designed and implemented in a case study. Continuous gas monitoring will contribute to an overall hazard assessment.

UAV-Rescue aims to raise underlying technologies for person detection and debris analysis from TRL 2 (Technology Readiness Level) to TRL 4 or TRL 5. The market environment for potential applications in the search-and-rescue area is characterized by accelerating dynamics, which will benefit the industrial partners once the technology has matured and is ready to be deployed in marketable products.

Projektleiter:
Dipl.-Ing. Michael Hofstätter
AIT Austrian Institute of Technology GmbH

Kontakt:
Dipl.-Ing. Michael Hofstätter
Thematic Coordinator
Sensing & Vision Solutions Center for Digital Safety & Security
AIT Austrian Institute of Technology GmbH Giefinggasse 4 | 1210 Vienna | Austria
T +43 (0) 50550-4202 | M +43 664 2351858 | F +43 (0) 50550-4150
michael.hofstaetter@ait.ac.at | www.ait.ac.at 

Projektpartner:
Bundesministerium für Landesverteidigung (BMLV)
Bundesministerium für Inneres (BMI)
Disaster Competence Network Austria – Kompetenznetzwerk für Katastrophenprävention
CBRN Protection GmbH
Johanniter Österreich Ausbildung und Forschung gemeinnützige GmbH
Berufsfeuerwehr Linz
Berufsfeuerwehr Wien
Skyability GesmbH
Rosenbauer International AG
Technische Universität Graz – Institut für Tragwerksentwurf