UGV-CBRN: An Unmanned Ground Vehicle for Chemical, Biological, Radiological, and Nuclear Disaster Response

Simon Schwaiger^1,2, Lucas Muster^1,3, Georg Novotny¹, Michael Schebek¹, Wilfried Wöber¹, Stefan Thalhammer¹ and Christoph Böhm¹

^* Equal Contribution

This work was supported by the Austrian Research Promotion Agency (FFG) through the research project UGV-ABC-Probe (FFG project Call 2020) and the Austrian Armed Forces.

¹ University of Applied Sciences Technikum Wien, Faculty of Industrial Engineering, 1200 Vienna, Austria

² Graz University of Technology, Faculty of Computer Science and Biomedical Engineering, Institute of Software Technology, Inffeldgasse 16b/II, 8010 Graz, Austria

³ University of Natural Resources and Life Sciences, Department of Biotechnology, Institute for Computational Biology, Muthgasse 18, 1190 Vienna, Austria

schwaige@technikum-wien.at, muster@technikum-wien.at

Paper

Code

arXiv

Abstract

Robotic search and rescue (SAR) supports response teams by accelerating disaster assessment and by keeping operators away from hazardous environments. In the event of a chemical, biological, radiological, and nuclear (CBRN) disaster, robots are deployed to identify and locate radiation sources. Human responders then assess the situation and neutralize the danger. The presented system takes a step toward enhanced integration of robots into SAR teams. Integrating autonomous radiation mapping with semi-autonomous substance sampling and online analysis of the CBRN threat lets the human operator localize and assess the threat from a safe distance. Two LiDARs, an IMU, and a Geiger counter are used for mapping the surrounding area and localizing potential radiation sources. A mobile manipulator with six Degrees of Freedom manipulates valves and samples substances that are analyzed by an onboard Raman spectrometer. The human operator monitors the mission’s progression from a remote location defining target locations and directing the semi-autonomous manipulation processes. Diverse recovery behaviours aid robot deployment, system state monitoring, as well as recovery of hard- and software. Field tests showcase the capabilities of the presented system during trials at the CBRN disaster response challenge European Robotics Hackathon (EnRicH).

Directory Structure

/operator: Contains Docker-based workspace run on human operator's PC
/robot: Contains workspace to run on robot's on-board PCs
- 3dparty: Third party methods integrated into the robot
  - exploration: Compose setup running explorer and switching between goalsources
  - mapping: Compose setup starting 2D and 3D mapping (rad mapping is on operator PC)
  - perception: Compose setup for projecting and filtering LiDAR measurements and reading preprocessed camera images from Jetson single board PC
  - sensing: Compose setup for reading LiDAR and IMU data
  - systemd: Blueprint for systemd setup that monitors each 3d party Compose workspace
- catkin_ws: Local, non containerized catkin workspace for components that need to be run bare metal
- manipulation: Compose setup for arm control
- navigation: Compose setup for semi-autonomous behavior and fully autonomous navigation
- systemd: Blueprint for systemd setup that monitors each Compose workspace

Dataset Download

We provide datasets from practical field trials in Rosbag format. (Coming Soon)

Citation

If you use this work in your research, please cite our paper:

@misc{SchwaigerMuster2024UGVCBRN,
    title               = {UGV-CBRN: An Unmanned Ground Vehicle for Chemical, Biological, Radiological, and Nuclear Disaster Response. \textit{arXiv preprint arXiv:2406.14385}}, 
    author              = {Simon Schwaiger and Lucas Muster and Georg Novotny and Michael Schebek and Wilfried Wöber and Stefan Thalhammer and Christoph Böhm},
    year                = {2024},
    url                 = {https://arxiv.org/abs/2406.14385}
}