ROBDEKON2

Robotic systems for decontamination in hostile environments Phase II

The walking excavator ARTER which has been equipped with autonomy functions by DFKI (Photo: DFKI GmbH)
The walking excavator ARTER which has been equipped with autonomy functions by DFKI (Photo: DFKI GmbH)

Within the scope of the ROBDEKON competence centre, phase II aims in using robots and humans together in deconstruction or decontamination processes to complement each other in order to simplify the necessary processes or even make them possible in the first place. The first phase of the competence centre has shown that both autonomy functions and close cooperation between humans and robots on site as well as teleoperation scenarios are essential components for efficient processing. The individual components are not to be seen as separate units, but are to be used in mixed forms.

Duration: 01.12.2022 till 30.11.2026
Donee: German Research Center for Artificial Intelligence GmbH
Sponsor: Federal Ministry of Research, Technology and Space (BMFTR)
Grant number: Funded by the Federal Ministry of Education and Research as part of the Federal Government's "Research for Civil Security" program with the funding number 13N16537.
Website: https://robdekon.de/
Partner:

Fraunhofer IOSB, Karlsruher Institut für Technologie (KIT), FZI Forschungszentrum Informatik, Götting KG, Kraftanlagen Heidelberg GmbH, ICP Ingenieurgesellschaft Prof. Czurda und Partner mbH, HKA Hochschule Karlsruhe

Application Field: Agricultural Robotics
SAR- & Security Robotics
Related Projects: ROBDEKON
Robot systems for decontamination in hostile environments (06.2018- 06.2022)
Related Robots: ARTER
Autonomous Rough Terrain Excavator Robot
SherpaTT

Project details

The mobile control station can be loaded from the walking excavator ARTER (Photo: DFKI GmbH).
Interior view of the mobile control station, which can accommodate up to four people (Photo: DFKI GmbH)
The virtual reality laboratory at the DFKI. On 21m² the information coming from the systems can be visualised and by means of OmniDeck (in the middle of the room) the operator can move continuously in the virtual environment.

In this project, the DFKI is pursuing two core objectives, the sampling of the soil of a landfill on the one hand and the holistic representation of the environment in which the robotic systems move on the other. Building on existing work on sampling the subsoil in landfill scenarios, existing technologies are being expanded, generalised and thus made transferable.

In the first phase, cross-institutional cooperation in the use of robotics and software was successfully demonstrated. The possibility of displaying and operating a large number of different systems on a map, on the other hand, was not part of the first phase as planned, but is a basic requirement if several heterogeneous systems are to carry out tasks simultaneously autonomously and semi-autonomously.

A strong focus of the work is therefore now on the processing and provision of the environment and environmental information for human operators. In a mobile control centre, a kind of situation map is created by merging the recorded mapping sensors of the robotic systems; these common maps are made available to the robots.

In the case of safety-relevant tasks that are dangerous for humans, which a robotic system takes over, it must be possible for humans not only to grasp the overall situation at a glance, but also to take over the control of subtasks and thus contribute their experience and cognitive abilities. This will be realised with the new type of control centre. Human operators can immerse themselves in the situation on site and intuitively operate the systems, regardless of the size or type of robotic system, by using state-of-the-art teleoperation methods and cross-system interfaces. This is where the step from pure teleoperation to telepresence is made.

While the systems are operating, the human can intervene at any time and move from the overview level down to the individual systems.

Publications

2025

Beyond Teleoperation: Autonomous-Assisted Soil Sampling for Decontamination in Hazardous Environments Using a Walking Excavator
Yuhan Jin, Haider Khan Lodhi, Fabian Maas genannt Bermpohl, Pierre Willenbrock, Christoph Hertzberg, Steffen Planthaber, Babu Ajish, Daniel Kuehn, Frank Kirchner
In 2025 IEEE International Symposium on Safety Security Rescue Robotics (SSRR), (SSRR-2025), 29.10.-31.10.2025, Galway, IEEE, Oct/2025.
Control of robots with hybrid locomotion capabilities
Babu Ajish
Aug/2025. DFKI.
Stepping Locomotion for a Walking Excavator Robot Using Hierarchical Reinforcement Learning and Action Masking
Babu Ajish, Frank Kirchner
In 2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2025), (IROS-2025), 18.10.-25.10.2025, Hangzhou, IEEE, 2025.
GSeg3D: A High-Precision Grid-Based Algorithm for Safety-Critical Ground Segmentation in LiDAR Point Clouds
Haider Khan Lodhi, Christoph Hertzberg
In 2025 7th International Conference on Robotics and Computer Vision (ICRCV), (ICRCV-2025), 24.10.-26.10.2025, Hong Kong, China, IEEE, 2025.

Back to the list of projects
© DFKI GmbH
last updated 10.06.2025