Main research areas
Research at the chair is based on three closely interlinked pillars.
- Basic research addresses central scientific issues of autonomous robot systems that are independent of the system and application.
- Building on this, the applications pillar focuses on the use of autonomous systems in demanding scenarios, for example in off-road applications, forestry, firefighting and robots in care.
- This is complemented by the development of high-performance technologies, including adaptive architectures, middleware and various robotic platforms, which serve as an experimental and methodological basis for research and transfer.
Adaptive autonomy
Adaptive autonomy is a central research focus of the chair. The aim is to develop autonomy for robotic systems that can flexibly adapt their perception, decision-making and action behavior to changing environmental conditions, tasks and system states. Adaptation is both data-driven on the basis of current sensor data (bottom-up) and situational and knowledge-based through context, mission goals and experience (top-down).
Dependable robot systems (Dependable Robotics)
The chair is concerned with the development of reliable autonomous robot systems that act reliably, safely and comprehensibly in the face of uncertainties, disruptions and incomplete information. The underlying engineering approach encompasses both software and hardware development and considers the entire life cycle of autonomous systems.
Behavior Networks
Behavioral networks and methods of behavior-based robotics are a central concept of the chair's research on modeling and implementing adaptive and reliable behavior of autonomous robot systems. The central idea is the strict decomposition of autonomy into behavioral modules and the targeted design of their respective interactions. This enables decentralized adaptation of control and perception, allowing autonomous systems to react flexibly to changes in the environment, task and system state.
Autonomics
Autonomics describes a young, interdisciplinary scientific discipline that deals comprehensively with the design, understanding and use of autonomous systems. The focus is not only on the classic technical aspects such as perception, decision-making and action of autonomous systems, but also on overarching issues from science and society. These include questions of reliability, safety, human-machine interaction, ethical and legal frameworks and the integration of autonomous systems into complex, real-world environments. The aim of autonomics is to research autonomous systems holistically, validate them and transfer them sustainably into innovative applications.
Off-road environments
The off-road domain places particularly high demands on autonomous robot systems, as it is characterized by unstructured, dynamic and only partially predictable environments. Ground conditions, vegetation, weather conditions and visibility can change continuously and lead to strongly varying requirements for perception, decision-making and control. The Chair's research addresses these challenges through adaptive autonomous concepts, robust perception and control strategies and reliable system architectures that enable the safe and efficient use of autonomous systems even under changing operating conditions.
Robotics middleware FINROC
Finroc is a powerful, modular software framework for the development and operation of complex robotic and control systems in research and application, which has been under development since 2008. The aim is to efficiently build autonomous systems from reusable, quality-tested components. In addition, extensive tools are available, for example for visualization, data acquisition and simulation support, which considerably facilitate the development, analysis and use of robotics software. Finroc is used in both research and industry and is constantly being developed further.
Self-adaptive robot architecture REACTiON
The REACTiON architecture is a modular, behavior-based system design specifically developed to make the perception and decision-making capabilities of autonomous robots in unstructured off-road environments significantly more robust and adaptive. The core idea is not to use classic data-driven algorithms such as neural networks in isolation, but to process and dynamically evaluate their results in the context of behavior, environmental knowledge and situational meaning. The architecture allows automatic reconfiguration and parameterization of the perception and processing procedures according to the characteristics of different robot platforms and sensor technology, which enables a high degree of reusability and flexibility across different systems and areas of application.
Research data repository
This research data repository provides multimodal data sets that were collected as part of extensive off-road field tests under demanding real-life operating conditions. The data was collected using a Unimog equipped with a synchronized sensor system consisting of a laser scanner, LiDAR, cameras, IMU and GNSS. The data records document complex scenarios in different terrain and environmental conditions. The repository thus serves as a valuable basis for research in the fields of autonomous off-road navigation, sensor fusion, localization and robust terrain perception.
News from research
At the Chair of Adaptive Autonomy and Off-Road Robotics, we are guided by the conviction that intelligent systems must interact dynamically with the real world instead of being limited to abstract models. Our current research combines theoretical foundations with practical applications and focuses on robots that can reliably perceive, adapt and decide in highly complex and unpredictable environments. By combining reliable software architectures such as REACTiON with robust platforms, we are developing the next generation of autonomous systems for the most demanding off-road challenges.
Scientific contributions
Discover our latest findings. We transparently share our research findings on adaptive systems, autonomy and off-road technologies with the global scientific and industry community.