Search our Publications
Latest Publications
Tim Puphal, Malte Probst , Julian Eggert , "Probabilistic Uncertainty-Aware Risk Spot Detector for Naturalistic Driving", Transactions on Intelligent Vehicles, 2019.
AbstractRisk assessment is a central element for the development and validation of Autonomous Vehicles (AV). It comprises a combination of occurrence probability and severity of future critical events. Time Headway (TH) as well as Time-To-Contact (TTC) are commonly used risk metrics and have qualitative relations to occurrence probability. However, they lack theoretical derivations and additionally they are designed to only cover special types of traffic...
Download Bibtex file
Matti Krüger , "The Embodied Vehicle", 11th International ACM Conference on Automotive User Interfaces and Interactive Vehicular Applications, 2019.
AbstractDriver assistance system development commonly targets a substitution of driver responsibilities. Such a substitutive approach which ignores the potential of utilizing available human resources has at least two downsides: 1. The human mind has not evolved to stay idle such that people tend to engage in secondary tasks and disengage from the driving task when not needed, causing out-of-the-loop effects. 2. People have strengths in domains that h...
Download Bibtex file Per Mail Request
Rodrigo Canaan, Julian Togelius, Andy Nealen, Stefan Menzel , "Diverse Agents for Ad-Hoc Cooperation in Hanabi", IEEE Conference on Games, 2019.
AbstractIn complex scenarios where a model of other actors is necessary to predict and interpret their actions, it is often desirable that the model works well with a wide variety of previously unknown actors. Hanabi is a card game that brings the problem of modeling other players to the forefront, but there is no agreement on how to best generate a pool of agents to use as partners in ad-hoc cooperation evaluation. This paper proposes Quality Diversity ...
Download Bibtex file
Thomas Schnürer, Stefan Fuchs, Horst-Michael Groß , "Real-Time 3D Pose Estimation from Single Depth Images ", 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, 2019.
AbstractTo allow for safe Human-Robot-Interaction in industrial scenarios like construction plants, it is essential to always be aware of the location and pose of humans in the shared workspace. We introduce a real-time 3D pose estimation system using single depth images that is aimed to run on limited hardware, like, e.g. , a mobile robot. For this, we optimize a CNN-based 2D pose estimation architecture to achieve high frame rates while simultaneou...
Download Bibtex file
Britta Wrede, Anna-Lisa Vollmer, Lukas Hindemith, Frank Joublin , "People use interaction patterns to teach robots", ICDL-EpiRob 2019 Workshop on Personal Robotics and Secure Human-Robot Collaboration, 2019.
AbstractIn the last years the interest on social robots increased sharply. Thereby we facing the desire to bring robots in private households. In this highly dynamic environment, robots need the ability to learn and act after the needs of the owner. Therefore, the development of flexible interaction protocols for Human-Robot Interaction (HRI) is needed. In 2016, Rohlfing et. al. reintroduced a concept in learning, that can be observed in language acquisi...
Download Bibtex file Per Mail Request
Antonello Ceravola, Frank Joublin, Heiko Wersing, Stephan Hasler, Behzad Dariush, Yi-Ting Chen , "A Data Management Infrastructure for Intelligent Systems", Japanese Society of Artificial Intelligence (JSAI), pp. 45-54, 2018.
AbstractIn this paper we describe the design principles, implementation choices and general challenges we encountered in the creation of a data management infrastructure for recording data streams from test vehicles, robots and other platforms. The trigger for this data management infrastructure project was twofold: First from the proper setup of new test cars equipped with many sensors, delivering high bandwidth data recordings and second from achieving...
Download Bibtex file
Steffen Limmer and Tobias Rodemann , "Robust Evolutionary Optimization Based on Coevolution", Proceedings of EvoApplications 2018, pp. 813-831, 2018.
AbstractA way to deal with uncertainties in the fitness function of an optimization problem is robust optimization, which optimizes the expected value of the fitness. In the context of evolutionary optimization, it is a common practice to compute the expected value of the fitness approximately with the help of Monte-Carlo simulation. This approach requires a lot of evaluations of the fitness function in order to evaluate an individual and thus it can be ...
Download Bibtex file
Simon Manschitz, Michael Gienger, Jens Kober, Jan Peters , "Mixture of Attractors: A novel Movement Primitive Representation for Learning Motor Skills from Demonstrations", IEEE Robotics and Automation Letters (RA-L), 2018.
AbstractMost learning from demonstration approaches still require too many demonstrations to learn a skill or fail to generalize it to new situations. In this paper, we introduce Mixture of Attractors, a novel movement primitive representation, which allows for learning complex skills from very few demonstrations. The movement primitive representation inherently supports multiple coordinate frames, enabling the system to generalize a skill to unseen obje...
Download Bibtex file
Michael Gienger and Thomas Buschmann , "The TUM Bipeds Johnnie and Lola", Humanoid Robotics: a Reference (Book Chapter), Springer, Springer, 2018.
Download Bibtex fileJulian Eggert , "Risk estimation for driving support and behavior planning in intelligent vehicles", Automatisierungstechnik Sonderheft "The car becomes a robot", vol. 66, no. 2, pp. 119-131, 2018.
AbstractGeneral overview on risk estimation and prediction in driving situations and review of our past work on the topic, especially related to survival functions and risk maps....
Download Bibtex file
1 ... 82 83 84 85 86 87 ... 183
Follow us
