Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung IOSB

Filters

8 7
Reset filters

Settings
Now showing 1 - 10 of 15
  • Publication
    Extending Reward-based Hierarchical Task Network Planning to Partially Observable Environments
    ( 2024)
    Mannucci, Tommaso
    ;
    Zimmermann, Robert
    ;
    Frese, Christian
    Rapid, recent developments in robotic applications demand feasible task planning algorithms capable of handling large search spaces. Hierarchical task network (HTN) planning complies with such demand by extending classical planning with task decomposition. Recent advances have extended HTN planners to include the use of reward functions, increasing their flexibility. Nonetheless, such planners assume a fully observable environment, which is often violated in realistic domains. This work contributes to this challenge by presenting POST-HTN, a tree-search based solver which accounts for partial observable environments. A qualitative comparison of POST-HTN with the PC-SHOP HTN solver is given in multiple domains, such as industrial inspection, which is executed on a mobile robot in the real world.
  • Publication
    Cooperative Automated Driving for Bottleneck Scenarios in Mixed Traffic
    ( 2023-06)
    Baumann, M.V.
    ;
    Beyerer, Jürgen
    ;
    Buck, H.S.
    ;
    Deml, Barbara
    ;
    Ehrhardt, Sofie
    ;
    Frese, Christian
    ;
    Kleiser, Dominik
    ;
    Lauer, Martin
    ;
    Roschani, Masoud
    ;
    Ruf, Miriam
    ;
    Stiller, Christoph
    ;
    Vortisch, Peter
    ;
    Ziehn, Jens
    Connected automated vehicles (CAV), which incorporate vehicle-to-vehicle (V2V) communication into their motion planning, are expected to provide a wide range of benefits for individual and overall traffic flow. A frequent constraint or required precondition is that compatible CAVs must already be available in traffic at high penetration rates. Achieving such penetration rates incrementally before providing ample benefits for users presents a chicken-and-egg problem that is common in connected driving development. Based on the example of a cooperative driving function for bottleneck traffic flows (e.g. at a roadblock), we illustrate how such an evolutionary, incremental introduction can be achieved under transparent assumptions and objectives. To this end, we analyze the challenge from the perspectives of automation technology, traffic flow, human factors and market, and present a principle that 1) accounts for individual requirements from each domain; 2) provides benefits for any penetration rate of compatible CAVs between 0 % and 100 % as well as upward-compatibility for expected future developments in traffic; 3) can strictly limit the negative effects of cooperation for any participant and 4) can be implemented with close-to-market technology. We discuss the technical implementation as well as the effect on traffic flow over a wide parameter spectrum for human and technical aspects.
  • Publication
    Ghost: getting invisibly from position A to position B
    ( 2023)
    Häufel, Gisela
    ;
    Bulatov, Dimitri
    ;
    Emter, Thomas
    ;
    Frese, Christian
    ;
    Kottler, Benedikt
    ;
    Petereit, Janko
    ;
    Schmidt, Arno
    ;
    Solbrig, Peter
    ;
    Stütz, Dominik
    ;
    Wernerus, Peter
    The Ghost GR Vision 60 V5 robot is a walking robot from Ghost Robotics, which has extensive sensor and autonomy capabilities. The autonomy functions are provided by Fraunhofer IOSB’s Algorithm Toolbox (ATB) and enable the robot to perform localization, obstacle mapping, path planning, and path control. While Ghost autonomously moves from position A to position B, the environment is captured by laser scanners. Normally, the shortest path between the two points is taken. But what happens if this path can be seen by enemy observers and the walking robot is disturbed or, in the worst case, destroyed? At Fraunhofer IOSB, a fusion of the sensed environment based on laser scanners, existing height models and a coverage map was implemented on the walking robot’s system. This enables the robot to navigate tactically: It will - whenever possible - actively and autonomously seek cover in the shelter of houses and other objects on its way to the given target point.
  • Publication
    A Cooperative HCI Assembly Station with Dynamic Projections
    ( 2020)
    Lengenfelder, Christian
    ;
    Frese, Christian
    ;
    Zube, Angelika
    ;
    Voit, Michael
    ;
    Beyerer, Jürgen
    This paper presents a cooperative human-computer interaction (HCI) assembly station which is assisting a worker during a manual assembly process. The worker's identity, body pose and height is determined to provide individualized assistance like a robot arm which is holding workpieces in a position which is ergonomic for the worker. A second robot arm is equipped with a camera and projector to precisely project information directly on the workpiece. Safe and intuitive human-robot collaboration is achieved by means of workspace monitoring, force detection, compliant control, and hand-guiding. A distinctive feature of this assembly station is that new assembly steps and documentation can be added interactively directly at the workstation during an assembly by interacting with the projected GUI through hand gestures. This paper will detail an assembly station that was developed in a laboratory environment.
  • Publication
    Generic Convoying Functionality for Autonomous Vehicles in Unstructured Outdoor Environments
    ( 2020)
    Albrecht, Alexander
    ;
    Heide, Nina Felicitas
    ;
    Frese, Christian
    ;
    Zube, Angelika
    Autonomously following a leading vehicle is a major step towards fully autonomous vehicles. The contribution of this work consists in the development, implementation, and validation of two following modes: 1) Exact following: accurate compliance with the reference path. 2) Flexible following: tolerate deviation from the reference path in order to avoid obstacles. The proposed method can easily be integrated into existing frameworks for autonomous vehicles. Therefore our approach is flexible enough to be applied to a large variety of different vehicles. To demonstrate the feasibility of our approach an experimental validation is carried out on two autonomous vehicles with major differences in kinematics, weight, and size: A cross-country wheelchair and an off-road truck. Both exact and flexible following have been successfully demonstrated in unstructured outdoor environments.
  • Publication
    Real-time hyperspectral stereo processing for the generation of 3D depth information
    ( 2018)
    Heide, Nina
    ;
    Frese, Christian
    ;
    Emter, Thomas
    ;
    Petereit, Janko
    We present a local stereo matching method for hyperspectral camera data, allowing multiple usage of camera hardware and imaging data such as for object classification or spectral analysis and multichannel input to the correspondence problem. The matching process combines correlation-based similarity measures for pixel windows utilizing all 16 spectral channels followed by a consistency check for disparity selection. We evaluate stereo-processing methods focusing on effectiveness and runtime of the processing on a CPU and analyze parallelization possibilities. Based on the results of the evaluation on the CPU, we implement the optimized stereo matching for images with 16 channels on a graphics processing unit (GPU) utilizing the Compute Unified Device Architecture (CUDA). The parallel processing of the calculation steps to obtain the disparity image on the GPU achieves more than 27× speed up, resulting in calculation and post-processing of hyperspectral images with 8 - 13 Hz, depending on the selection of maximum disparity. The 3D reconstruction achieves a mean square error of 0.0267 m 2 in distance measurements from 5 - 10 m2.
  • Publication
    A non-invasive cyberrisk in cooperative driving
    ( 2017)
    Bapp, F.
    ;
    Becker, J.
    ;
    Beyerer, Jürgen
    ;
    Doll, J.
    ;
    Filsinger, Max
    ;
    Frese, Christian
    ;
    Hubschneider, C.
    ;
    Lauber, A.
    ;
    Müller-Quade, J.
    ;
    Pauli, M.
    ;
    Roschani, Masoud
    ;
    Salscheider, O.
    ;
    Rosenhahn, B.
    ;
    Ruf, Miriam
    ;
    Stiller, C.
    ;
    Willersinn, Dieter
    ;
    Ziehn, J.
    This paper presents a hacking risk arising in fully automated cooperative driving. As opposed to common cyber risk scenarios, this scenario does not require internal access to an automated car at all, and is therefore largely independent of current on-board malware protection. A hacker uses a wireless mobile device, for example a hacked smartphone, to send vehicle to- vehicle (V2V) signals from a human-driven car, masquerading it as a fully-automated, cooperating vehicle. It deliberately engages only in high-risk cooperative maneuvers with other cars, in which the unwitting human driver is expected to perform a specific maneuver to avoid collisions with other vehicles. As the human driver is unaware of the planned maneuver, he fails to react as expected by the other vehicles; depending on the situation, a severe collision risk can ensue. We propose a vision-based countermeasure that only requires state-of-the-art equipment for fully-automated vehicles, and assures that such an attack without internal access to an automated car is impossible.
  • Publication
    Model predictive contact control for human-robot interaction
    ( 2016)
    Zube, Angelika
    ;
    Hofmann, J.
    ;
    Frese, Christian
    For shared human-robot workspaces and safe physical human-robot interaction, robots have to react adequately to intended and unintended contacts with their environment. In this contribution, a Nonlinear Model Predictive Control approach is presented that allows to move the robot end-effector along Cartesian trajectories while at the same time the robot reacts compliantly to contacts based on the estimated contact force. The impairment of the trajectory following task during contacts is minimized by exploiting the robot redundancy. The controller is based on the kinematic robot model. That means no exact knowledge of the robot dynamics is required and the approach is applicable to both fixed-base and mobile manipulators. In contrast to classical control strategies, joint constraints and self-collisions can be directly considered. In order to reduce the amount of unintended contacts, the control approach can also be combined with a collision avoidance extension. The developed algorithms are validated in experimental results on a KUKA LWR IV.
  • Publication
    Sampling-based path planning to cartesian goal positions for a mobile manipulator exploiting kinematic redundancy
    ( 2016)
    Seyboldt, Ruben
    ;
    Frese, Christian
    ;
    Zube, Angelika
    A robot path planner is presented which integrates both collision-free path planning and finding an inverse kinematics solution in a single search stage. Thereby the degrees of freedom resulting from redundant robot kinematics with regard to goal positions specified in the Cartesian workspace are used to optimize the path length, whereas the usual two-stage approach of decoupled inverse kinematics computation and path planning in the configuration space fails to exploit this potential. In addition to sampling-based exploration of the configuration space, the proposed algorithm uses a computationally efficient gradient descent method for approaching the Cartesian goal pose. Furthermore, the planner is extended by the possibility to respect end-effector path orientation constraints. A detailed evaluation of the planning performance in comparison with a two-stage planner is presented.
  • Publication
    AgriApps: an App-based Solution for Field-Robot-Based Agriculture
    ( 2015)
    Frese, Christian
    ;
    Frey, Christian
    ;
    Meßmer, F.
    ;
    Pfeiffer, K.
    ;
    Sander, S.
    ;
    Marco, Daniel di
    ;
    Wenger, M.
    ;
    Albert, A.
    ;
    Wopfner, M.
    ;
    Burkhardt, A.
    ;
    Hochdorfer, S.
    ;
    Bittner, M.
    ;
    Bosch, J.
    ;
    Strobel, M.
    · Today agriculture has to deal with an increasing cost pressure. Innovative Field-Robot-Based solutions can help to handle this challenge. Until now, such systems are not profitable and technically mature for commercial use. In the project AgriApps these issues are addressed. An application based system is developed where a carrier platform can be equipped with different apps. This allows the usage of the system throughout the year to handle different tasks. Thus such a system is profitable for end users. To evaluate the developed concepts. an app for mechanical weed control in special cultures was developed. It consists of a multi-sensor setup for plants and weed detection and a manipulator for digging the soil around the plants. This not only removes the weed but also provides better water absorption for the plants. The app is finally tested on the BoniRob platform, which is the reference platform in this project. A novelty and major contribution of the project is the general definition and specification of the interface between robot and the apps, which covers mechanical and electrical issues as well as the information flow, static and dynamic working area descriptions and energy demand management. The general approach of the interface specification is not limited on robotics, especially it can be transferred to working implements on mobile machinery.