"This site requires JavaScript to work correctly"
Header minimieren

Prof. Dr.-Ing. Berthold Bäuml

Fakultät Angewandte Informatik

Professor:innen

Professor

BITZ

0991/3615-8712

Sortierung:
Beitrag in Sammelwerk/Tagungsband

  • Lennart Röstel
  • Leon Sievers
  • Johannes Pitz
  • Berthold Bäuml

Learning a State Estimator for Tactile In-Hand Manipulation

  • (2022)

DOI: 10.1109/IROS47612.2022.9981730

We study the problem of estimating the pose of an object which is being manipulated by a multi-fingered robotic hand by only using proprioceptive feedback. To address this challenging problem, we propose a novel variant of differentiable particle filters, which combines two key extensions. First, our learned proposal distribution incorporates recent measurements in a way that mitigates weight degeneracy. Second, the particle update works on non-euclidean manifolds like Lie-groups, enabling learning-based pose estimation in 3D on SE(3). We show that the method can represent the rich and often multi-modal distributions over poses that arise in tactile state estimation. The models are trained in simulation, but by using domain randomization, we obtain state estimators that can be employed for pose estimation on a real robotic hand (equipped with joint torque sensors). Moreover, the estimator runs fast, allowing for online usage with update rates of more than 100 Hz on a single CPU core. We quantitatively evaluate our method and benchmark it against other approaches in simulation. We also show qualitative experiments on the real torque-controlled DLR-Hand II.
  • TC Plattling MoMo
  • DIGITAL
Beitrag in Sammelwerk/Tagungsband

  • Johannes Tenhumberg
  • D. Winkelbauer
  • D. Burschka
  • Berthold Bäuml

Self-Contained Calibration of an Elastic Humanoid Upper Body Using Only a Head-Mounted RGB Camera

pg. 702-707.

  • (2022)

DOI: 10.1109/Humanoids53995.2022.10000184

  • TC Plattling MoMo
  • DIGITAL
Beitrag in Sammelwerk/Tagungsband

  • Johannes Tenhumberg
  • D. Burschka
  • Berthold Bäuml

Speeding Up Optimization-based Motion Planning through Deep Learning

pg. 7182-7189.

  • (2022)

DOI: 10.1109/IROS47612.2022.9981717

  • TC Plattling MoMo
  • DIGITAL
Beitrag in Sammelwerk/Tagungsband

  • D. Winkelbauer
  • Berthold Bäuml
  • M. Humt
  • N. Thuerey
  • R. Triebel

A Two-stage Learning Architecture that Generates High-Quality Grasps for a Multi-Fingered Hand

pg. 4757-4764.

  • (2022)

DOI: 10.1109/IROS47612.2022.9981133

  • Angewandte Informatik
  • DIGITAL
Beitrag in Sammelwerk/Tagungsband

  • J. Tenhumberg
  • Berthold Bäuml

Calibration of an Elastic Humanoid Upper Body and Efficient Compensation for Motion Planning

pg. 98-103.

  • (2021)

DOI: 10.1109/HUMANOIDS47582.2021.9555793

High absolute accuracy is an essential prerequisite for a humanoid robot to autonomously and robustly perform manipulation tasks while avoiding obstacles. We present for the first time a kinematic model for a humanoid upper body incorporating joint and transversal elasticities. These elasticities lead to significant deformations due to the robot’s own weight, and the resulting model is implicitly defined via a torque equilibrium. We successfully calibrate this model for DLR’s humanoid Agile Justin, including all Denavit–Hartenberg parameters and elasticities. The calibration is formulated as a combined least-squares problem with priors and based on measurements of the end effector positions of both arms via an external tracking system. The absolute position error is massively reduced from 21 mm to 3.1 mm on average in the whole workspace. Using this complex and implicit kinematic model in motion planning is challenging. We show that for optimization-based path planning, integrating the iterative solution of the implicit model into the optimization loop leads to an elegant and highly efficient solution. For mildly elastic robots like Agile Justin, there is no performance impact, and even for a simulated highly flexible robots with 20 times higher elasticities, the runtime increases by only 30%.
  • TC Plattling MoMo
  • GESUND
Vortrag

  • J. Tenhumberg
  • Berthold Bäuml

Calibration of an elastic humanoid upper body and efficient compensation for motion planning

In: 2020 IEEE-RAS 20th International Conference 2021

IEEE München

  • 19.-21.07.2021 (2021)
High absolute accuracy is an essential prerequisite for a humanoid robot to autonomously and robustly perform manipulation tasks while avoiding obstacles. We present for the first time a kinematic model for a humanoid upper body incorporating joint and transversal elasticities. These elasticities lead to significant deformations due to the robot’s own weight, and the resulting model is implicitly defined via a torque equilibrium. We successfully calibrate this model for DLR’s humanoid Agile Justin, including all Denavit–Hartenberg parameters and elasticities. The calibration is formulated as a combined least-squares problem with priors and based on measurements of the end effector positions of both arms via an external tracking system. The absolute position error is massively reduced from 21 mm to 3.1 mm on average in the whole workspace. Using this complex and implicit kinematic model in motion planning is challenging. We show that for optimization-based path planning, integrating the iterative solution of the implicit model into the optimization loop leads to an elegant and highly efficient solution. For mildly elastic robots like Agile Justin, there is no performance impact, and even for a simulated highly flexible robots with 20 times higher elasticities, the runtime increases by only 30%.
  • TC Plattling MoMo
  • GESUND
Zeitschriftenartikel

  • J. Vogel
  • D. Leidner
  • A. Hagengruber
  • M. Panzirsch
  • Berthold Bäuml
  • M. Denninger
  • U. Hillenbrand
  • L. Suchenwirth
  • P. Schmaus
  • M. Sewtz
  • A. Bauer
  • T. Hulin
  • M. Iskandar
  • G. Quere
  • A. Albu-Schaffer
  • A. Dietrich

An Ecosystem for Heterogeneous Robotic Assistants in Caregiving: Core Functionalities and Use Cases

In: IEEE Robotics & Automation Magazine

  • (2020)

DOI: 10.1109/MRA.2020.3032142

Demographic change and its various implications will offer some of the biggest challenges faced by society and our health-care systems in the coming decades. While the number of people in need of caregiving is steadily growing in most industrial nations, the number of caregivers is not keeping up with this increasing demand. Robotic assistance systems have the potential to mitigate this problem and support caregivers, people in need, and, thus, the health-care systems in numerous ways. We present the concept and demonstrate the first applica-tion scenarios of a holistic ecosystem for robotic assistants in caregiving. This ecosystem involves various robots covering individual demands and combines several robotic technolo-gies, ranging from autonomous operation over shared control to telepresence modes, for dealing with the wide variety of situ-ations in everyday caregiving. Working toward this ecosystem, we have already implemented its core functionalities on the basis of our robotic prototypes and demonstrate exemplary scenarios to showcase the feasibility of the approach.
  • TC Plattling MoMo
  • DIGITAL
  • GESUND
Beitrag in Sammelwerk/Tagungsband

  • A. Tulbure
  • Berthold Bäuml

Superhuman Performance in Tactile Material Classification and Differentiation with a Flexible Pressure-Sensitive Skin

  • (2018)

DOI: 10.1109/HUMANOIDS.2018.8624987

In this paper, we show that a robot equipped with a flexible and commercially available tactile skin can exceed human performance in the challenging tasks of material classification, i.e., uniquely identifying a given material by touch alone, and of material differentiation, i.e., deciding if the materials in a given pair of materials are the same or different. For processing the high dimensional spatio-temporal tactile signal, we use a new tactile deep learning network architecture TactNet-II which is based on TactNet [1] and is significantly extended with recently described architectural enhancements and training methods. TactN et- Iireaches an accuracy for the material classification task as high as 95.0 %. For the material differentiation a new Siamese network based architecture is presented which reaches an accuracy as high as 95.4 %. All the results have been achieved on a new challenging dataset of 36 everyday household materials. In a thorough human performance experiment with 15 subjects we show that the human performance is significantly lower than the robot's performance for both tactile tasks.
  • TC Plattling MoMo
  • Angewandte Informatik
  • DIGITAL
Beitrag in Sammelwerk/Tagungsband

  • S. Baishya
  • Berthold Bäuml

Robust material classification with a tactile skin using deep learning

  • (2016)

DOI: 10.1109/IROS.2016.7758088

Attaching a flexible tactile skin to an existing robotic system is relatively easy compared to integrating most other tactile sensor designs. In this paper we show that material classification purely based on the spatio-temporal signal of a flexible tactile skin can be robustly performed in a real world setting. We compare different classification algorithms and feature sets, including features adopted and extended from previous works in tactile material classification and that are based on the signal's Fourier spectrum. Our convolutional deep learning network architecture, which we also present here, is directly fed with the raw 24000 dimensional sensor signal and performs best by a large margin, reaching a classification accuracy of up to 97.3%.
  • TC Plattling MoMo
  • Angewandte Informatik
  • DIGITAL
Beitrag in Sammelwerk/Tagungsband

  • R. Wagner
  • U. Frese
  • Berthold Bäuml

Unified treatment of sparse and dense data in graph-based least squares

  • (2016)

DOI: 10.1109/HUMANOIDS.2016.7803393

In this paper, we present a novel method of incorporating dense (e.g., depth, RGB-D) data in a general purpose least-squares graph optimization framework. Rather than employing a loosely coupled, layered design where dense data is first used to estimate a compact SE(3) transform which then forms a link in the optimization graph as in previous approaches [28, 10, 26], we use a tightly coupled approach that jointly optimizes over each individual (i.e. per-pixel) dense measurement (on the GPU) and all other traditional sparse measurements (on the CPU). Concretely, we use Kinect depth data and KinectFusion-style point-to-plane ICP measurements. In particular, this allows our approach to handle cases where neither dense, nor sparse measurements separately define all degrees of freedom (DoF) while taken together they complement each other and yield the overall maximum likelihood solution. Nowadays it is common practice to flexibly model various sensors, measurements and to be estimated variables in least-squares frameworks. Our intention is to extend this flexibility to applications with dense data. Computationally, the key is to combine the many dense measurements on the GPU efficiently and communicate only the results to the sparse framework on the CPU in a way that is mathematically equivalent to the full least-squares system. This results in <;20 ms for a full optimization run. We evaluate our approach on a humanoid robot, where in a first experiment we fuse Kinect data and odometry in a laboratory setting, and in a second experiment we fuse with an unusual “sensor”: using the embodiedness of the robot we estimate elasticities in the kinematic chain modeled as unknown, time-varying joint offsets while it moves its arms in front of a tabletop manipulation workspace. In both experiments only tightly coupled optimization will localize the robot correctly.
  • TC Plattling MoMo
  • Angewandte Informatik
  • DIGITAL
Zeitschriftenartikel

  • O. Birbach
  • U. Frese
  • Berthold Bäuml

Rapid calibration of a multi-sensorial humanoid’s upper body: An automatic and self-contained approach

In: The International Journal of Robotics Research vol. 34 pg. 420-436.

  • (2015)

DOI: 10.1177/0278364914548201

This paper addresses the problem of calibrating a pair of cameras, a Microsoft Kinect sensor and an inertial measurement unit (IMU) mounted at the head of a humanoid robot with respect to its kinematic chain. As complex manipulation tasks require an accurate interplay of all involved sensors, the quality of calibration is crucial for the outcome of the intended tasks. Typical procedures for calibrating are often time-consuming, involve multiple people overseeing a series of subsequent calibration steps and require external tools. We therefore propose to auto-calibrate all sensors in a single, completely automatic and self-contained procedure, i.e. without a calibration plate. By automatically detecting a single point feature on each wrist while moving the robot’s head, the stereo cameras’, the Kinect’s infrared camera’s intrinsic and extrinsic and an IMU’s extrinsic parameters are calibrated while considering the arm joint elasticities and joint angle offsets. All parameters are obtained by formulating the calibration problem as a single least-squares batch-optimization problem. The procedure is integrated on DLR’s humanoid robot Agile Justin allowing to obtain an accurate calibration in around 5 minutes by simply “pushing a button”. The proposed approach is experimentally validated by means of standard metrics of the calibration errors.
  • Angewandte Informatik
  • TC Plattling MoMo
  • DIGITAL
Zeitschriftenartikel

  • T. Hammer
  • Berthold Bäuml

The Communication Layer of the aRDx Software Framework: Highly Performant and Realtime Deterministic

In: Journal of Intelligent & Robotic Systems vol. 77 pg. 171-185.

  • (2015)

DOI: 10.1007/s10846-014-0095-9

Communication between software components is one of the most important functionalities a software framework for modern complex robotic systems has to provide. Here, we present the highly performant realtime communication layer of our new robotic software framework aRDx (agile robot development – next generation), with, e.g., zero-copy semantics, realtime determinism and detailed control of the QoS (quality of service). In addition, we give an in- depth performance comparison to other popular robotic frameworks, namely ROS, YARP, Orocos and aRD.
  • Angewandte Informatik
  • TC Plattling MoMo
  • DIGITAL