arXiv subjects

Florian Müller

Publications and source records attributed to Florian Müller.

22 records · Page 2Linked to original sources

arXiv2022

Experimental validation of a model for a self-adaptive beam-slider system

A system consisting of a doubly clamped beam with an attached body (slider) free to move along the beam has been studied recently by multiple research groups. Under harmonic base excitation, the system has the capacity to passively adapt itself (by slowly changing the slider position) to yield either high or low vibrations. The central contributions of this work are the refinement of the recently developed system model with regard to the finite stiffness of the beam's clamping, followed by a thorough validation of this model against experimental results. With the intent to achieve repeatable and robust self-adaption, a new prototype of the system was designed, featuring, in particular, a continuously adjustable gap size and a concave inner contact geometry. The initial beam model is updated based on the results of an Experimental Nonlinear Modal Analysis of the system (without slider). By varying the excitation level and frequency in a wide range, all known types of behavior were reproduced in the experiment. The simulation results of the updated model with slider are in excellent agreement with the measurements, both qualitatively (type of behavior) and quantitatively. Minor deviations are attributed to the system's sensitivity to inevitable uncertainties, in particular with regard to the friction coefficient and the linear natural frequency. It is thus concluded that the proposed model is well-suited for further analysis of its intriguing dynamics and for model-based optimization for technical applications such as energy harvesting.

nlin.AO↗

arXiv2022

Proper Posture: Designing Posture Feedback Across Musical Instruments

There is a recommended body posture and hand position for playing every musical instrument, allowing efficient and quick movements without blockage. Due to humans' limited cognitive capabilities, they struggle to concentrate on several things simultaneously and thus sometimes lose the correct position while playing their instrument. Incorrect positions when playing an instrument can lead to injuries and movement disorders in the long run. Previous work in HCI mainly focused on developing systems to assist in learning an instrument. However, the design space for posture correction when playing a musical instrument has not yet been explored. In this position paper, we present our vision of providing subtle vibrotactile or thermal feedback to guide the focus of attention back to the correct posture when playing a musical instrument. We discuss our concept with a focus on motion recognition and feedback modalities. Finally, we outline the next steps for future research.

cs.HC↗

arXiv2022

Lateral Ego-Vehicle Control without Supervision using Point Clouds

Existing vision based supervised approaches to lateral vehicle control are capable of directly mapping RGB images to the appropriate steering commands. However, they are prone to suffering from inadequate robustness in real world scenarios due to a lack of failure cases in the training data. In this paper, a framework for training a more robust and scalable model for lateral vehicle control is proposed. The framework only requires an unlabeled sequence of RGB images. The trained model takes a point cloud as input and predicts the lateral offset to a subsequent frame from which the steering angle is inferred. The frame poses are in turn obtained from visual odometry. The point cloud is conceived by projecting dense depth maps into 3D. An arbitrary number of additional trajectories from this point cloud can be generated during training. This is to increase the robustness of the model. Online experiments show that the performance of our method is superior to that of the supervised model.

cs.CV↗

arXiv2018

A novel DOI Positioning Algorithm for Monolithic Scintillator Crystals in PET based on Gradient Tree Boosting

Monolithic crystals are examined as an alternative to segmented scintillator arrays in positron emission tomography (PET). Monoliths provide good energy, timing and spatial resolution including intrinsic depth of interaction (DOI) encoding. DOI allows reducing parallax errors (radial astigmatism) at off-center positions within a PET ring. We present a novel DOI-estimation approach based on the supervised machine learning algorithm gradient tree boosting (GTB). GTB builds predictive regression models based on sequential binary comparisons (decision trees). GTB models have been shown to be implementable in FPGA if the memory requirement fits the available resources. We propose two optimization scenarios for the best possible positioning performance: One restricting the available memory to enable a future FPGA implementation and one without any restrictions. The positioning performance of the GTB models is compared with a DOI estimation method based on a single DOI observable (SO) comparable to other methods presented in literature. For a 12 mm high monolith, we achieve an averaged spatial resolution of 2.15 mm and 2.12 mm FWHM for SO and GTB models, respectively. In contrast to SO models, GTB models show a nearly uniform positioning performance over the whole crystal depth.

physics.med-ph↗