Tim Sziburis

• Dott. Mag. M.Sc. M.Sc.
• Ruhr University Bochum

The identification of individual movement characteristics sets the foundation for the assessment of personal rehabilitation progress and can provide diagnostic information on levels and stages of movement disorders. This work presents a preliminary study for differentiating individual motion patterns using a dataset of 3D upper-limb transport traje...

The Ruhr Hand Motion Catalog of Human Center-Out Transport Trajectories [1] is a compilation of three-dimensional task-space motion data simultaneously measured by two motion tracking systems. The first one, an optical motion capture system, provided robust reference data. The second recording system consisted of a single state-of-the-art IMU to de...

This work presents the design, implementation and validation of learning techniques based on the kNN scheme for gesture detection in prosthetic control. To cope with high computational demands in instance-based prediction, methods of dataset reduction are evaluated considering real-time determinism to allow for the reliable integration into battery...

Variability analysis bears the potential to differentiate between healthy and pathological human movements [1]. Our study is conducted in the context of developing a portable glove for the diagnosis of movement disorders. This proposal has methodical as well as technical requirements. Generally, the identification of movement disorders via an analy...

This work presents the design, implementation and validation of learning techniques based on the kNN scheme for gesture detection in prosthetic control. To cope with high computational demands in instance-based prediction, methods of dataset reduction are evaluated considering real-time determinism to allow for the reliable integration into battery...

In our research, we model human upper-limb motion by means of the attractor dynamics approach as a promising candidate for the generation of human-like trajectories. For this purpose, we introduce a systematic dataset of 3D center-out hand movements measured by an Intertial Measurement Unit (IMU) attached to a cylindric transport object. Former stu...

The goal of this work is the development of a motion model for sequentially timed movement actions in robotic systems under specific consideration of temporal stabilization, that is maintaining an approximately constant overall movement time (isochronous behavior). This is demonstrated both in simulation and on a physical robotic system for the tas...

Timing plays a vital role in the generation of naturalistic behavior satisfying all constraints arising from interacting with a dynamic environment while adapting the planning and execution of action sequences online. In biological systems, many of the physiological and anatomical functions follow a particular level of periodicity and stabilization...

This work has been conducted in the context of pattern-recognition-based control for electromyographic prostheses. It presents a k-nearest neighbour (kNN) classification technique for gesture recognition, extended by a proportionality scheme. The methods proposed are practically implemented and validated. Datasets are captured by means of a state-o...

This work has been conducted in the context of pattern-recognition-based control for electromyographic prostheses. It presents a k-nearest neighbour (kNN) classification technique for gesture recognition, extended by a proportionality scheme. The methods proposed are practically implemented and validated. Datasets are captured by means of a state-o...

Current systems of electromyographic prostheses are controlled by machine learning techniques for gesture detection. Instance-based learning showed promising results concerning classification accuracy and robustness without explicit model training. However, it suffers from high computational demands in the prediction phase, which can be problematic...

This work has been conducted in the context of pattern-recognition-based control for electromyographic prostheses. It presents a k-nearest neighbour (kNN) classification technique for gesture recognition, extended by a proportionality scheme. The methods proposed are practically implemented and validated. Datasets are captured by means of a state-o...

Development of kinematics model based control algorithms for the CERN Universal Adjustment Platform to automatize device adjustment in the High Luminosity LHC project

In this project we implemented a real-time audio processing application on the ZedBoard with the Xilinx Zynq-7000. The app is working as a digital effect board that includes five different effect models. Both the input and the output are audio streams reading from and writing to the audio jack ports. The software has a graphical user interface whic...

This master's thesis is related to control and application concepts as well as realizations in the context of rehabilitation robotics. It firstly gives a comparative overview of techniques and tools for implementing controller functionalities with a focus on solutions applicable to programmable logic controllers (PLCs). This comprises hardware plat...

This paper analyzes selected machine learning and pattern recognition techniques in the field of speaker recognition. Firstly, it provides a state-of-the-art analysis of the most common and established speaker recognition methods. Secondly, the implementation of feature extraction and classification methods is introduced, to be trained on sentences...

Project report for contributions to the Working Posture Controller

This thesis’ main concern is the development of a possibility to link data of arbitrary depth camera systems into the framework OpenNI2 for subsequent processing. The implementation of this virtual camera system has been realized by means of a device driver which is able to read in photographical colour picture data as well as depth camera data fro...