arXiv subjects

Mark Wielitzka

Publications and source records attributed to Mark Wielitzka.

2 recordsLinked to original sources

arXiv2024

Nonlinear Two-Track Model of a Semitrailer with Experimental Validation of Lateral and Vertical Tire Forces

As part of the automation of commercial vehicles, the number of assistance systems in this field is continuously increasing. The semitrailer plays an important role for the vehicles driving dynamics due to its highly varying loads and the large proportion to the total mass of the truck-semitrailer, especially when it is fully loaded. To create a basis for further development of assistance systems for the semitrailer, this paper presents a two-track model which includes the lateral and roll dynamics of the semitrailer. This enables the future development of observer and filter-based estimation of vehicle states and parameters, which are impossible or very difficult to measure. For offline identification of the unknown model parameters, a Particle-Swarm-Optimization (PSO) algorithm will be used. The validation of the model is based on measurements from a test vehicle. The focus is on the lateral and vertical tire forces of the semitrailer, which are measured at the test vehicle using strain gauges.

eess.SY↗

arXiv2021

Autoencoder-based Representation Learning from Heterogeneous Multivariate Time Series Data of Mechatronic Systems

Sensor and control data of modern mechatronic systems are often available as heterogeneous time series with different sampling rates and value ranges. Suitable classification and regression methods from the field of supervised machine learning already exist for predictive tasks, for example in the context of condition monitoring, but their performance scales strongly with the number of labeled training data. Their provision is often associated with high effort in the form of person-hours or additional sensors. In this paper, we present a method for unsupervised feature extraction using autoencoder networks that specifically addresses the heterogeneous nature of the database and reduces the amount of labeled training data required compared to existing methods. Three public datasets of mechatronic systems from different application domains are used to validate the results.

cs.LG↗