KI am Tatort
Im Vortrag wird uns Martin eine grundlegende Einführung zu datenbasierter Modellbildung, oft als „KI“ bezeichnet geben. Was sind die Vorteile gegenüber konventionellen Verfahren und was entwickelt die AI-UI GmbH.
Anschließend geht es um:
„KI am Tatort“.
AI for forensic trace detection and classification
Boosting Mask R-CNN Performance for long, thin Forensic Traces with Pre-Segmentation and IoU Region Merging Abstract Mask R-CNN has recently achieved great success in the field of instance segmentation. However, weaknesses of the algorithm have been repeatedly pointed out as well, especially in the segmentation of long, sparse objects whose orientation is not exclusively horizontal or vertical. We present here an approach that significantly improves the performance of the algorithm by first pre-segmenting the images with a PSPNet algorithm. To further improve its prediction, we have developed our own cost functions and heuristics in the form of training strategies, which can prevent so-called (early) overfitting and achieve a more targeted convergence. Furthermore, due to the high variance of the images, especially for PSPNet, we aimed to develop strategies for a high robustness and generalization, which are also presented here. Link to article
AI for live process monitoring in industry 4.0
In-situ monitoring of hybrid friction diffusion bonded EN AW 1050/EN CW 004A lap joints using artificial neural nets Abstract In this work, a dissimilar copper/aluminum lap joint was generated by force-controlled hybrid friction diffusion bonding setup (HFDB). During the welding process, the appearing torque, the welding force as well as the plunge depth are recorded over time. Due to the force-controlled process, tool wear and the use of different materials, the resulting dataseries varies significantly, which makes quality assurance according to classical methods very difficult. Therefore, a Convolutional Neural Network was developed which allows the evaluation of the recorded process data. In thisstudy, data from sound welds as well as data from samples with weld defects were considered. In addition to the different welding qualities, deviations from the ideal conditions due to tool wear and the use of different alloys were also considered. The validity of the developed approach is…
AI-UI and Coral for Edge AI
coral.ai for edge deployment In this tutorial, we try to showcase the way how to use our software in combination with a coral development board to directly deploy self developed AI-models on edge devices. From data to deployment within one package. Overview As you can see, this robotic arm grasps two different kinds of components from the blue basket and puts them in the left. These two kinds of components are plastic and aluminium parts. The machine must find the components and distinguish them. The challenge is, the two different material reflect lights differently, because the light changes during the day inside a production line, which causes the traditional image processing algorithm to not work well often. In this project, we will demonstrate how an industrial AI dedicated for this problem is developed and deployed on the Coral Dev Board, with AI-UI and a few mouse clicks. Coral Dev Board…
AI for brake related, non-exhaust emissions
Artificial Neural Network regression models for the prediction of brake-related emissions Abstract The growth of electric propulsion systems motivates the automotive industry to transfer the focus from exhaust to non-exhaust emissions, with special attention to brake-related emissions. The literature lacks well-established approaches that describe the particulate emissions through reliable analytical correlations. Moreover, the mechanisms of brake particulate formation entail highly stochastic phenomena, which cannot be captured by means of traditional deterministic modelling tools. Machine learning algorithms have been recently used as an alternative method to seek for a branched correlation between tribological properties (i.e. friction coefficient and wear rate), pad composition, environmental and operating conditions. In this regard, the presented work focuses on the study and identification of sophisticated stochastic meta-models for the prediction of the number of emitted brake particulate and associated uncertainty. Specifically, artificial neural networks are developed and validated against brake emission data collected in real driving…