Refine
Year of publication
- 2021 (229) (remove)
Institute
- Fachbereich Gestaltung (55)
- Fachbereich Medizintechnik und Technomathematik (46)
- IfB - Institut für Bioengineering (32)
- Fachbereich Elektrotechnik und Informationstechnik (26)
- Fachbereich Luft- und Raumfahrttechnik (23)
- Fachbereich Wirtschaftswissenschaften (21)
- Fachbereich Energietechnik (19)
- Fachbereich Bauingenieurwesen (13)
- ECSM European Center for Sustainable Mobility (10)
- Fachbereich Maschinenbau und Mechatronik (10)
Has Fulltext
- no (229) (remove)
Document Type
- Article (90)
- Conference Proceeding (52)
- Bachelor Thesis (49)
- Part of a Book (17)
- Book (6)
- Master's Thesis (4)
- Report (4)
- Doctoral Thesis (2)
- Review (2)
- Conference: Meeting Abstract (1)
Keywords
- Animation (3)
- Holz (3)
- Mode (3)
- Nachhaltigkeit (3)
- Redesign (3)
- UX Design (3)
- App (2)
- Bookazine (2)
- Corporate Design (2)
- Corporate Identity (2)
In this paper we investigate the use of deep neural networks for 3D object detection in uncommon, unstructured environments such as in an open-pit mine. While neural nets are frequently used for object detection in regular autonomous driving applications, more unusual driving scenarios aside street traffic pose additional challenges. For one, the collection of appropriate data sets to train the networks is an issue. For another, testing the performance of trained networks often requires tailored integration with the particular domain as well. While there exist different solutions for these problems in regular autonomous driving, there are only very few approaches that work for special domains just as well. We address both the challenges above in this work. First, we discuss two possible ways of acquiring data for training and evaluation. That is, we evaluate a semi-automated annotation of recorded LIDAR data and we examine synthetic data generation. Using these datasets we train and test different deep neural network for the task of object detection. Second, we propose a possible integration of a ROS2 detector module for an autonomous driving platform. Finally, we present the performance of three state-of-the-art deep neural networks in the domain of 3D object detection on a synthetic dataset and a smaller one containing a characteristic object from an open-pit mine.
A new method for improved autoclave loading within the restrictive framework of helicopter manufacturing is proposed. It is derived from experimental and numerical studies of the curing process and aims at optimizing tooling positions in the autoclave for fast and homogeneous heat-up. The mold positioning is based on two sets of information. The thermal properties of the molds, which can be determined via semi-empirical thermal simulation. The second information is a previously determined distribution of heat transfer coefficients inside the autoclave. Finally, an experimental proof of concept is performed to show a cycle time reduction of up to 31% using the proposed methodology.