Refine
Document Type
- Bachelor thesis (2)
- Working paper (1)
Has Fulltext
- yes (3)
Is part of the Bibliography
- no (3)
Year of publication
- 2021 (3)
Keywords
- Nichtlineare Optimierung (3) (remove)
Institute
Die Bachelorarbeit beschäftigt sich mit der Entwicklung effizienter Methoden für die Klassifizierung der Stadien von Zellen. Hierbei liegt der Fokus auf sogenannten Deep-Learning-Algorithmen. Diese haben sich unter anderem in der Bilderkennung als sehr leistungsfähig erwiesen und können genutzt werden, um große Mengen von mikroskopischen Zellbildern in kurzer Zeit zu klassifizieren. Aufgezeigt werden Möglichkeiten zur Optimierung solcher Algorithmen mit dem Ziel, Genauigkeit und Speichergröße zu verbessern. Es wurden hierbei der Einfluss von verschiedenen Parametern auf die Performance eines Algorithmus untersucht und gegenübergestellt, verschiedene etablierte Modelle miteinander verglichen und eine Auswahl gängiger Methoden zur Modell-Optimierung getestet. Genutzt wurde die Software-Bibliothek TensorFlow, welche über die Programmiersprache Python angesprochen wird. Reale Anwendungsdaten wurden von der Firma Medipan zur Verfügung gestellt.
We consider the mission and flight planning problem for an inhomogeneous fleet of unmanned aerial vehicles (UAVs). Therein, the mission planning problem of assigning targets to a fleet of UAVs and the flight planning problem of finding optimal flight trajectories between a given set of waypoints are combined into one model and solved simultaneously. Thus, trajectories of an inhomogeneous fleet of UAVs have to be specified such that the sum of waypoint-related scores is maximized, considering technical and environmental constraints. Several aspects of an existing basic model are expanded to achieve a more detailed solution. A two-level time grid approach is presented to smooth the computed trajectories. The three-dimensional mission area can contain convex-shaped restricted airspaces and convex subareas where wind affects the flight trajectories. Furthermore, the flight dynamics are related to the mass change, due to fuel consumption, and the operating range of every UAV is altitude-dependent. A class of benchmark instances for collision avoidance is adapted and expanded to fit our model and we prove an upper bound on its objective value. Finally, the presented features and results are tested and discussed on several test instances using GUROBI as a state-of-the-art numerical solver.
Machine learning is a field that has been the object of study of many researchers around the globe during the last decades. Very often to solve machine learning challenges like classification problems for example, one needs to train an artificial neural network. To train this network a certain loss function has to be minimized. There is a ubiquitous approach to achieve this which consists of using variants of the stochastic gradient descent combined with the backpropagation algorithm. In our work, we aimed at testing a rather non-conventional scheme consisting of making use of the solvers a software called AMPL offers.