Browsing NTNU Open by Author "Larsen, Thomas Nakken"
Now showing items 1-8 of 8
-
Comparing Deep Reinforcement Learning Algorithms’ Ability to Safely Navigate Challenging Waters
Larsen, Thomas Nakken; Teigen, Halvor Ødegård; Laache, Torkel; Varagnolo, Damiano; Rasheed, Adil (Peer reviewed; Journal article, 2021)Reinforcement Learning (RL) controllers have proved to effectively tackle the dual objectives of path following and collision avoidance. However, finding which RL algorithm setup optimally trades off these two tasks is not ... -
Merging Classical Control and Deep Reinforcement Learning for Dynamic Collision Avoidance for a Quadcopter
Carlsen, Ørjan (Master thesis, 2023)Selvstyrte ubemannede luftfartøy (UAVs, engelsk: Unmanned Aerial Vehicles), som eksempelvis kvadrokoptere, kan forbedre effektiviteten ved leveranser, utføre risikofylte inspeksjoner av strukturelle eiendeler, eller til ... -
On the applicability of a perceptually driven generative-adversarial framework for super-resolution of wind fields in complex terrain
Larsen, Thomas Nakken (Master thesis, 2020)Det har nylig blitt gjort store fremskritt innenfor datasyn for konstruksjon av høyoppløste bilder fra referansebilder med lav oppløsning ved å benytte høy-\\dimensjonale aktiveringer fra ``feature-extractors" for å danne ... -
Path Following and Collision Avoidance for Quadcopters using Deep Reinforcement Learning
Sundøen, Ludvig Løken (Master thesis, 2022)Klassiske kontrollmetoder avhenger av nøyaktige modeller. Slike modeller eksisterer ikke alltid for komplekse systemer, og kontroll er begrenset til enkle oppgaver på lavt nivå. Modellfrie Reinforcement Learning (RL) ... -
Path-following and Collision Avoidance using World Models
Brudeli, Kristian (Master thesis, 2023)Denne masteroppgaven presenterer en applikasjon av den modellbaserte forsterkningslæringsagen- ten DreamerV2 for banefølging og kollisjonsunngåelse med autonome overflatsfartøy. Fartøyet lærer gjennom interaksjon med ... -
A receding horizon approach for curriculum management in higher education
Larsen, Thomas Nakken; Busetto, Riccardo; Varagnolo, Damiano; Formentin, Simone; Rasheed, Adil (Peer reviewed; Journal article, 2022)We propose a dynamic approach for curriculum management in university programs, i.e., for deciding which teaching and learning activities should be performed and in which order, as classes are being executed, to better aid ... -
Risk-based Convolutional Perception Models for Collision Avoidance in Autonomous Marine Surface Vessels using Deep Reinforcement Learning
Larsen, Thomas Nakken; Hansen, Hannah; Rasheed, Adil (Chapter, 2023)In this work, we propose a novel policy network architecture for model-free Reinforcement Learning (RL)-based path-following and collision avoidance in marine surface vessels. By applying convolutional neural networks ... -
Risk-based implementation of COLREGs for autonomous surface vehicles using deep reinforcement learning
Heiberg, Amalie; Larsen, Thomas Nakken; Meyer, Eivind; Rasheed, Adil; San, Omer; Varagnolo, Damiano (Peer reviewed; Journal article, 2022)Autonomous systems are becoming ubiquitous and gaining momentum within the marine sector. Since the electrification of transport is happening simultaneously, autonomous marine vessels can reduce environmental impact, lower ...