• Approximating a deep reinforcement learning docking agent using linear model trees 

      Gjærum, Vilde Benoni; Rørvik, Ella-Lovise H.; Lekkas, Anastasios M. (Journal article, 2021)
      Deep reinforcement learning has led to numerous notable results in robotics. However, deep neural networks (DNNs) are unintuitive, which makes it difficult to understand their predictions and strongly limits their potential ...
    • Autonomous navigation along power lines using monocular camera 

      Gjærum, Vilde Benoni (Master thesis, 2019)
      Denne rapporten tar for seg problemet med autonom deteksjon og navigasjon langs kraftlinjer i landlige omgivelser ved bruk av UAV, som et steg mot å oppnå fullkommen autonomitet i arbeidet med inspeksjon av kraftlinjer, ...
    • Explainable AI methods on a deep reinforcement learning agent for automatic docking 

      Løver, Jakob; Gjærum, Vilde Benoni; Lekkas, Anastasios M. (Peer reviewed; Journal article, 2021)
      Artifical neural networks (ANNs) have made their way into marine robotics in the last years, where they are used in control and perception systems, to name a few examples. At the same time, the black-box nature of ANNs is ...
    • Robust Reasoning for Autonomous Cyber-Physical Systems in Dynamic Environments 

      Håkansson, Anne; Saad, Aya; Sadanandan Anand, Akhil; Gjærum, Vilde Benoni; Robinson, Haakon; Seel, Katrine (Peer reviewed; Journal article, 2021)
      Autonomous cyber-physical systems, CPS, in dynamic environments must work impeccably. The cyber-physical systems must handle tasks consistently and trustworthily, i.e., with a robust behavior. Robust systems, in general, ...
    • Safe Learning for Control using Control Lyapunov Functions and Control Barrier Functions: A Review 

      Sadanandan Anand, Akhil; Seel, Katrine; Gjærum, Vilde Benoni; Håkansson, Anne; Robinson, Haakon; Saad, Aya (Peer reviewed; Journal article, 2021)
      Real-world autonomous systems are often controlled using conventional model-based control methods. But if accurate models of a system are not available, these methods may be unsuitable. For many safety-critical systems, ...