Blar i NTNU Open på forfatter "Lekkas, Anastasios M."
-
A Voronoi-diagram-based dynamic path-planning system for underactuated marine vessels
Candeloro, Mauro; Lekkas, Anastasios M.; Sørensen, Asgeir Johan (Journal article; Peer reviewed, 2017)The main contribution of this paper is the development of a rapid, dynamic path-planning system for 3-DOF marine surface vessels navigating in environments where other marine vehicles might be operating too. The method is ... -
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 docking using direct optimal control
Martinsen, Andreas Bell; Lekkas, Anastasios M.; Gros, Sebastien (Journal article; Peer reviewed, 2019)We propose a method for performing autonomous docking of marine vessels using numerical optimal control. The task is framed as a dynamic positioning problem, with the addition of spatial constraints that ensure collision ... -
Causal versus Marginal Shapley Values for Robotic Lever Manipulation Controlled using Deep Reinforcement Learning
Remman, Sindre Benjamin; Strumke, Inga; Lekkas, Anastasios M. (Peer reviewed; Journal article, 2022)We investigate the effect of including application knowledge about a robotic system states’ causal relations when generating explanations of deep neural network policies. To this end, we compare two methods from explainable ... -
COLREGs-aware automatic docking for autonomous surface vessels
Nordhus, Henrik Rohde (Master thesis, 2023)I en æra hvor autonome systemer i økende grad former industrier og hverdagsliv, har området for maritim navigasjon opplevd en betydelig transformasjon. Denne masteroppgaven viser fremskrittene som er gjort i dokkingsprosedyrer, ... -
Curved Path Following with Deep Reinforcement Learning: Results from Three Vessel Models
Martinsen, Andreas Bell; Lekkas, Anastasios M. (Chapter, 2018)This paper proposes a methodology for solving the curved path following problem for underactuated vehicles under unknown ocean current influence using deep reinforcement learning. Three dynamic models of high complexity ... -
Dynamic Wireless Charging of Autonomous Vehicles: Small-scale demonstration of inductive power transfer as an enabling technology for self-sufficient energy supply
Guidi, Giuseppe; Lekkas, Anastasios M.; Stranden, Jon Eivind; Suul, Jon Are Wold (Peer reviewed; Journal article, 2020)Recent developments toward selfdriving cars combined with technology for wireless inductive power transfer can enable electric vehicles that are fully autonomous with respect to operation and energy requirements. Furthermore, ... -
Energy-Optimized Path Planning for Autonomous Ferries
Bitar, Glenn Ivan; Breivik, Morten; Lekkas, Anastasios M. (Journal article, 2018)The topic of energy-optimized path planning using pseudospectral optimal control is considered. An optimal control problem (OCP) is formulated to produce an energy-optimized path between two points among static obstacles. ... -
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 ... -
Explaining a deep reinforcement learning docking agent using linear model trees with user adapted visualization
Gjærum, Vilde Benoni; Strumke, Inga; Alsos, Ole Andreas; Lekkas, Anastasios M. (Peer reviewed; Journal article, 2021)Deep neural networks (DNNs) can be useful within the marine robotics field, but their utility value is restricted by their black-box nature. Explainable artificial intelligence methods attempt to understand how such ... -
Guidance and Path-Planning Systems for Autonomous Vehicles
Lekkas, Anastasios M. (Doktoravhandlinger ved NTNU, 1503-8181; 2014:126, Doctoral thesis, 2014)This thesis is concerned with two interconnected and very important problems regarding the autonomy of vehicles, namely, path planning and guidance. By adopting a modular approach, path planning and guidance can be viewed ... -
Hybrid Collision Avoidance for ASVs Compliant with COLREGs Rules 8 and 13-17
Eriksen, Bjørn-Olav Holtung; Bitar, Glenn Ivan; Breivik, Morten; Lekkas, Anastasios M. (Journal article; Peer reviewed, 2020)This paper presents a three-layered hybrid collision avoidance (COLAV) system for autonomous surface vehicles, compliant with rules 8 and 13–17 of the International Regulations for Preventing Collisions at Sea (COLREGs). ... -
Identification of Failure Modes in the Collision Avoidance System of an Autonomous Ferry using Adaptive Stress Testing
Hjelmeland, Hanna Waage; Eriksen, Bjørn-Olav Holtung; Mengshoel, Ole Jakob; Lekkas, Anastasios M. (Peer reviewed; Journal article, 2022)As complex autonomous systems emerge in the maritime sector, measures must be taken in order to ensure thorough safety assessment. Real world testing can be costly and potentially dangerous, and therefore there is a need ... -
milliAmpere: An Autonomous Ferry Prototype
Brekke, Edmund Førland; Eide, Egil; Eriksen, Bjørn-Olav Holtung; Wilthil, Erik Falmår; Breivik, Morten; Skjellaug, Even; Helgesen, Øystein Kaarstad; Lekkas, Anastasios M.; Martinsen, Andreas Bell; Thyri, Emil Hjelseth; Torben, Tobias Valentin Rye; Veitch, Erik Aleksander; Alsos, Ole Andreas; Johansen, Tor Arne (Peer reviewed; Journal article, 2022)In this paper, we summarize the experiences with the autonomous passenger ferry development prototype milliAmpere, which has been used as a test platform in several research projects at the Norwegian University of Science ... -
Online Path Planning for Surface Vehicles Exposed to Unknown Ocean Currents Using Pseudospectral Optimal Control
Lekkas, Anastasios M.; Roald, Ann Louise; Breivik, Morten (Journal article; Peer reviewed, 2016)This paper investigates the feasibility of using pseudospectral (PS) optimal control in real-time path planning for marine surface vehicles in environments where both obstacles and unknown disturbances are present. In ... -
Optimization-Based Automatic Docking and Berthing of ASVs Using Exteroceptive Sensors: Theory and Experiments
Martinsen, Andreas Bell; Bitar, Glenn Ivan; Lekkas, Anastasios M.; Gros, Sebastien (Peer reviewed; Journal article, 2020)Docking of autonomous surface vehicles (ASVs) involves intricate maneuvering at low speeds under the influence of unknown environmental forces, and is often a challenging operation even for experienced helmsmen. In this ... -
A Path-Velocity Decomposition Approach to Collision Avoidance for Autonomous Passenger Ferries in Confined Waters
Thyri, Emil Hjelseth; Breivik, Morten; Lekkas, Anastasios M. (Peer reviewed; Journal article, 2020)A deliberate collision avoidance approach for autonomous surface vehicles operating in confined waters with high traffic is presented. The approach focuses on dynamic obstacles, by assuming a predefined set of paths that ... -
Reinforcement Learning-Based Tracking Control of USVs in Varying Operational Conditions
Martinsen, Andreas Bell; Lekkas, Anastasios M.; Gros, Sebastien (Peer reviewed; Journal article, 2020)We present a reinforcement learning-based (RL) control scheme for trajectory tracking of fully-actuated surface vessels. The proposed method learns online both a model-based feedforward controller, as well an optimizing ... -
Robotic Lever Manipulation using Hindsight Experience Replay and Shapley Additive Explanations
Remman, Sindre Benjamin; Lekkas, Anastasios M. (Peer reviewed; Journal article, 2021)This paper deals with robotic lever control using Explainable Deep Reinforcement Learning. First, we train a policy by using the Deep Deterministic Policy Gradient algorithm and the Hindsight Experience Replay technique, ... -
A Rudimentary Mission Planning System for Marine Autonomous Surface Ships
Hinostroza, Miguel; Lekkas, Anastasios M. (Peer reviewed; Journal article, 2022)Maritime autonomous surface ships (MASS) have become a topic of intensive research in academia and industry, with focus areas such as path following and tracking, low-level control, collision avoidance, situational awareness, ...