Abstract
The automation of ships will in the long run enable new mobility concepts and profitable business models both for offshore and inland vessels. However, automation in shipping can already today offer significant benefits for the environment as well as for the safety of people, infrastructure and the vessel itself. To gain these advantages, sophisticated sensors, reliable propulsion systems and intelligent algorithms are required. In this context, the project AKOON is investigating the potential of the automation of a river ferry. The ferry under investigation operates on the Rhine near Mainz in Germany and is being converted to a fully automated vessel as part of the research project. This article gives an overview of the guidance, navigation and control (GNC) modules of the automation with focus on the trajectory planning and trajectory tracking modules. The trajectory planning uses hybrid-state A* to find safe and fuel-efficient routes. The subsequent trajectory tracking is achieved by a two-level approach that uses a state-space controller and an optimization-based thrust allocation to efficiently operate the over-actuated propulsion system consisting of four cycloidal propellers. Results from a high-fidelity vessel simulator based on state-of-the-art hydrodynamic simulations show that the presented approach allows the generation of collision-free reference trajectories, that can be tracked safely even in cases of model mismatches and disturbances.
Zusammenfassung
Die Automatisierung von Schiffen lässt auf lange Sicht die Entstehung neuer Mobilitätskonzepte und Geschäftsmodelle sowohl in der See- als auch in der Binnenschifffahrt erwarten. Jedoch kann schon heute durch die Schiffsautomatisierung ein signifikanter Nutzen sowohl für die Umwelt als auch für die Sicherheit von Mensch, Maschine und Infrastruktur erzielt werden. Dieses Potential kann insbesondere mittels moderner Sensoren, zuverlässiger Antriebssysteme und intelligenter Algorithmen ausgeschöpft werden. In diesem Kontext untersucht das Projekt AKOON das Potenzial der Automatisierung einer Flussfähre. Die untersuchte Fähre, die im Rahmen des Forschungsprojekts zu einem vollautomatischen Schiff umgebaut wird, verkehrt auf dem Rhein in der Nähe der Stadt Mainz. Der vorliegende Artikel gibt einen Überblick über die dafür notwendigen GNC-Module (Guidance, Navigation and Control), wobei ein Schwerpunkt auf die Module Trajektoriengenerierung und Trajektorienfolgeregelung gelegt wird. Die Trajektoriengenerierung nutzt einen Hybrid-State A*-Ansatz, um sichere und energieeffiziente Routen zu planen. Die nachgelagerte Trajektorienfolgeregelung ist als zweistufiges Verfahren umgesetzt, das einen Zustandsregler und einen optimierungsbasierte Schuballokation verwendet, um das vorliegende überaktuierte Antriebssystem mit vier Voith-Schneider-Propellern effizient anzusteuern. Ergebnisse eines realitätsnahen Schiffssimulators auf Basis moderner hydrodynamischer Simulationen zeigen das Potenzial des vorgestellten Ansatzes, kollisionsfreie Referenztrajektorien zu generieren, die selbst im Falle von Modellfehlern und Störgrößen sicher realisiert werden können.
Funding source: Bundesministerium für Wirtschaft und Energie
Funding statement: This work was funded by the Federal Ministry for Economic Affairs and Energy of Germany in the joint research project AKOON (project number 03SX479B).
About the authors
Philipp Koschorrek is development engineer at the Research and Development department of Voith Turbo Marine. His field of work and research interests cover automatic control systems for marine vessels and optimal control of marine propulsion, with a strong focus on Dynamic Positioning simulations and control systems.
Martin Kosch received a M. Sc. degree in Energy Engineering from the Department of Mechanical Engineering at RWTH Aachen University in 2017. He is currently pursuing a Ph. D. degree at Institute of Automatic Control, where he is head of the Mobility research group. His research interest includes energy-optimal trajectory planning and model predictive control.
Maximilian Nitsch received a M. Sc. degree in Electrical Engineering, Information Technology and Computer Engineering from the Department of Electrical Engineering at RWTH Aachen University in 2018. At the same university, he is pursuing a Ph. D. degree at Institute of Automatic Control. His research interest includes navigation algorithms for maritime vehicles.
Dirk Abel is head of the Institute of Automatic Control, RWTH Aachen University. His research topics cover Automation Engineering in general with a focus on model-based predictive control, non-linear control, simulation of dynamic systems and Rapid Control Prototyping.
Dr. Dirk Jürgens studied shipbuilding at the University of Rostock from 1983-1987. Afterwards he was a scientific assistant at the Universities of Rostock and Hamburg. From 1993 to 1999, Dirk Jürgens worked in research and development at the Hamburg shipyard Blohm & Voss. Since 1999, he has been Vice President Research and Development at Voith Turbo Marine.
Acknowledgment
Our appreciation goes to the project partners, who were not actively involved in this publication: Argonav GmbH and Rheinfähre Maul GmbH. Map data copyrighted OpenStreetMap contributors and available from https://www.openstreetmap.org. Simulated current field data and measured depth data were kindly provided by the Federal Waterways Engineering and Research Institute (BAW) and Rheinfähre Maul GmbH, respectively.
References
1. Alfheim, H.L., K. Muggerud, M. Breivik, E.F. Brekke, E. Eide and Ø. Engelhardtsen. 2018. Development of a dynamic positioning system for the ReVolt model ship. IFAC-PapersOnLine, 51 (29): 116–121. 11th IFAC Conference on Control Applications in Marine Systems, Robotics, and Vehicles (CAMS) 2018.10.1016/j.ifacol.2018.09.479Search in Google Scholar
2. Safety Allianz. 2012. Shipping 1912–2012, From Titanic to Costa Concordia. Allianz Global Corporate & Specialty, Munich, Germany.Search in Google Scholar
3. Arditti, F., H. Cozijn, F.G. Van Daalen and E.A. Tannuri. 2018. Dynamic positioning simulations of a thrust allocation algorithm considering hydrodynamic interactions. IFAC-PapersOnLine 51 (29): 122–127. 11th IFAC Conference on Control Applications in Marine Systems, Robotics, and Vehicles (CAMS) 2018.10.1016/j.ifacol.2018.09.480Search in Google Scholar
4. Bitar, G., Eriksen B.-O.H., A.M. Lekkas and M. Breivik. 2021. Three-phase automatic crossing for a passenger ferry with field trials. In 2021 European Control Conference (ECC), pp. 2271–2277.10.23919/ECC54610.2021.9655139Search in Google Scholar
5. Bitar, G., A.B. Martinsen, A.M. Lekkas and M. Breivik. 2020. Two-stage optimized trajectory planning for ASVs under polygonal obstacle constraints: Theory and experiments. IEEE Access 8: 199953–199969.10.1109/ACCESS.2020.3035256Search in Google Scholar
6. Dolgov, D., S. Thrun, M. Montemerlo and J. Diebel. 2008. Practical search techniques in path planning for autonomous driving. Ann Arbor 1001 (48105): 18–80.Search in Google Scholar
7. Fossen, T.I. 2011. Handbook of Marine Craft Hydrodynamics and Motion Control. Wiley.10.1002/9781119994138Search in Google Scholar
8. DNV GL. 2018. Autonomous and remotely operated ships. Technical Report DNVGL-CG-0264, DNV GL.Search in Google Scholar
9. Groves, P.D. 2013. Principles of GNSS, inertial, and multisensor integrated navigation systems, 2nd edn. GNSS technology and application series. Artech House, Boston.Search in Google Scholar
10. Kosch, M., A. Elkhashap, P. Koschorrek, R. Zweigel and D. Abel. 2021. Hardware-in-the-loop trajectory tracking and collision avoidance of automated inland vessels using model predictive control. In 2021 European Control Conference (ECC), pp. 2251–2256.10.23919/ECC54610.2021.9655183Search in Google Scholar
11. Kosch, M., R. Zweigel and D. Abel. 2020. Online estimation of near-surface water current fields using horizontally mounted adcps on inland vessels. In 2020 European Navigation Conference (ENC), pp. 1–10.10.23919/ENC48637.2020.9317463Search in Google Scholar
12. Koschorrek, P. and M. Kosch. 2021. An Approach to QP-based Thrust Allocation considering Inflow. In Control Applications in Marine Systems, Robotics, and Vehicles (CAMS) 2021.10.1016/j.ifacol.2021.10.083Search in Google Scholar
13. Koschorrek, P., M. Palm and T. Jeinsch. 2017. A Dynamic Allocation Strategy for Voith Schneider Propeller. IFAC-PapersOnLine, 50 (1): 1127–1132.10.1016/j.ifacol.2017.08.395Search in Google Scholar
14. Laub, A. 1979. A Schur method for solving algebraic Riccati equations. IEEE Trans. Automat. Control 24 (6): 913–921.10.1109/CDC.1978.267893Search in Google Scholar
15. Nitsch, M., J.-J. Gehrt, R. Zweigel and D. Abel. 2021. Tightly Coupled INS-GNSS Navigation Filter for the Automation of a River Ferry. In Control Applications in Marine Systems, Robotics, and Vehicles (CAMS) 2021.10.1016/j.ifacol.2021.10.085Search in Google Scholar
16. Rivkin, B. 2021. Unmanned ships: Navigation and more. Gyroscopy Navig. 12: 96–108.10.1134/S2075108721010090Search in Google Scholar
17. Ruth, E.. 2008. Propulsion control and thrust allocation on marine vessels. Doctoral thesis, Fakultet for ingeniørvitenskap og teknologi, NTNU.Search in Google Scholar
18. Sontag, E. 2013. Mathematical Control Theory: Deterministic Finite Dimensional Systems. Texts in Applied Mathematics. Springer, New York.Search in Google Scholar
19. Thyri, E.H., M. Breivik and A.M. Lekkas. 2020. A path-velocity decomposition approach to collision avoidance for autonomous passenger ferries in confined waters. IFAC-PapersOnLine, 53 (2): 14628–14635, 21st IFAC World Congress.10.1016/j.ifacol.2020.12.1472Search in Google Scholar
20. Wan, E.A. and R. Van Der Merwe. 2000. The unscented Kalman filter for nonlinear estimation. In IEEE 2000 Adaptive Systems for Signal Processing, Communications, and Control Symposium (Cat. No. 00EX373), pp. 153–158.10.1109/ASSPCC.2000.882463Search in Google Scholar
© 2022 Walter de Gruyter GmbH, Berlin/Boston
