J.R. Marx et al. 
Channels are sufficiently well tuned to reflect the physical capabilities of 
che propulsion system, ensuring that the control performance remains 
nigh in practice and disturbances are rejected robustly. 
6. Conclusions and future works 
The contribution described first successful tests of automatic, coop- 
erative maneuvering of three surface vehicles within confined waters. 
The results represent an important step in the development of succes- 
3ive maneuver automation, since not only a single ship was maneuvered, 
but the automatic collision avoidance could be implemented for several 
ships in real environment. Based on expert nautical knowledge and com- 
non navigation tools, the assistance systems and the entire test design 
are characterized by a high level of transparency in all automatic func- 
donalities to ensure responsible handling for nautical personnel. 
The following applicable rules in the port, an adaptation of the COL- 
REGs, have been incorporated into the optimization in the calculation 
of the evasion manoeuvres, safe speed, safe distance to fixed objects and 
other vessels, collision avoidance in head-on and crossing situations. The 
adaptive MPC has proven to be well suited for trajectory tracking for all 
‘hree vehicles and in combination with XN or XYN allocation depending 
an the speed range. In this phase of development, it was a good solution 
‘hat trajectory optimization was based on the entire initial trajectories 
and fixed parameterization for distances to the port infrastructure and 
to other ships. But this static methodology has a low tolerance to local 
disturbances in the environment, faults in the automation system or pro- 
z:edure. However, for more flexible methods that execute recalculations 
ınder real-time requirements, high-performance computing technology 
ıs required. The presented approach is well tailored for ferry connections 
with two or three landings, so that the environment, the usual weather 
:onditions as well as the routes and maneuvers from pier to pier are 
<nown. Dynamic obstacles that are not in the network and whose eva- 
sion trajectories cannot therefore be automatically adjusted must first 
be detected in real time using an array of optical sensors, such as lidar 
and radar. A system setup for tracking dimensions, states and classes 
af dynamic obstacles with the new technology of frequency modulated 
zontinuous wave LiDARs was given in Karez et al. (2025). Subsequently, 
che successive motion planning method, based on the contributions in 
Damerius and Jeinsch (2022) and in Damerius et al. (2023), can be used 
co locally adapt the evasion paths and trajectories of the network vehi- 
cles. In order to modify the approach for major disturbances from the 
environment, such as higher wind speeds, an automatic scenario su- 
zervisor is required, which then only allows operations in DP mode. 
However, for safety reasons, there are limits to this, especially in nar- 
:ow fairways, so that the vessel traffic service will prohibit automatic 
maneuvering above certain wind speeds. 
The tests have shown a small selection of possible scenarios in the 
port. In future developments, the remaining basic rules identified in Sec- 
ion 2.2 should also be taken into account for automatic maneuvering 
ın the port. It is particularly important to automatically recognize the 
applicable rule depending on the type of traffic participants or other ob- 
iects, such as buoys, in the environment. As entries and departures from 
‘he port are organized by the vessel traffic service, it is to be integrated 
into the central communication system. 
With regard to the control approach with adaptive MPC, the aim is 
to make development work more effective by automatically identifying 
parameters for the model, controller and allocation on the basis of mea- 
surement data. In future tests, more flexible methods for calculating the 
evasion trajectories are to be used, allowing shorter sections to be recal- 
culated if conditions in the environment have changed or errors have 
occurred within the automation system. 
CRediT authorship contribution statement 
Johannes R. Marx: Writing — original draft, Visualization, Vali- 
dation, Software, Methodology: Agnes U. Schubert: Writing — origi- 
Ocean Engineering 343 (2026) 123388 
nal draft, Visualization, Methodology, Conceptualization; Nick Eisen- 
blätter: Validation, Software, Methodology; Robert Damerius: Su- 
pervision, Software, Resources, Methodology, Investigation, Concep- 
ualization; Michael Gluch: Visualization, Software, Methodology; 
Sapt. Michael Quandt: Supervision, Methodology, Conceptualization; 
Torsten Jeinsch: Supervision, Project administration, Funding acquisi 
von. 
Fundings 
The work was funded by the German Federal Ministry for Economic 
Affairs and and Climate Action (BMWK) and the German Federal Mar- 
time and Hydrographic Agency (BSH) as the owner of research vessel 
DENEB as well as supported by the DLR Space Administration under the 
registration number 50NA2304B. 
Declaration of competing interest 
The authors declare that they have no known competing financial 
.nterests or personal relationships that could have appeared to influence 
‚he work reported in this paper. 
Acknowledgments 
Special thanks go to the crew of the research vessel DENEB and the 
‚esponsible staff from the Federal Maritime and Hydrographic Agency 
BSH) for their openness towards the developments and their tireless 
support during the trials as well as the fruitful discussions on ship han- 
dling. 
References 
3aldauf, M., Benedict, K., Fischer, S., Gehrke, M., Finger, G., Gluch, M., Kirchhoff, M., 
1024. Manoeuvring prediction technologies in ship handling for training and use on- 
»oard. Universitat Polit&cnica de Catalunya. Iniciativa Digital Polit&cnica. 10. https: 
‚/doi.org/10.5821/mt.12828 
zetts, J.T., 1998. A survey of numerical methods for trajectory optimization. J. Guid. 
Control Dyn. 21 (2), 193-207. https://doi.org/10.2514/2.4231 
3lanke, M., Hansen, N.P., Dittmann, K., Enevoldsen, T.T., Dagdilelis, D., Schöller, F.E.T.S., 
>lenge-Feidenhans’l, M.K., Becktor, J., Papageorgiou, D., Galeazzi, R., 2024. Green- 
‘opper: the danish spearhead towards autonomous waterborne mobility. J. Phys. Conf. 
Ser. 2867 (1), 012035. https: //doi.org/10.1088/1742-6596/2867/1/012035 
Brekke, E.F., Eide, E., Eriksen, B.O.H., Wilthil, E.F., Breivik, M., Skjellaug, E., Helgesen, 
A.K., Lekkas, A.M., Martinsen, A.B., Thyri, E.H., Torben, T., Veitch, E., Alsos, O.A., 
Johansen, T.A., 2022. Milliampere: an autonomous ferry prototype. J. Phys. 2311 (1). 
attps://doi.org/10.1088/1742-6596/2311/1/012029 
SH, 2024. German Traffic Regulations for Navigable Maritime Waterways, English 
version of the Seeschifffahrtsstraßen-Ordnung. Federal Maritime and Hydrographic 
Agency (BSH), Referat N2. https://www.bsh.de/DE/PUBLIKATIONEN/_Anlagen/ 
Jownloads/Nautik_und_Schifffahrt/Seehandbuecher ueberregional/SeeschStrO_engl 
ATML. 
7amacho, E.F., Alba, C.B., 2013. Model Predictive Control. Springer Science & Business 
Media. 
Chen, L., Shan, Y., Tian, W., Li, B., Cao, D., 2018. A fast and efficient double-tree RRT* 
"ike sampling-based planner applying on mobile robotic systems. IEEE/ASME Trans. 
Mechatron. 23 (6), 2568-2578. https://doi.org/10.1109/TMECH.2018.2821767 
"ıassNK, 2020. Guidelines for automated/autonomous operation on ships (ver.1.0) - de- 
sign development, installation and operation of automated operation systems/remote 
operation systems. ClassNK NIPPON KAIJI KYOKAI. 
Jamerius, R., Hahn, T., Karez, I., Schubert, A., Kolewe, B., Jeinsch, T., 2024. Guidance, 
ıavigation and control of couplable unmanned surface vehicles. In: OCEANS 2024 
Singapore. https://doi.org/10.1109/0CEANS51537.2024.10682217 
lamerius, R., Jeinsch, T., 2022. Real-time path planning for fully actuated autonomous 
surface vehicles. In: 2022 30th Mediterranean Conference on Control and Automation 
(MED), pp. 508-513. https://doi.org/10.1109/MED54222.2022.9837178 
Jamerius, R., Marx, J.R., Jeinsch, T., 2023. Fast trajectory generation on a path using 
“eedback linearization. IFAC-PapersOnLine 56 (2), 10990-10995. 22nd IFAC World 
Congress. https://doi.org/10.1016/j.ifacol.2023.10.796 
har, A., Bhasin, S., 2018. Adaptive MPC for uncertain discrete-time 1ti mimo systems with 
ncremental input constraints. IFAC-PapersOnLine 51 (1), 329-334. 5th IFAC Confer- 
ence on Advances in Control and Optimization of Dynamical Systems ACODS 2018. 
nttps://doi.org/10.1016/j.ifacol.2018.05.040 
Jittmann, K., Blanke, M., 2022. Risk mitigation by design of autonomous maritime au- 
tomation systems. at - Automatisierungstechnik 70 (5), 469-481. https: //doi.org/10. 
1515/aute-2021-0151