"Why This Avoidance Maneuver?" Contrastive Explanations in Human-Supervised Maritime Autonomous Navigation

Joel Jose; Andreas Madsen; Andreas Brandsæter; Tor A. Johansen; Erlend M. Coates

cs.AIApr 9, 2026

"Why This Avoidance Maneuver?" Contrastive Explanations in Human-Supervised Maritime Autonomous Navigation

Authors: Joel Jose, Andreas Madsen, Andreas Brandsæter, Tor A. Johansen, Erlend M. Coates

Organizations: Department of ICT and Natural Sciences, Norwegian University of Science and Technology (NTNU), Ålesund, Norway. · Department of Ocean Operations and Civil Engineering, Norwegian University of Science and Technology (NTNU), Ålesund, Norway. · Department of Science and Mathematics, Volda University College (HVO), Volda, Norway. · Department of Engineering Cybernetics, Norwegian University of Science and Technology (NTNU), Trondheim, Norway.

Abstract

Automated maritime collision avoidance will rely on human supervision for the foreseeable future. This necessitates transparency into how the system perceives a scenario and plans a maneuver. However, the causal logic behind avoidance maneuvers is often complex and difficult to convey to a navigator. This paper explores how to explain these factors in a selective, understandable manner for supervisors with a nautical background. We propose a method for generating contrastive explanations, which provide human-centric insights by comparing a system's proposed solution against relevant alternatives. To evaluate this, we developed a framework that uses visual and textual cues to highlight key objectives from a state-of-the-art collision avoidance system. An exploratory user study with four experienced marine officers suggests that contrastive explanations support the understanding of the system's objectives. However, our findings also reveal that while these explanations are highly valuable in complex multi-vessel encounters, they can increase cognitive workload, suggesting that future maritime interfaces may benefit most from demand-driven or scenario-specific explanation strategies.

Figures & tables

Fig. 1 : Concept illustration. Contrastive explanations for a collision avoidance maneuver visualized in an Electronic Chart Display and Information System (ECDIS) (Explanations visualized on top of a figure from [ 9 ] ). Here, a proposed maneuver (red) is compared with the original route (black), and an alternative route (blue) based on an objective of keeping a safe closest point of approach (CPA).

Fig. 2 : The adapted PSW model from [ 24 ] . We believe that comparing alternative routes with contrastive explanations fits best in supporting supervision of the action planning stage (highlighted in red).

Cost	Description
$f_{\mathrm{DO}}$	Collision risk with dynamic obstacles
$f_{\mathrm{CLRG}},f_{\mathrm{{Trans}}}$	COLREG compliance and transition costs
$f_{U^{\mathrm{ref}}},f_{\chi^{\mathrm{ref}}}$	Deviation from autopilot speed/course references
$f_{\Delta U},f_{\Delta\chi}$	Variation from previous speed/course states

TABLE I : SB-MPC Cost Function Components.

Fig. 3 : Contrastive Explanations Framework. The costs are extracted from the cost function for the solution and foil trajectories and compared to generate a contrastive explanation, which is visualized on an interface to the supervisor.

Characteristic	Constraint ( $C$ )
Reduced speed	$U^{\tau^{\mathrm{alt}}}<U^{\tau^{*}}$
Port Turn	$\chi^{\tau^{\mathrm{alt}}}<0$
Starboard Turn	$\chi^{\tau^{\mathrm{alt}}}>0$
Closer to original route	$\|\chi^{\tau^{\mathrm{alt}}}\|<\|\chi^{\tau^{*}}\|$
Farther from original route	$\|\chi^{\tau^{\mathrm{alt}}}\|>\|\chi^{\tau^{*}}\|$

TABLE II : Choices of the characteristic and associated constraint for the alternative.

Fig. 4 : The Visualization Interface used in this study. The design draws inspiration from the transparency layers in [ 6 ] and the explanations depicted in Fig. 1 . (Left) Map View illustrates the proposed fact (white), alternative (orange), with speed-proportional spacing of dots, and the nominal path (faint red). Obstacle predictions are shown as dots of the corresponding ship color. (Top right) Action table for all three trajectories. (Center right) Contrastive explanations comparing the fact against the foils. (Bottom right) Target vessel information

Cost	Associated Cost Measure
$f_{DO}$	CPA distance to Dynamic Obstacle (DO)
$f_{CLRG}$	COLREG rule number
$f_{Trans}$	Transition behavior
$f_{U_{ref}}$	Speed offset
$f_{\chi_{ref}}$	Course offset

TABLE III : Mapping of Costs to Cost Measures. (CPA refers to the closest point of approach between the ownship and the obstacle along their corresponding trajectories)

Panel of Figure 8 — Fig. 5 : Snapshots of traffic scenarios utilized in the user study, with the alternative maneuvers and associated explanations. Top right: Single-vessel, Head-on Encounter; Top left: Single-vessel, Crossing Encounter; Bottom left: Multi-vessel, two head-on encounters; Bottom Right: Multi-vessel, one crossing Give-Way, one crossing Stand-on.

Trial	P1	P2	P3	P4
Trial 1	Accept	Accept	Accept	Accept
Trial 2	Accept	Accept	Accept	Accept
Trial 3	Decline	Decline	Decline	Accept
Trial 4	Accept	Decline	Accept	Accept

TABLE IV : Participant responses across trials. A majority of participants declined the proposed maneuver in Trial 3 due to a need to communicate with other vessels.

Explainable Motion Planning

Explainable motion planning makes robot and vehicle trajectories understandable, auditable, and safer to supervise.

See all 17 papers

Explainable Failure Prediction and Prevention in Maritime
Oct 6, 2026Dionisis Kalogeropoulos, Georgia Sovatzidi, Panagiotis G. Kalozoumis +1Predictive Maintenance Explainable Artificial Intelligence
Dept. of Computer Science and Biomedical Informatics, University of Thessaly
Understanding Operator Attitudes Toward AI-Supported Decision Making in Maritime Operations
Sep 10, 2026Doreen Jirak, Armeen Saroukanoff, Dirk van RooyAI-Assisted Decision Making Explainable Artificial Intelligence
University of Antwerp, Paardenmarkt 94, 2000 Antwerp, Belgium · Antwerp Maritime Academy (AMA), Noordkasteel Oost 6, 2030 Antwerp, Belgium
Explainable Neuro-Fuzzy Prediction for Trustworthy Decision-Making in Maritime
Sep 21, 2026Dionisis Kalogeropoulos, Georgia Sovatzidi, Dimitris K. IakovidisExplainable Artificial Intelligence Predictive Maintenance
Dept of Computer Science and Biomedical Informatics University of Thessaly Lamia, Greece

Oct 6, 2026J/K move · Enter open · S save

Explainable Failure Prediction and Prevention in Maritime

Dionisis Kalogeropoulos, Georgia Sovatzidi, Panagiotis G. Kalozoumis, Dimitris K. Iakovidis

Dept. of Computer Science and Biomedical Informatics, University of Thessaly

Predictive Maintenance Explainable Artificial Intelligence

Open paper AskPDF arXiv

Maritime systems operate in highly dynamic environments where unexpected equipment failures can compromise safety, reliability, and operational efficiency. Recent advances in artificial intelligence (AI), machine learning, digital twins, and predictive maintenance enable proactive failure prediction and prevention. However, ensuring trustworthy and explainable decision-making remains a major challenge in safety-critical maritime applications. This chapter reviews key AI technologies required for explainable failure prediction and prevention in maritime systems and presents a conceptual architecture capable of supporting autonomous or human-in-the-loop corrective actions. This architecture integrates data acquisition, time-series forecasting, anomaly detection, risk assessment, decision-making, and explainable AI into a closed-loop framework. With reference to the architectural components, a review and discussion of relevant maritime studies is performed, outlining their methods, advantages, and limitations. Furthermore, it highlights current challenges, including uncertainty and robustness, model generalization, explainability, limited availability of maritime datasets, and operational deployment, and identifies future research directions toward trustworthy AI-assisted maritime decision-making.

Abstract

Explainable Motion Planning

Explore similar work

Explainable Failure Prediction and Prevention in Maritime

Understanding Operator Attitudes Toward AI-Supported Decision Making in Maritime Operations

Explainable Neuro-Fuzzy Prediction for Trustworthy Decision-Making in Maritime