Rotterdam, 25-09-2023 (updated on, 19-05-2026) – VSTEP and Captain AI are partnering to transform maritime training through advanced simulator data. With NAUTIS, VSTEP enables the cost-effective generation of datasets essential for training an AI model capable of navigating an autonomous vessel. This collaboration focuses on enhanced computer vision, dynamic state sampling and AI-driven traffic simulation, underscoring VSTEP’s commitment to innovation at the frontier of autonomous shipping.
Overview of the Project
VSTEP is developing a system that enables training in a virtual world populated with autonomous ship models. This hybrid environment, where real human captains and AI-driven vessels operate side by side, represents a significant development in how the maritime industry prepares for the future of navigation.
The project aligns with multiple points outlined in the Kennis- en Innovatie-Agenda from the Dutch Topsectors initiative, aimed at advancing the Netherlands as a competitive maritime nation. It directly addresses four of the most pressing challenges the maritime industry currently faces:
- Growing shortage of trained and capable staff
- Safety
- Sustainability and carbon reduction
- Cost reduction
VSTEP takes great pride and enthusiasm in its contribution towards alleviating the pressing issues within the maritime industry and enabling learning by simulation.
Collaboration Between VSTEP and Captain AI
VSTEP and Captain AI are using NAUTIS to train an AI autopilot model through the generation of synthetic data. NAUTIS enables the creation of a wide range of scenarios and datasets, enhancing both model quality and training efficiency. Through the NAUTIS API, scenarios can be generated procedurally and automatically, without requiring human involvement in dataset creation.
Because NAUTIS adheres to standard maritime protocols, the autonomous ship model can connect to the simulator in the same way it would interact with a real vessel. When the model is later deployed on an actual ship and receives live data, it can apply the patterns learned from synthetic data to real-world conditions, producing consistent and reliable outcomes. This approach significantly reduces development costs while maintaining a safe training environment, even when simulating collisions or accidents.
Diving into VSTEP and Captain AI’s Collaboration
VSTEP and Captain AI’s collaboration have three key aspects:
1. Enhanced computer vision capabilities
Beyond NAUTIS’s existing ability to measure data such as velocity, depth, acceleration and distance between vessels, VSTEP is developing a video signal feature. This enables the acquisition of additional visual data and facilitates information fusion, cross-referencing video with AIS and radar datasets for further validation and analysis.
2. Dynamic state sampling
NAUTIS is being optimised to eliminate warm-up time in simulation sessions. Fast-forward functionality ensures both visual and hydrodynamic accuracy are maintained, enabling the generation of multiple datasets within a significantly reduced timeframe. This accelerates the development cycle for autonomous vessel AI models.
3. AI traffic integration
Autonomous ship models integrated into NAUTIS can self-train by observing and interacting with other vessels in simulated traffic scenarios. These models can set navigation goals, determine optimal routes and avoid collisions in real time. For human trainees, this creates a more realistic training environment that prepares them for the hybrid future of autonomous shipping, where AI-controlled and human-controlled vessels share the same waterways.
“Working with VSTEP has been transformative for maritime training for humans and AI algorithms alike. Our ongoing collaboration with VSTEP aims to enhance simulator technology to offer more realistic training experiences in the future. The results of our collaboration extend beyond traditional methods, setting new ambitions for the industry. Together, we’re better equipping maritime professionals and ships for the challenges ahead.”
– Gerard Kruisheer, Co-Founder & CTO at Captain AI
The collaboration with Captain AI plays a role in VSTEP’s development of a distinct module featuring innovative content designed specifically for autonomous ships. This advancement gains further momentum through the integration of Captain AI’s AI auto pilot, infusing the simulations with genuine traffic scenarios to enhance the overall sense of realism. This further enhances the educational experience for sailors while also underscoring the invaluable role of artificial intelligence in facilitating this process.
“As a progressive organization, we acknowledge the immense potential of autonomous ships in the near future. Embracing and supporting this transformative technology aligns with our commitment to innovation and progress. Our active participation in this initiative demonstrates our determination to be at the forefront of autonomous ship development and contribute to the advancement of cutting-edge technology in the maritime industry. By supporting autonomous ship systems, we aim to lead the way in shaping the future of maritime transportation, creating safer, more efficient and sustainable solutions for the industry.“
– Maarten Meulemans, Software Engineer Director at VSTEP
VSTEP’s dedication to creating safer, more efficient and sustainable solutions for the industry emphasizes their mission of enabling learning through simulation. Guided by innovation and the pursuit of effective training through simulation, VSTEP positions itself as a significant contributor to the emergence of autonomous vessels.
VSTEP x Captain AI: Project overview
| Project pillar | Technical approach | Training benefit |
|---|---|---|
| Computer vision | Video signal integrated with AIS and radar data | Richer, more accurate datasets for AI model training |
| Dynamic state sampling | Fast-forward simulation with hydrodynamic accuracy | Faster generation of diverse training scenarios |
| AI traffic integration | Autonomous vessels self-train in simulated traffic | Realistic mixed-traffic environments for human and AI trainees |
| NAUTIS API | Procedural scenario generation without human input | Scalable, cost-efficient dataset creation |
| Protocol compatibility | NAUTIS adheres to standard maritime protocols | AI model transfers learning from simulation to real vessel seamlessly |
| Safety | Collisions and accidents can be simulated without risk | Full scenario range available without operational hazard |
Interested in exploring a collaboration with VSTEP on simulation-driven innovation? Get in touch with the VSTEP team to discuss the possibilities.
About VSTEP
VSTEP creates innovative simulation software for professional organizations worldwide. We are a team of 60 explorers who are always up for challenges. We are not constrained by standards and are flexible in our approach to fulfilling the mission: enable learning by simulation.
We are all dedicated to the same vision: ensure that at least one-third of the maritime industry has obtained their skills through our simulation technology. Hence, we strive to make a great contribution to creating a more sustainable world. We invest in continuous R&D of our software and work passionately as a team to get there.
What is autonomous vessel simulation and why does it matter for maritime training?
Autonomous vessel simulation uses advanced maritime simulators to train AI models that can navigate ships without human input. It matters because the maritime industry is moving toward hybrid operations where autonomous and human-controlled vessels share waterways. Simulation allows AI models to be trained safely and cost-effectively before deployment on real vessels, while also preparing human crews for this new operational reality.
How does NAUTIS support the development of autonomous ship technology?
NAUTIS generates synthetic training data for AI autopilot models through procedural scenario creation via its API. The simulator replicates standard maritime protocols, allowing AI models trained in NAUTIS to transfer their learned behaviours directly to real vessel operations. VSTEP’s collaboration with Captain AI uses this capability to develop AI models capable of navigating autonomously in complex maritime environments.
What is the difference between synthetic data and real-world maritime data in AI training?
Real-world maritime data is limited by operational constraints. You cannot safely generate data from collision scenarios or extreme conditions on a live vessel. Synthetic data generated in a maritime simulator like NAUTIS can cover the full range of scenarios including high-risk situations, at scale and at a fraction of the cost. The challenge is ensuring the synthetic data accurately reflects real-world vessel behaviour, which is where NAUTIS’s realistic hydrodynamics and protocol compliance are critical.
What does a hybrid maritime environment mean for seafarer training?
A hybrid environment is one where autonomous vessels and human-controlled vessels operate on the same waterways. Seafarers training for this future need exposure to AI-driven traffic behaviour, understanding how autonomous vessels navigate, respond to obstacles and make routing decisions. VSTEP’s integration of Captain AI’s autopilot into NAUTIS creates exactly this kind of training environment.
Updated on 09-08-2026.
Why this collaboration reads differently in 2026
The regulatory context has moved since the project started. In May 2026 the IMO adopted its first Code for maritime autonomous surface ships, in effect since 1 July 2026, covering areas that include navigation, connectivity and remote operations (IMO). Mixed traffic, in which autonomous and conventionally crewed vessels share the same fairway, is becoming a planning assumption rather than a thought experiment.
That changes what autonomous vessel simulation is for. Two groups need the same synthetic environment. The autonomous system has to meet rare encounters at a volume live sailing cannot supply, and the bridge team has to learn how autonomous traffic behaves before meeting it at sea. Running both in one environment is the practical argument for developing an autopilot inside a training simulator rather than testing it in isolation.
Autonomous simulation with NAUTIS
Explore the NAUTIS ships simulator and our maritime training solutions, or book a demo to discuss autonomous and mixed-traffic scenarios. The same approach applied to uncrewed vessels is described in our USV remote operator training story.
