Registration

To participate, teams can register by submitting a form on our website. As part of the submission, teams should include details about any previous experience working with autonomous systems. This marks the 4th physical Njord Challenge, and we encourage all eligible teams to register and take part in this exciting event.
You can find the handbook containing the rules and guidelines for this years competition below.

Sign up here.‍
‍

Info for competing teams

This years competition is being held in Trondheim from the 10. to the 14. of august 2026.

The guide below has been designed to provide you with all the information you need to participate in the upcoming autonomous ship challenge.

Inside, you'll find information about the physical challenge. You can learn about registration, time schedule, and important deadlines, as well as technical specifications, test criteria, and the technical report.

We hope this handbook will serve as a valuable resource as you prepare for the Njord Challenge 2026. The Handbook may be updated up until the Challenge in order to ensure clear and reasonable guidelines. If changes are made, all participants will be notified.

Link to github handbook: Team_handbook_Njord_2026

Software & Online training

Link to access software and start online training here

Self-paced online courses (from basic to intermediate, with a lot of attention to pedagogics and applicability):

MATLAB Onramp: basic MATLAB introduction for those who are unfamiliar with the core parts of MATLAB, our programming and numeric computing platform.
Simulink Onramp: learning the basics of block-based models to simulate complex real-world systems, like autonomous ships, and deploy into languages such as C++, directly. If needed, the is a more advanced course for Simulink, called Simulink Fundamentals.
Simscape Onramp: how to rapidly include models of physical systems within the Simulink environment, ranging from realistic batteries to physical representations of the ensembled ship coming directly from CAD files. This would be key for a realistic simulator of an autonomous ship facing the challenges of Njord 2025.
Stateflow Onramp: how to leverage the high-level control and decision logic possibilities of Simulink, like switching from navigation to docking modes with predefined logics in a visual platform which will make debugging significantly easier.
Control Design Onramp with Simulink: basics of feedback control in Simulink, greatly improving the performance of all teams as this was left as future work for most of them. It includes strategies to tune basic controllers, which they informed me was challenging with the limited amount of practice time they had with a new system.
Computer Vision Onramp: introductory course focused on a typical workflow of object detection and tracking.
Deep Learning Techniques in MATLAB for Image Applications: collection of online courses focused on Deep Learning for computer vision, including object detection, which would be useful for docking.a

General resources for robotics

Robotics Learning Resources for Students: a collection of resources including some of the previous courses and, more importantly, examples on how to some useful things for Njord like importing the ship from CAD into Simulink/Simscape, co-simulating with Gazebo, using ROS with Simulink, or generate C++ code to be deployed into the computer onboard. Ideal for bridging the introductory courses and the actual application.
PID Control Made Easy: webinar on more advanced controllers with focus on deployment.
MATLAB to C/C++ Made Easy: webinar on how to deploy your MATLAB code directly into C/C++ using MATLAB Coder.

Advanced materials that will be helpful for Njord

Getting Started with MATLAB, Simulink, and ROS: blog on how to connect Simulink and ROS for complex systems, full of links to other relevant resources videos.
Deep Learning with MATLAB, NVIDIA Jetson, and ROS: short video with example code to use Jetsons and ROS, which was the most common hardware/middleware choice among the competitors.
Implement Deep Learning Applications for NVIDIA GPUs with GPU Coder: connected with the previous resource, a webinar to implement Deep Learning algorithms with GPU Coder, which works with the NVIDIA Jetson.

Navier dataset

We are excited to announce the availability of the Navier dataset! This dataset has been created to support teams in accessing valuable data for their projects. If you’re interested in using the Navier dataset, you can find it here.

Additionally, we’re in discussions with MathWorks to explore ways to highlight their solutions and demonstrate how their tools can assist in solving complex problems.
‍
‍
‍