Kristjan Tabri, Heigo Mõlder: Robots could guard marine infrastructure

Command center of an autonomous robotic ship (MindChip).
Command center of an autonomous robotic ship (MindChip). Source: TalTech

Marine infrastructure security monitoring solutions can also be used for broader exploration of the marine environment as well as for servicing offshore wind energy and aquaculture. Increasingly smarter artificial intelligence (AI) could open up yet unanticipated applications at sea, write Kristjan Tabri, tenured associate professor of marine structures and technology at Tallinn University of Technology (TalTech) and Heigo Mõlder, researcher at TalTech's Department of Electrical Power Engineering and Mechatronics.

Hundreds of kilometers of vital infrastructure, from power and communications cables to gas pipelines, meander along the Baltic Sea floor.

Underwater infrastructure is of vital importance both for daily life and from a security aspect, and the volume of such infrastructure is increasing rapidly. For example, dozens of new offshore wind farms are going up in Europe alone in the coming years.

While underwater infrastructure was previously invisible and inconspicuous for the most part, recent incidents involving Nord Stream, the Balticconnector gas pipeline and various communications cables have demonstrated that greater attention needs to be paid to the security of underwater infrastructure.

Fully protecting infrastructure from natural or human damage, either intentional or unintentional, is either very costly or impossible. The solution is continuous infrastructure monitoring. It's unfortunately clear that without the help of modern technology, humans cannot manage to monitor such infrastructure. Essentially, four options exist, which can complement one another if necessary.

1. Underwater sensors: ears below the surface

Sensors installed to the seafloor, underwater infrastructure or to float in its vicinity that would detect unusual activity and disturbances. These sensors would provide real-time data, providing continuous information about changes in the underwater environment.

As with any solution, it has its challenges. Major storms and the constantly changing seafloor can cause false alarms or damage sensors. Power supply for and data acquisition from underwater sensors is difficult and costly. Development costs for such a surveillance solution would likely be in the millions even for relatively small-scale infrastructure. Add to this running costs.

2. Underwater drones: spies of the underwater world

Underwater drones are small yet powerful underwater robots equipped with sonar, cameras as well as various other sensors. Their flexibility enables them to respond rapidly and interact with the underwater world.

The lack of radio communications in deep water can cause issues with piloting, however, when a drone is far from its base station. The high price of such drones also makes losing one especially painful.

3. Autonomous robot ships: robots that patrol at sea

The ability of robotic ships to operate 24/7 makes them ideal sentries. One can imagine, for example, how an autonomous ship might carry out daily sonar inspections of the Estlink cables between Finland and Estonia, detecting changes and registering suspicious underwater noise. In addition to the data collected, frequent surveillance would also have a deterrent effect.

A robotic ship doesn't get bored and doesn't tire, nor does it make human navigational errors. Robotic ships are relatively cheap, with a small environmental footprint and low operating costs.

4. Maritime surveillance: eyes and ears on the water

One emerging trend we're seeing in maritime infrastructure security is the integration of artificial intelligence, radar, satellite imagery and advanced communication systems for the real-time monitoring of maritime traffic.

These technologies allow for the creation of a dynamic and detailed picture of activity at sea level, detecting deviations from routine traffic patterns. By linking these systems to a central coordinating command center, it is possible to detect almost immediately if any ship stops suspiciously close to critical infrastructure.

If the AI detects such an anomaly, the command center can step in, either communicating directly with the vessel or dispatching a patrol to the site. This system also helps reduce human error, while at the same time offering faster and more efficient ways to respond, which in turn increases the security of the entire Baltic Sea.

Estonia's chance

In addition to a precisely and independently navigating robotic ship, the monitoring of underwater infrastructure also requires machine learning-based sonar or camera image processing that detects changes from previous measurements.

Such algorithms detect suspicious objects and record changes either to the infrastructure itself or on the seabed within the vicinity thereof. The interpretation of detected anomalies will remain at least for the time being the responsibility of the operator of the robotic ship, and if necessary, an anomaly can be given a closer look using underwater robots or divers. Sonar data can likewise be supplemented if necessary with underwater acoustics, which can help detect suspicious sounds as well as the movements of other drones.

A monitoring solution based on autonomous systems technologically combines several rapidly developing disciplines ranging from AI, machine learning and cybersecurity to satellite-based precision navigation.

Estonia is a compact country with a long coastline already surrounded and slated to be further surrounded in several directions by various vital underwater connections. Local technical expertise with active marine industry and marine technology competencies provides us with the necessary basis for developing a suitable autonomous solution.

TalTech spinoff company MindChip is already developing autonomous robotic ship-based solutions that can also be adapted for use in underwater infrastructure monitoring. There are likewise research institutes and companies in Estonia involved in marine robotics, underwater robot and underwater sensing development. Such research and development activity would not constitute merely the solving of a specific monitoring task, but would also help us move in the same direction as the rapidly developing marine and blue economy.

Infrastructure security monitoring solutions can also be used for the broader exploration of the marine environment as well as in servicing offshore wind energy and aquaculture. Increasingly smarter artificial intelligence (AI) could open up fields of application at sea that we can't even yet anticipate.

Maritime activity on the Baltic Sea is set to accelerate in the coming years, bringing with it more and more new engineering challenges. By taking direct part in solving these challenges with the help of new technologies, we will be investing in growing sectors and establishing critically important competencies right here — in Estonia.


Follow ERR News on Facebook and Twitter and never miss an update!

Editor: Aili Vahtla

Hea lugeja, näeme et kasutate vanemat brauseri versiooni või vähelevinud brauserit.

Parema ja terviklikuma kasutajakogemuse tagamiseks soovitame alla laadida uusim versioon mõnest meie toetatud brauserist: