Mutual communication is one of the substantial issues that needs to be sorted out before swarms of underwater robots can work together. Wi-Fi doesn’t work underwater. In the EU SWARMs project, TNO is responsible for developing the wireless underwater communication system.
SWARMs (Smart and networking underWAter Robots in cooperation Meshes), the successor of Artemis, is the first project within the new EU ECSEL programme, to start on 1 July 2015. The aim of the project is to demonstrate a number of scenarios in which a swarm of underwater robots carry out a task together.
Collaborating partners SWARMs
This extensive project, in which 30 partners from 10 countries collaborate, accounts for namely 17 million euros, which is made up of funding by both Europe and the national governments of the participating countries. The project is divided into nine work packages. TNOmainly forms part of the Communication and Demonstration work packages.
Autonomously Underwater communication data performance
Where currently a frogman is required, underwater operations can save expensive costs by having robots run operations and take action on the basis of their collected information. Of course, the ideal situation is if the robots aren’t attached to the ship by cables but that they are able to perform their tasks autonomously. However, for the time being, human presence is still needed, who should be able to in particular follow what is happening underwater live.
Underwater communication data
In dredging work, for instance, the current may not carry a cloud of sand on and deposit it on a coral, causing it to die. Sensors that position themselves against each other can measure the density and motion of such a cloud, but when doing so, they have to communicate with each other underwater to do this. Range, speed of data exchange and standardisation are a number of things which are important for underwater communication.
Robots of different origin can only collaborate if they understand each other, in other words 'speak' the same language. In that respect, the fact that the SWARMs consortium comprises so many different parties – manufacturers of sensors and actuators, companies developing acoustic communication technology and autonomous decision-making – is promising. Nowhere else in the world is there such a diverse group collaborating on this topic.
“Nowhere else in the world is there such a diverse group collaborating on this topic”
Wi-Fi has an issue with salt
The consortium’s ambition to tackle this problem has been simmering for a while, but before TNO got involved, none of the partners had any experience with acoustic data communication underwater. Henry Dol is one of the TNO researchers engaged in the SWARMs project. When we started working on this project, the other partners thought that regulating communication would only take a few modems and place them into the interfaces of the robots. But it is not as simple as that. Electromagnetic signals, like Wi-Fi, have a real issue with seawater salt.
High data speeds, short distances
In water, which is not compressible, acoustic signals, or sound waves, can be transmitted well. Unfortunately, standard modems exist for this. However, our work for the Ministry of Defence gave our group the experience with low-speed information transfer across long distances using modulated acoustic signals in seawater. So it was a logical next step for us to extend this to high data speeds across shorter distances using the techniques we had already used for electromagnetic signals.
High data speeds underwater
- The bandwidth – extend the frequency range of the acoustic signals used.
- The modulation – make the way information is packaged in the acoustic signal more efficient.
- Limiting the overhead – for instance, by optimising the self- learning algorithm that adjusts the modem to the underwater channel.
It all comes down to seeking an optimum between the data speed, the range and the robustness of the channel, whereby this last is related to all kinds of physical aspects (salt content, temperature, pressure) of the seawater, the medium in which the acoustic signal is transmitted.
End consumer added to consortium
Another aspect that the consortium lacked before TNO joined in, was the fact that there was no real end-user participating. The consortium comprised research institutes, universities and various industrial parties, but no real end-user. So we got the Dutch dredging company Boskalis involved, a company that was not only interested in the final result but also in playing a crucial role in which demonstrations will have to be known and be carried out.
Three demonstrations at sea
In all, three demonstrations at sea are planned in the SWARMs project: in the Canary islands (2016), the Black Sea (2017) and the Norwegian Sea (2018). The last demonstration will have to show that the idea works, in principle, for a number of robots. The industrialisation will then be left to commercial parties.
We already see a whole variety of possible applications, from measuring the quantity of fish in a particular area to the safe transport of sand and searching for fossil fuels in the sea. SWARMs could also help inspect sea-based wind farms or detect sea-borne drugs transport.