Skip to main content

Bike counting at docks could save fuel

18 March 2024

A prototype that counts the number of passengers with bicycles at the dock. Could that be something? Well, it could actually save fuel in archipelago traffic. It's one of the results from the innovation project "Data Collection with IoT on Passenger Ferries" conducted within the Swedish Transport Administration's industry program "Sustainable Shipping," managed by Lighthouse.

Smart homes, smart buildings, and smart cities. Most have probably heard of concepts that use sensors, IoT, and cloud-based data storage to create more comfortable, efficient, and sustainable living and working environments for people. But does it work at sea? To test this, researchers in the project "Data Collection with IoT on Passenger Ferries" installed sensors and other IoT technology on M/S Dessi, a passenger ferry operating between Kalmar and Färjestaden.

"It doesn't work far out at sea, but it worked well along the coast. We connected and tested several networks, including Stadshubben, which is an infrastructure for IoT applications covering almost all of Kalmar and other Swedish municipalities," says Fredrik Ahlgren at Linnaeus University, who led the project.

IoT sensors, also installed at docks, collected data on important environmental and operational parameters such as air quality, noise levels, movement, and weather conditions.

"Today's ships have plenty of measuring points from the start. There's access to a lot of machine data, data from fan systems, temperatures in cabins, and so on. With IoT technology, these can be complemented in a simple and inexpensive way as long as you're within reach of the existing networks. It creates significant opportunities," explains Ahlgren.

A sensor typically costs between a few hundred to over a thousand Swedish kronor and can be attached with double-sided tape. For example, on M/S Dessi, they installed an air quality meter measuring particles in the passenger area, but they also did something as simple as measuring the temperature for sandwiches in the refrigerated display – which the shipping company is required to do according to food law.

"What didn't work so well with the technology was quite expected. It was difficult to get coverage down in the engine room, while, for example, the weather station we placed up on the ship's mast sent fine signals to the land," says Ahlgren.

A significant outcome of the project is that it generated a prototype for bicycle identification using machine learning.

"The idea came from Ressel Rederi, with whom we collaborated. We hadn't thought about it at all, but if the one operating a passenger ferry in shuttle traffic knows how many passengers with bicycles are waiting at the next dock, the speed can be adjusted accordingly. Taking on and securing bicycles takes quite a bit of time. So, you can slow down when there are few bicycles and thereby save quite a bit of fuel," Ahlgren explains.

The best way to identify bicycles is with a camera and machine learning.

"The measurement doesn't need to be exact. It doesn't matter so much if it's about eight bicycles or ten. What you want to know is whether it's about a dozen or dozens. This is, of course, a technology that would also be used on land, for example, in bus traffic," says Ahlgren.

The report "Data Collection with IoT on Passenger Ferries" was authored by Fredrik Ahlgren, Oxana Lundström, and David Mozart from the Department of Computer Science and Media Technology, Linnaeus University, in collaboration with Linnaeus University, RISE, Sensative AB, and Ressel Rederi AB.


Dela på