MIT engineers have designed a modular, unmanned boat specifically for cities with lots of inland water, such as Amsterdam. The new design follows an earlier extensive test with a prototype in our capital city.
MIT expects autonomous boats to play an important role in future transport within water-rich cities such as Amsterdam, Bangkok and Venice. Researchers at MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL), in collaboration with the Senseable City Lab in the Department of Urban Studies and Planning (DUSP), have taken another step towards that future with the design of a fleet of autonomous boats that can easily manoeuvre and be controlled very precisely. The boats can also be quickly 3D-printed with a low-cost printer. This will make mass production a lot easier.
Decongesting traffic
The designed boats can carry both people and goods, thus taking more of the burden off street traffic. In the future, the researchers also envision the unmanned boats being adapted to run city services at night, rather than during busy daylight hours, further reducing congestion on both roads and canals.
"Imagine a number of infrastructure services that usually take place on the road during the day - deliveries, waste management, waste management - being carried out in the middle of the night, on the water, by a fleet of autonomous boats," says CSAIL director Daniela Rus , co-author of a paper on the technology presented last week at the IEEE International Conference on Robotics and Automation.
Moreover, the boats - rectangular 4-by-2-metres whose hulls are equipped with sensors, microcontrollers, GPS modules and other hardware - can be programmed to assemble themselves into floating bridges, concert stages, platforms for food markets and other structures in a matter of hours. "Again, some of the activities that usually take place on land, causing disruption in how the city moves, can be done temporarily on water," says Rus.
The boats can also be equipped with environmental sensors to monitor a city's waters and gain insights into urban and human health.
Better design and control
The work was carried out as part of the "Roboat" project, a collaboration between the MIT Senseable City Lab and the Amsterdam Institute for Advanced Metropolitan Solutions (AMS). In 2016, the researchers tested a prototype in Amsterdam's canals as part of the project. The boat traveled a pre-programmed path forward, backward and sideways.
The ICRA paper includes a number of important new innovations: rapid manufacturing technology, a more efficient and agile design, and advanced tracking algorithms that improve control, precision docking and locking, and other tasks. Quite a technical story, but interesting.
To make the boats, they first 3D printed a rectangular hull with a commercial printer and produced 16 separate sections that were spliced together. The printing took about 60 hours. The completed hull was then sealed by bonding several layers of fibreglass.
Integrated into the hull are a power supply, Wi-Fi antenna, GPS and a minicomputer and microcontroller. For accurate positioning, the researchers have incorporated an ultrasonic beacon system and real-time open-air kinematic GPS modules that enable centimetre-level localisation, as well as an inertial measurement unit (IMU) module that monitors the boat's turn and angular velocity, among other statistics.
The boat has a rectangular shape, rather than traditional kayak or catamaran shapes, to allow the craft to move sideways and attach to other boats when assembling other structures. Another simple but effective design element was the placement of the bow thruster. Four thrusters are placed in the centre of each side, rather than at the four corners, generating forward and aft forces. This makes the boat more manoeuvrable and efficient, say the researchers.
In-water testing
The team also developed a method that allows the boat to track its position and orientation faster and more accurately. To do this, they developed an efficient version of a predictive control (NMPC) algorithm, which is generally used to control and navigate robots within various constraints.
NMPC and similar algorithms have previously been used to control autonomous boats. But usually those algorithms are only tested in simulation or do not take into account the dynamics of the boat. Instead, the researchers incorporated non-linear mathematical models that included a few known parameters, such as boat drag, centrifugal and Coriolis forces, and added mass due to accelerating or decelerating in water. The researchers also used an identification algorithm that then identifies unknown parameters while training the boat on a path.
It is not yet known when the boats will actually operate in Amsterdam.
By: Kelly Bakker
Source: MIT
