Five years since its inception, MIT's autonomous floating vessel has seen a size upgrade and has been taught new methods of navigating waters on its own.
Roboat II moved around Amsterdam's canals over three hours, learning how to map and chart the famous Dutch waterways before docking with just a small seven-inch (17 cm) error.
MIT researchers from MIT‘s Computer Science and Artificial Intelligence Laboratory (CSAIL) are behind this project, along with the Amsterdam Institute for Advanced Metropolitan Solutions.
The vessel the team used in Amsterdam isn't the Roboat II's final size, at 6.5 feet (2 meters), the trial boat is roughly half of its intended size.
This second version of Roboat weighs roughly 110 pounds (50kg) and moves smoothly around thanks to its four propellers, LiDAR, GPS, and other sensors. It can move in any direction and has the ability to transport two persons as well as the equivalent weight in cargo.
Thanks to its algorithm for Simultaneous Localization and Mapping (SLAM), the autonomous vessel can be assigned specific tasks such as picking up people. Once the system knows its task, it automatically charts its course dependent on weather conditions and traffic.
Given how much focus has been placed on autonomous land vehicles as well as drones, it's great to see more attention on water-based autonomous movement.
"The development of an autonomous boat system capable of accurate mapping, robust control, and human transport is a crucial step towards having the system implemented in the full-scale Roboat. We also hope it will eventually be implemented in other boats in order to make them autonomous," said Wei Wang, a senior postdoctoral associate from the Senseable City Lab and CSAIL, and author of the study.
The modular system can be arranged to have multiple boats coordinating together, with one 'leader' boat taking charge. This means a larger set of cargo can be transported at once, with the 'follower' boats moving alongside the 'leader' boat thanks to a technique called "distributed control system."
The plan the team has in mind is now to create a larger autonomous version of the boat, which can carry between four and six passengers, and will measure 13 feet (four meters) long. The team is also going to focus on improving real-time object detection and identification, so as to be able to handle more complex situations that are likely to occur in a city's canal system.