Roboat is a 5 year research project and collaboration between
the Amsterdam Institute for Advanced Metropolitan Solutions and the Massachusetts
Institute of Technology. In developing the world's first fleet of autonomous
floating vessels for the city of Amsterdam, it investigates the potential of
self-driving technology to change our cities and their waterways.
Roboat is a new kind of on-demand infrastructure: autonomous platforms will combine together
to form floating bridges and stages, collect waste, deliver goods, and transport
people, all while collecting data about the city. How can we re-imagine urban
infrastructures with cutting-edge technologies?
Technical details
Perception
Perception
Roboat is equipped with a LiDAR time-of-flight sensor and a camera to view its surrounding environment. Perception methods such as clustering and neural network classifiers are used on sensor readings to recognize objects in the canal environment. This enables Roboat to avoid other boats while navigating and to interact with other Roboats while docking.
Motion Planning with Obstacle Avoidance
Motion Planning with Obstacle Avoidance
Roboat can wisely plan its path to avoid collision with known obstacles when moving throughout the urban canals. The motion planner always considers the boat dynamics and geometry, canal boundaries, and obstacle dynamics simultaneously to calculate out the optimal obstacle-free path for the Roboat in real time. The motion planner will become even intelligent by actively learning its strategies and the surrounding dynamics in the near future.
Predictive trajectory tracking
Predictive trajectory tracking
Roboat can always track on the planned obstacle-free path with minimal effort and errors by implementing a nonlinear model predictive controller (NMPC). The NMPC iteratively optimizes the control action by collectively considering the reference trajectories, the nonlinear dynamic boat model, and the thruster force constraints in a finite time horizon. More constraints such as passenger comfort level could be added to the controller as required.
Route planning
Route planning
Roboat operation involves navigating through Amsterdam’s canals between origins and destinations. Key challenges for navigation stem from the presence of other vessels and the dynamic models of Roboat units. We develop a data-driven computational method to build a reactive system that can predict vessel movement and find an optimal route for canal navigation, avoiding collision with other vessels.
Latching
Latching
The latching system enables Roboats to create dynamic united floating infrastructure such as bridges, markets or concert stages while overcoming water disturbances. It is based on the spherical joint (ball and socket) that allows rotation and free movement in two planes at the same time.
Multi vessel coordination
Multi vessel coordination
Roboat units are required to self-assemble to build larger platforms and to navigate throughout Amsterdam canal as a team in order to serve various purposes and use cases. We create a multi-vessel coordination platform that includes communication, sensing, and control of multiple units in the Roboat network. We expect that the platform will enable autonomous latching of multiple Roboat units and effective management of those to provide services in a timely manner.
Environmental Monitoring
Environmental Monitoring
Roboat can collect mobile measurements of a variety of environmental conditions, including but not limited to: water quality, air quality, and weather conditions. Roboat's mobility allows us to collect environmental conditions at a much finer spatial resolution when compared to traditional stationary monitoring sites. The information provided by Roboat will enable local policy makers to make better, data-driven, decisions on environmental and public health issues.
Urban Interfaces
Transportation
Transportation
Roboats can be autonomous water taxis or buses, adding to the experience of both tourists and locals. And by leveraging a Roboat as a tugboat, existing private boats can enjoy an on-demand transportation service. This system allows individuals to moor their boats outside the city, and request it on demand.
Food delivery
Food delivery
Roboat units can form floating food markets or become pop-ups stalls that appear on the canal’s edge for Amsterdammers to collect crates of fresh produce. We imagine this use of autonomous boats could promote healthier food and create new public spaces for Amsterdam. Roboat activates the canals while tapping into the resources located on the pervasive regional network of waterways.
Garbage collection
Garbage collection
Roboat units can collect waste for the Centrum district of Amsterdam, where residents currently deposit trash on the curb for collection. To alleviate the many problems caused by large trash trucks on historic streets - congestion, pollution, noise, etc. - Roboat can serve as floating dumpsters that autonomously transfer waste. Our research shows that such a system could serve 70% of the Centrum district.
Infrastructure
Infrastructure
Roboat units can join together to create temporary bridges, alleviating congestion on Amsterdam’s centuries-old bridges and canal-side streets. As an autonomous system, Roboat can respond in real time to the ebb and flow of rush hour traffic. Individual units can also tessellate to form floating stages and public squares on the canals, enhancing Amsterdam’s strong tradition of water-based events with a 21st-century technology.
Visualizations
Laserscape
Laserscape
Experience how Roboat sees the world as it travels down Amsterdam’s canals. Using LiDAR technology, it builds a picture of the city from millions of data points, which is then used for navigation and classification of objects. This first person view of the canals could yield new insights into the life of Amsterdam’s waterways.
Roboat AR Experience
Roboat AR Experience
Experience Roboat use cases through augmented reality! Open the link below to download the apps and see how Roboat can enrich Amsterdam's urban environment.
A Summer Day
A Summer Day
This application visualizes the traffic on Amsterdam’s canals on an August day. Using data from MarineTraffic, it identifies locations in the canal network that experience the highest volume of boats. These types of traffic patterns optimize the global path planning algorithms of the Roboat system, which will learn from real-time data how to navigate the canals.
Timing Amsterdam
Timing Amsterdam
Use this application to evaluate travel times in Amsterdam. It uses the single largest continuous waterway to compare Roboat’s speed with that of cars, bikes, and public transportation. It considers time of day and speed limit as well as density of jobs, population, and attractions. In some cases, traveling by Roboat may be faster than public transportation or car.