Programmable Cities: Using Roboat to Create a responsive Autonomous Infrastructure in Amsterdam
By Tom Benson, Stephan Van Dijk and Michael Batty
The Emergence of a New AI: Autonomous Infrastructure
Cities emerged more than 5000 years ago and have always been dynamic creations whose populations continually interact with the artefacts of built form that are designed to enable them to function as places of exchange and social interaction. Until the industrial revolution, which began in Europe in the middle of the 18th century, cities were dominated by continual adaptation to new technologies generating high rates of growth. Innovations were introduced, including steam power, which allowed national transportation networks, seen in Figure 1, to be developed worldwide. Infrastructure and networks evolved, enabling populations to respond quickly and effectively to change from within and without (O’Brien, 1983). Cities then were not necessarily prosperous, but they were highly responsive to human needs. Individuals could move quite fluently, despite technological limits on how far they could travel. The built form that emerged reflected the unself-conscious architecture that was well adapted to the way people moved and communicated prior to the machine age.
Now in the dawn of the age of Intelligent machines and digital technologies, we have the potential to dramatically affect a city’s infrastructure and other forms of human behaviour that urban planners view as determining the form and function of spatial and social structures (Batty, 1997). In fact, infrastructure is just beginning to witness the integration of embedded intelligence to increase the efficiency of commuting and living in cities and become adaptive itself. Current infrastructure developments are complex, lengthy processes and often produce slow transitions in the form of cities. Still, they are beginning to provide intelligence through the convergence of their software with their physical form and are starting to offer programmable environments for urban living.
Autonomous vehicles (AV) and the rise of active travel can aid the development of cities as we quest for a low carbon future; they can provide a multitude of opportunities to rebuild our cities in ways that mirror an earlier era allowing people are able to adapt to the infrastructure of the city with that infrastructure being much more responsive to the needs of the people. The software that now defines the digital transformation that is all-pervasive in contemporary society is being rapidly embedded in our infrastructure and within our person in the form of smart devices. This is leading to infrastructure that in itself is mobile, continually responsive to human needs and demands, acting in intelligent ways in providing new forms of pop-up infrastructures that enable improved mobility across the many locations that define where we work, live, entertain and learn (Derrible and Chester, 2021).
In this essay, we explore the potential and the implications of introducing intelligence and autonomy to elements of the ‘classical’ infrastructure of the city (like bridges, roads, canals ao.) and how this is enabling the creation of a more resilient and programmable city (Siller and Stibe, 2016). We discuss how the existing infrastructure of cities can be complemented with intelligence and autonomy that can transform it from its initial passive state to a more dynamic and responsive state never possible before.
Our case study will be the city of Amsterdam, where we begin by exploring its infrastructure, which for 50 years has been dominated by automobiles and how this has affected the urban environment. Following this, we shed light on the historic canals and waterway systems that have formed the spatial configuration of the city. We then investigate how the introduction of fully autonomous, reconfigurable vessels can bring back and restore the importance of the canals, with the canals themselves being privy to the transportation and primary city services. The integration of the intelligent vessels can become a new conduit, creating on-demand connectivity between spaces, nudging us to blur the borders between ‘vehicle’ and ‘infrastructure’, thus shaping an upgraded urban system that is responsive and programmable, allowing the potential for new space and interaction. It will show how far-reaching the impact and implications of new technologies, like AI and robotics, are for the milieu of future cities, enabling us to explore the potential of a reciprocal dialogue between infrastructure and human populations. Finally, we set this essay within its broader social, political and cultural context.
Amsterdam’s Canals
Amsterdam is recognised worldwide for its historical infrastructure, much of which can be traced back to the 17th century. Actually, this continues to constrain the development of the city and its social interactions. The strategic arrangement of interconnected canals still defines the city, but the vital significance of the waterways has slowly shifted in its purpose over time. When first constructed, the canals were mainly used for water management and served as a defence for the city in continuous battles that dominated the city’s history from the 16th to the 18th centuries. But as the area of Amsterdam grew, the canals became a fundamental feature for city services as they were used to transport goods and food, also playing an essential role in the city for trade and global shipping. These changes increased the heterogeneity of social activities and the diversity of commerce, shown in Figure 2. When the bicycle was invented, this added yet another dimension to the geometry of movement, providing a gentle and fluid sense of how the different parts of the city could be connected. But with the coming of the railway and then the automobile, all this became vulnerable to modernism that sought to do away with local movement in the interests of mass transit and ever longer travel times between places. In the last century, as the automobile became dominant in cities worldwide, the canals were under the palpable threat of being paved over to adjust the city to a new transport system primarily motivated to ease traffic congestion in the city (Cox and Koglin, 2020).
The city’s transformation to fit the automobile was highlighted in 1967 when the ‘Jokinen Plan’, portrayed in Figure 3, was recommended for Amsterdam by urban planner David Jokinen. The plan intended to re-design the city for the car by paving over the canals, building over them with six-lane roadways and high-rise structures. Even though the complete plans failed to materialise, the canal space was halved during the 20th century to accommodate more space for the automobile, thus alleviating some street traffic. Consequently, the canals lost their significance in being a central part of the processes of moving assets and services to delivery trucks and various vehicle types. As a direct result of all this, the city’s streets generated high traffic congestion rates and traffic-related polluting emissions. Today, the canals’ water surface is still equal to 35% of Amsterdam’s municipal area, and about 70% of all city districts can be reached over water. While the current function of the canals is merely for tourists or social boat trips, seen in Figure 4, the potential to utilise the waterways to improve urban living is still genuine.
Roboat: Autonomous and Responsive Vessels for Amsterdam’s Waterways
The city of Amsterdam has the potential to regenerate its canals, reduce road transport issues, and utilise them to their potential. Here, we present Roboat [robotic boats], a research and development project between the Massachusetts Institute of Technology and the Amsterdam Institute for Advanced Metropolitan Solutions, which strives to bring back the value of the canals to the city with its first fleet of modular and self-driving boats (Duarte et al., 2020). Roboat explores the intersection between computer science, robotics, AI, maritime, civil and environmental engineering that has profound implications for the juxtaposition of mobile infrastructure in the city, powered by autonomous boats that can be fashioned to contain moveable land-use activities. Initially, the work was focused on researching and developing full-autonomy for the vessels (through perception, controls, learning) and the interface between vessels and quays and piers. However, the project evolved into designing multi-vessel coordination to orchestrate the interaction between multiple vessels and objects and connect or ‘latching’ ships to each other and the water infrastructure. The vessels’ novel functionalities and capabilities blur the boundaries between ‘vehicle and infrastructure’ and create an ‘autonomous and responsive system’ for transport, accessibility, and land use.
The autonomous vessel units can work singularly as transport; for instance, moving household waste, construction waste, or people, goods and assets. But in addition, the vessels can also work together and coordinate their actions as ‘swarms’ of Roboats. With a novel latching mechanism, Figure 5 showcases how the Roboats can create rigid connections linking them in various shapes and formations that create a programmable infrastructure. The Roboat technology pushes the limits of vehicle-to-vehicle communication into vehicle-to-infrastructure communication, allowing the city’s infrastructure to respond to human behaviour, creating two-way communication between Roboat and any citizen of or visitor to Amsterdam who avails themselves of this new technology.
roundAround: Extending the System
To examine the potential of Roboat to create an on-demand and responsive infrastructure in more detail, the concept roundAround — the world’s first dynamic bridge — has been explored (Leoni et al., 2019). Figure 6 shows how roundAround aimed to connect two popular locations, the Nemo Science Museum and the Marineterrein in Amsterdam, which are currently separated by a waterway. It takes nearly 15 minutes to walk between these places, but Roboat can decrease this travel distance to 2 minutes by connecting vessels that circle from one edge of the waterway to the other. Additionally, the concept allows for unique feedback loops and communication between humans and infrastructure, affording Roboat’s potential as a fleet to dynamically adjust physical space to the citizens and city’s needs.
In this way, the infrastructure can be manipulated to provide new transport links and contain new land use activities that can be moved in response to changing human demands. Clearly, there are other constraints on such pop-up land uses. Still, the Roboat and roundAround system have enormous potential for increasing the diversity and accessibility of the city to its residents and visitors.
During the morning peak, the units can be self-assembled to construct bridges to reduce traffic congestion during rush hour. In the afternoon, they can form a plaza or a stage for food markets or various cultural events that create new spaces joining the water and canal edge; in the evening, they can make a space for families, offering them a wide variety of public activities. Depending on demand and the required functionality, the Roboats can go where needed, freely moving throughout the extensive canal system. The spaces created by the vessels can provide the city with spaces to enhance social life, enabling new places for people to meet people and bringing various cultures together (Gehl, 2011).
Conclusions
This essay’s ambition has been to demonstrate how advances in autonomous technology can upgrade urban mobility and transform urban space in Amsterdam. With Roboat and roundAround, we decipher the opportunity to use AV in a waterway setting to create programmable autonomous infrastructure and build two-way communication between the boats and human behaviour. Despite the possible economic, environmental, and social advantages of introducing Roboat in Amsterdam, implementing the project has complex challenges that are not only technical in nature. The possible and potential interactions between citizens and Roboats provides a context and environment whose properties are still largely unknown. In addition, some citizens are universally sceptical or concerned about AI applications and autonomous vehicles or robots in their cities.
The relationship between humans and machines has been a regular debate between people since ‘automated’ manufacturing processes came to fruition during the industrial revolution. In 1942, Isaac Asimov introduced three laws of robotics aiming to create an outline to protect humans from robots (Asimov, 1941). The laws were: 1) A robot may not injure a human being or, through inaction, allow a human being to come to harm. 2) A robot must obey the orders given by human beings except where such orders would conflict with the first law. 3) A robot must protect its own existence as long as such protection does not conflict with the first or second law. When Asimov defined the laws, he saw a playful future with robots acting more like androids, acting as ‘slaves’ to humans. Since Asimov created his laws, notable robotic progress has happened. As we move forward as a society to implement new technological inventions, it may help revisit the ‘laws’ that complement developments such as Roboat.
As Roboat progresses, it is essential to deeply examine and determine its political and social identity and understand the duties, liabilities, and general civic status that such infrastructure defines. In addition, questions will need to be answered as we develop with and deal with machines daily; for example, who is accountable if a robot injures a human? Who governs AI and the programmable infrastructure? How are we to communicate the intentions and manoeuvres of Roboat in a straightforward way to people interacting with each other and make these socially acceptable and contestable? And how do we ensure accessibility and inclusivity of these new technologies to all citizens instead of limiting it to the happy few or digitally capable? As intelligent technology, machines and smart cities become more prevalent as the century wears on, these questions will be fundamental in their deployment in moving civilisation to the more desired state.
Tom Benson is a researcher and designer who trained in architecture in the UK. Since September 2018, he has been working as a Research Fellow at the Massachusetts Institute of Technology’s Senseable City Lab. Currently based in Amsterdam, his projects involve studying the built environment through large datasets using computational techniques.
Stephan van Dijk is Director of Innovation at the Amsterdam Institute for Advanced Metropolitan Solutions. He is also a member of the Innovation Board for Smart Mobility at the Amsterdam Metropolitan Region (AMA), member of the EIT Urban Mobility Nomination Committee, and member of the steering committee Impact Monitoring NZ Metroline Amsterdam.
Michael Batty is Bartlett Professor of Planning at UCL where he is Chair of the Centre for Advanced Spatial Analysis (CASA). His work is focussed on computer models of cities and their visualisation, and more recently how computation lies at the heart of the smart cities movement. His most recent books are The New Science of Cities (2013) and Inventing Future Cities (2018) published by MIT Press. Both have been translated into Chinese
To go further
Tom Benson, Michael Batty and Stephan van Dijk will give an online lecture on this topic on December 14th. To register : https://www.eventbrite.fr/e/urban-ai-conversations-tickets-190867859907?fbclid=IwAR29MvP0BjRgcL6ruO-7_V-Lcbqwe5P2CsyQ1So7NiUrH576jhvIdBMLoro
References
Asimov, I., 1941. Three laws of robotics. Asimov, I. Runaround.
Batty, M., 1997. The computable city. International Planning Studies, 2(2), pp.155–173.
Beggrov, K., 1840. Tsarskoe Selo Railway. [lithography].
Breitner, G., 1894–1898. [image] Available at: <https://www.rijksmuseum.nl/en/rijksstudio/artists/george-hendrik-breitner/objects#/RP-F-00-565,9> [Accessed 3 August 2021].
Cox, P. and Koglin, T. eds., 2020. The Politics of Cycling Infrastructure: Spaces and (in) equality. Policy Press.
Dai, T., Hein, C. and Zhang, T., 2019. Understanding how Amsterdam City tourism marketing addresses cruise tourists’ motivations regarding culture. Tourism Management Perspectives, 29, pp.157–165.
Derrible, S., and Chester, M. V., Editors. 2021. Urban Infrastructures: Reflections for 2100, Amazon Books, https://www.amazon.com/gp/product/B08LZV66YK/
Duarte, F., Johnsen, L. and Ratti, C., 2020. Reimagining urban infrastructure through design and experimentation. The Routledge Companion to Smart Cities, p.52.
Gehl, J., 2013. Cities for people. Island press.
Jacobs, J., 1961. The Death and Life of Great American Cities, Random House, New York.
Leoni, P,. Johnson, L., Duarte, F,. and Ratti, C,. 2021. roundAround. [online] Available at: https://senseable.mit.edu/roundaround/ [Accessed 15 Jun. 2021].
Nieuwenhuis, A., 2021. [image] Available at: <https://www.iamsterdam.com/en/plan-your-trip/getting-around/boating> [Accessed 3 August 2021].
O’Brien, P., 1983. Transport and economic development in Europe, 1789–1914. In Railways and the Economic Development of Western Europe, 1830–1914 (pp. 1–27). Palgrave Macmillan, London.
Planner, S., 2021. [image] Available at: <https://sandyjamesplanner.wordpress.com/2019/12/19/what-happened-when-the-netherlands-imported-american-road-engineers-a-true-story/> [Accessed 3 August 2021].
roboat.org. 2021. Roboat. [online] Available at: <https://roboat.org/> [Accessed 3 August 2021].
Siller, M. and Stibe, A., 2016. Programmable Cities: A New ICT Approach. In ECSW@ PERSUASIVE.
