AT OUR CURRENT moment in third decade of the 21st century, the relationship between technology and conservation could generously be described as uneasy: decades of greenwashing have seen the goodwill behind formerly sound concepts like “eco”, “organic”, and “autonomous” erode like so much sand. Current technological advances, from electric / autonomous vehicles and drones to the self-perpetuating slurry of metaverse and NFT content, pay occasional lip service to the impending wave of climate catastrophes, while rarely working to counteract it, and in some cases even accelerating it.

Robotic Landscapes, new from Park Books, begins the process of steering the diverging paths of technology and sustainability toward a common goal. Through a broad, multidisciplinary series of academic essays, the book brings both rigorous science and wide-ranging philosophy to its examination of autonomous robotics in landscape design. And while climate change and the compounding environmental and architectural stresses it will bring is often more implicit than explicit in the book, its presence can be felt throughout.

The book suggests a new model of architecture which adds a fourth dimension – time – to the standard spatial three. While every building begins aging as soon as its construction begins, in the traditional model there is nonetheless a clear division between architect/builder and owner/manager, marking – short of any catastrophic failures – the end of the former’s obligations via a handoff to the latter. The model put forth in Robotic Landscapes brings the passage of time into the equation from the beginning: “While it is possible to forecast likely scenarios using simulation, it is virtually impossible to anticipate natural events that lie far in the future. As such, a computational approach that dynamically accommodates change is vital.” The word “dynamically” is key, implying a process of self-correction on the part of an analytical intelligence – and while the human element is not written out of the equation, the presence of robotics and AI are implicit in the broader context of the book.

This new architectural impetus can also be seen as a shifting of debt: the years and decades though which a building is expected to last are not just the responsibility of its future owners and residents, but of its architects as well. Renewal and reconstruction are part of the DNA of such projects; indeed, the authors explicitly evoke the idea of chromosomes as an analog to such architecture by reversing Schrödinger’s concept of the chromosome-as-architect:

It was this same entropy-reversing ‘live’ that caused physicist Erwin Schrödinger to turn in 1944 to biology in search of his ‘negative entropy,’, and in particular to the chromosome with its uncanny ability to preserve information and resist formal decay. Tellingly, Schrödinger names the chromosome – the key protagonist of his gene action theory – ‘architect’s plan and builder’s craft in one,’ that is, both instruction and execution.

“Instruction and execution” in particular evoke the new era of AI, where computer intelligences are trained to work toward a certain goal within a certain set of parameters. With this conceptual framework in place, the question becomes how well it will work in the real world of earth, stone, and water.

Though it’s not overly emphasized in the text, the book is the result of high-level academic research into robotics, and in particular the ways in which robots can be programmed – “instructed” – to correct fluctuations in terrain, steering landscapes back toward stability from their inevitable entropic breakdowns. The gap in scale between the project’s lab and field research is notable, with the former represented mostly by trays of sand in controlled indoor environments, and the latter by intense and often violent upheavals of the natural world, in particular the landslides and flooding that devastated the Swiss village of Bondo in 2017. This gap is bridged by the aptly-acronymed HEAP (Hydraulic Excavator for an Autonomous Purpose), an autonomously guided earthmover that adjusts dynamically to changing terrain as it reinforces that same terrain. The challenge posed by climate change is daunting, when one considers how fast and unpredictable catastrophic weather and terrain events can be, not to mention how far away their origins may be from the sites that are ultimately affected. Nonetheless, the research shown in the book points to a solid initial foundation, so to speak, for the coming age of autonomous automated landscape design.

The book itself is remarkably beautiful, with a design that commits to a series of subtle off-white rectangles printed across each double-page spread, with text and images placed within, between, and across this unique grid structure. The binding uses a drop-spine-style dust jacket for stability, while laying open beautifully on its own during reading. The images shift subtly from photographs to illustrations and back, and like the text serve to unite the conceptual with the practical. The prospect of accelerating climate-based crises is grim enough on its own; that the project approaches its subject with such a heightened aesthetic is testament to the desire to unify art and science in a common goal.

Robotic Landscapes: Designing the Unfinished
Edited by Ilmar Hurkxkens, Fujan Fahmi, and Ammar Mirjan
Park Books, Paperback, 238 pages