Archinect - News

Drexel University researchers develop AI-guided robotic structural inspection system

2024-02-12T08:00:00-05:00

Researchers based at the Drexel University College of Engineering have devised a new method for performing structural safety inspections using autonomous robots aided by machine learning technology.

The article they published recently in the Elsevier journal Automation in Construction presented the potential for a new multi-scale monitoring system informed by deep-learning algorithms that work to find cracks and other damage to buildings before using LiDAR to produce three-dimensional images for inspectors to aid in their documentation.

The development could potentially work to benefit the enormous task of maintaining the health of structures that are increasingly being reused or restored in cities large and small across the country. Despite the relative age of America’s built environment, roughly two-thirds of today’s existing buildings will be in use in the year 2050, according to Gensler’s predictions.

Antoni Gaudí’s Casa Batlló overlaid with AI-informed artwork by Sofia Crespo

2024-01-30T11:46:00-05:00

Antoni Gaudí’s Casa Batlló in Barcelona has served as the host for a major projection mapping artwork by Sofia Crespo over the weekend. Titled Structures of Being, the event drew 95,000 visitors across its showing on January 27th and 28th.

Image credit: Claudia Maurino

Created by Crespo with music by Robert M. Thomas, the project is described by organizers as “a spectacular setting that immersed the public in the nature contained inside Casa Batlló as well as in Gaudí's creative universe.” Created with the aid of artificial intelligence and the Barcelona Supercomputer Center, the project used data on the behavior of marine currents.

Image credit: Claudia Maurino

The artwork draws on Crespo’s interest in the boundaries of human perception in addition to her research into Gaudí’s own sources of inspiration. The resulting methodology saw Crespo use artificial neural networks to extract visual patterns from data inspired by the bra...

Refik Anadol creates ‘world first’ AI model trained on data from nature instead of humans

2024-01-15T11:35:00-05:00

Refik Anadol Studio has announced the creation of the “world’s first open-source generative AI model dedicated to nature.” Titled 'Large Nature Model,' the model is trained “solely on nature’s inherent intelligence” as opposed to the conventional approach to AI large language model training, which consists of human-generated data such as art and text.

To train the model, Anadol’s studio used open-access information from institutions including the Smithsonian Institution and London’s Natural History Museum alongside various other universities, museums, foundations, government entities, and libraries. The studio is also visiting 16 rainforest locations around the world, deploying data collection technologies such as LiDAR, photogrammetry, and the capturing of ambisonic audio and high-resolution visuals of diverse ecosystems to further train the model.

“Our vision for the Large Nature Model goes beyond being a repository or a creative research initiative,” said Refik Anadol about the m...

Cambridge researchers develop AI model to help retrofit and decarbonize housing

2023-11-02T11:45:00-04:00

Researchers from the University of Cambridge have unveiled a “first-of-its-kind AI model” that can help policymakers identify and prioritize houses for retrofitting and other decarbonizing measures. The deep learning model, trained by researchers from the university’s Department of Architecture, promises “to make it far easier, faster, and cheaper to identify these high-priority problem properties and develop strategies to improve their green credentials,” the team says.

The model and wider research centers on ‘hard-to-decarbonize’ (HtD) houses that, while responsible for a quarter of all direct housing emissions, are rarely identified or targeted for improvement. According to the team, the age, structure, location, availability of data, and socioeconomic context can all lead to a house being classified as HtD.

In response, a team including urban researcher and data scientist Maoran Sun and the university’s Sustainable Design group lead, Dr. Ronita Bardhan, has developed their AI mo...

Refik Anadol's Machine Hallucinations generative AI piece acquired by MoMA

2023-10-11T15:59:00-04:00

Refik Anadol has made history as one of two prominent digital artists to have their works acquired into the permanent collection of the 94-year-old Museum of Modern Art (MoMA) in New York City.

The museum announced its acquisition of Anadol’s 2022 Generative AI piece Unsupervised – Machine Hallucinations, along with the blockchain-based piece 3FACE. 2022 from artist Ian Cheng, this week. Anadol's work was a gift to the museum by the 1OF1 Collection (which is led by Ryan Zurrer) and the RFC Collection, which is led by Pablo Rodriguez-Fraile and Desiree Casoni.

Writing on Instagram, Anadol commended the museum’s curatorial decision, saying it “inscribes a historic chapter as MoMA welcomes the first Generative AI and tokenized artwork into its esteemed permanent collection.” The work was one of three made for last year’s special solo exhibition using a machine learning algorithm to interpret publicly available data taken from the museum’s collection to produce reinterpreted images that...

Ohio State researchers use machine learning to create 3D digital models of lost historic neighborhoods

2023-07-21T12:38:00-04:00

A team of researchers from The Ohio State University has developed a machine learning technique that converts old urban maps into three-dimensional digital models. According to the team, the models could potentially revolutionize research involving historic neighborhoods and the economic impact of their demolition.

The study, recently published in the journal PLOS ON, involved extracting and digitizing data from Sanborn Fire Insurance maps. The maps, created during the 19th and 20th centuries and frequently updated, were used by fire insurance companies to estimate their liabilities in about 12,000 U.S. cities and towns.

"We now have the ability to unlock the wealth of data that are embedded in these Sanborn fire atlases," said Ohio State Geography Professor Harvey Miller, a co-author of the study. "It enables a whole new approach to urban historical res...

Archinect's Lexicon: "Semanticism"

2023-06-16T12:03:00-04:00

Archinect's Lexicon focuses on newly invented (or adopted) vocabulary within the architectural community. For this installment, we're featuring a term that was recently presented in an Archinect feature article on the architectural implications of generative AI tools such as Midjourney and DALL-E.

"Semanticism" is a term used within the realms of architecture and design, describing the theoretical approach that emphasizes the significant role of semantics, or meaning, in shaping architectural forms. It postulates that the forms and structures within architecture are shaped by their intended meanings and context, aligning with the dictum "form follows meanings." The approach dictates that the study and creation of architecture necessitate the unraveling of the various significations inherent in built forms.

Under the purview of Semanticism, the construction of new architecture corresponds to the crafting of forms that best express a design's intended meanings within its specific cont...

Archinect's Lexicon: "Hallucination"

2023-06-09T11:22:00-04:00

Archinect's Lexicon focuses on newly invented (or adopted) vocabulary within the architectural community. For this installment, we're featuring a term that has risen to prominence with the advent of generative AI tools such as ChatGPT and DALL-E.

"Hallucination" in the context of artificial intelligence refers to the phenomenon where an AI model generates or interprets something that isn't grounded in the actual input data it was given. This can happen when AI is asked to create content such as text, images, or speech. The AI may infer patterns or details based on its training but doesn't truly comprehend the information it is handling. It can produce results that seem plausible but might not correspond accurately with the provided input or real-world context.

In the field of building design and architecture, AI hallucinations might be relevant in several ways:

Genera...

SWAPP raises $11.5 million for tool that offers ‘AI-powered construction documents in minutes’

2023-05-23T11:27:00-04:00

The Israeli startup SWAPP has raised $11.5 million in funding for its solution for the streamlined creation and management of architectural construction documents. The platform, which has now raised $18.5 million since its founding, describes itself as providing clients with “AI-powered construction documents in minutes.”

SWAPP was founded by Noam Gat, Adi Shavit, and Eitan Tsarfati, who sold their last venture CmyCasa to Autodesk. Headquartered in Tel Aviv with offices in Houston, the company currently markets its software to the multifamily and education design sector. According to its founders, SWAPP’s AI platform integrates with BIM software such as Revit and automatically generates construction documents that can make automatic model-wide revisions.

"The product we developed at SWAPP tackles the unfairness I faced as an architect on a daily basis - the excessive amount of time spent on creating intricate plans, rather than concentrating on the design aspect, and the lack of co...

Autodesk unveils Forma, an AI-driven tool for generating and analyzing BIM models

2023-05-08T11:28:00-04:00

Autodesk has unveiled a new software program that integrates automation and artificial intelligence-powered insights. Autodesk Forma is described by the company as an “industry cloud” that will use AI to “simplify the exploration of design concepts, offload repetitive tasks, and help evaluate environmental qualities surrounding a building site, giving architects time to focus on creative solutions.”

Forma’s initially-released capabilities target the early-stage planning and design process, with new features and capabilities to be added on a continual basis. For existing Autodesk subscribers, Forma is included in the Autodesk AEC Collection.

Among Forma’s initial capabilities is contextual modeling, which allows users to set up a 3D BIM model of a project area “within minutes” in Forma by either inputting their own design data or buying data sets for selected geographies. Forma’s automation functions will also allow users to use parametric tools to “quickly develop many design conce...

Introducing the ‘Archinect In-Depth: Artificial Intelligence’ series

2023-05-01T08:00:00-04:00

Quick links

At Archinect, we are excited to announce the launch of Archinect In-Depth, a new initiative examining the most pressing, transformative, and integral subjects in the architectural community. Each series of Archinect In-Depth will feature three months of dedicated coverage across our platform, addressing each subject's relationship with architectural practice, employment, education, research, and beyond.

Our initiative kicks off today, May 1st, with the inaugural series Archinect In-Depth: Artificial Intelligence. Over the next three months, we will explore the ever-evolving relationship between architecture and artificial intelligence through dedicated news, features, interviews, calls for submissions, and a storytelling competition, with input from both Archinect’s editorial team and ou...

AI could help create more hurricane-resistant buildings, NIST research says

2023-04-14T15:06:00-04:00

Researchers from the National Institute of Standards and Technology (NIST) have devised a new method to digitally simulate hurricanes. Using data derived from 100 years of hurricane monitoring, infused with modern AI techniques, the researchers suggest that simulating the trajectory and wind speeds of storms can help develop improved guidelines for the design of buildings in hurricane-prone regions.

The research, published recently in Artificial Intelligence for the Earth Systems, envisions a scenario where designers are directed to standardized maps by their local building codes. On these maps, designers can find the level of wind their proposed structure must handle based on its location and relative importance. For example, the resilience standards of a hospital would be regarded as higher than that of a self-storage facility.

The wind speeds in the maps are derived from scores of hypothetica...

Refik Anadol repurposes his AI data sculpture 'Machine Hallucinations' for 2023 Grammy Awards stage

2023-02-07T18:54:00-05:00

Istanbul-born media artist and design innovator Refik Anadol has quickly become a household name with his mind-bending, data-driven art. This year, Anadol continues to push what's possible during his visual backdrop debut at the 65th annual Grammy Awards.

"The collaboration with the Recording Academy started almost six months ago with an exciting call from the designers who are in charge of the stage design and the whole flow of the ceremony," Anadol shared with Archinect.

Anadol's work can be seen in the background illuminating the space during the award's premier ceremony speech by the Recording Academy's CEO Harvey Mason Jr. Video courtesy of Recording Academy/Grammys.

Along with the event's lead production designer Julio Himede, the art direction team, and electronic graphics expert Hugh Grew, Anadol's distinct data-generated paintings from his AI artwork and NFT collections, Machine Hallucinations — Nature Dreams and Machine Hallucinations: Space Metaverse, could be spotted ...

Now Google Maps can see inside buildings thanks to AI

2022-05-13T12:06:00-04:00

Google launched a new mode for Maps on Wednesday, designed to give users a more real-life look at the places they’re going before they even go. The new Immersive View is sort of a Street View in the sky: you can look over a location from above to get a sense of the neighborhood and then drop to street level to see the specific spots you might want to hit up.

The feature was revealed at Google’s I/O 2022 Keynote address in which CEO Sundar Pichai also introduced a new flood forecasting feature and expanded capabilities of the 15-year-old Maps project made possible through advancements in AI technology. Pichai says the company now has 1.6 billion buildings and over 60 million kilometers of road to date as part of the project.

The Immersive View feature will now allow the interiors of a good portion of them to be seen using a machine learning technique called neural rendering. Pichai used London's Westminster area as an example and said the feature is useful in providing pedestrian and vehicle congestion information to users in real-time.

The 10 most promising construction startups of 2021, from AI contract reviewers to cloud-based material management

2021-11-24T14:02:00-05:00

The ten winners of the Construction Startup Competition 2021 have been announced by CEMEX Ventures, recognizing the most promising startup companies in the construction sector for this year. The overall winner of the competition was Document Crunch, an AI-powered software which can identify and explain critical issues in construction contracts.

Based in Atlanta, Georgia, Document Crunch allows users to upload documents online to be reviewed by an AI software to analyze risk, understand waivers, and highlight indemnity and insurance obligations. As well as reviewing construction contracts and insurance policies, the software can also analyze plans, drawings, and specifications, categorizing elements such as design responsibility, confidentiality requirements, hazardous materials, and warranty obligations.

Document Crunch was one of ten winning projects in the competi...

ETH Zurich unveils robotically-assembled sculpture inspired by the Hanging Gardens of Babylon

2021-11-23T16:59:00-05:00

Researchers at ETH Zurich in Switzerland have offered an insight into the design and fabrication of a new architectural sculpture, built with the help of artificial intelligence and four robots. The structure, standing 22.5 meters (74 feet) in height, will consist of five geometrically-complex wooden pods all slightly offset from each other and supported by eight thin steel pillars.

Image: Gramazio Kohler Research, ETH Zürich

Titled "Semiramis" after the Babylonian queen associated in folklore with the Hanging Gardens of Babylon, the sculpture is currently being manufactured at ETH Zurich’s Robotic Fabrication Laboratory. In what the team describes as a “delicate dance for the highest precision,” the wooden pods are being assembled by four suspended robotic arms, each collecting wooden panels and placing them in accordance with a computer-generated design.

Image: Pascal Bach / Gramazio Kohler Research, ETH Zurich

Once the robotic arms have placed four panels next to each other, cra...

Researchers have mapped every large solar plant in the world

2021-11-01T14:19:00-04:00

Researchers from the United States and United Kingdom have used machine learning to map every large solar plant in the world. The team behind the map sees it as an opportunity to consider the future trajectory of solar expansion and to inform decisions on what land uses can be best supplanted by solar farms by looking at historic trends. The findings were published in the journal Nature by researchers from the University of Oxford in the UK, Descartes Labs Inc in San Francisco, and the World Resources Institute in Washington DC.

A map of all the large solar facilities detected up to 2018 (lighter colors = more recent). Image: Kruitwagen et al, Nature via The Conversation

The map was made possible by the construction of a machine learning system which could detect solar facilities in satellite imagery; a system that the researchers scaled to a planet-wide search using 550 terabytes of imagery. To streamline their results, the researchers limited their search to solar facilities which ...

New York City as the world's biggest apartment

2020-03-09T13:28:00-04:00

Artists Sam Lavigne and Tega Brain and the Whitney Museum of American Art have collaborated to produce New York Apartment 2020, a mesmerizing website experience that "advertises a fictitious New York City apartment for sale that covers more than 300 million square feet and spans the five boroughs."

The piece, which exists as an interactive website, combines thousands of real estate listings into one massive aggregated apartment that features 65,764 bedrooms and 55,588 bathrooms. The mega-apartment clocks in at a cozy 36,672,535 square feet in size.

According to project text showcased on the Whitney Museum website, the artwork "collapses the high and low ends of the market, architectural periods and styles, and neighborhoods and affordability into a single space that cumulatively creates a portrait of New York's living spaces and the real estate market."

Screenshot of New York Apartment 3D virtual tour. Image courtesy of The Whitney Museum of American Art.

For the project, the artis...

Amazon opens first cashierless grocery store in Seattle

2020-03-09T11:19:00-04:00

Amazon [has opened] its first grocery store to pilot the use of the retailer’s cashierless “Just Walk Out” technology that has previously powered 25 Amazon Go convenience stores in a handful of major U.S. metros. Based in Amazon’s hometown of Seattle, the new Amazon Go Grocery store allows customers to shop for everyday grocery items like fresh produce, meat, seafood, bakery items, household essentials, dairy, easy-to-make dinner options, beer, wine and spirits, and more.

According to TechCrunch, the store is 7,700 square feet in the front of house and 10,400 square feet overall, making it the largest use of Amazon's Just Walk Out technology to date. With a similar model to the Amazon Go convenience stores, shoppers use the Amazon Go app to check themselves in as they enter the store. Once inside, users shop as they typically would. Sensors and cameras monitor the items removed from shelves and add them to the shopper's virtual cart. Once the shopper exits the store, the software charges the card the customer has on file, TechCrunch reports.

Artificial Intelligence helps mapping urban trees (all of them)

2019-01-07T14:19:00-05:00

To train the model, he identified known locations of tree canopy using lidar data and NAIP imagery over California. Using that as ground truth, the model was trained to classify which pixels contain trees in the corresponding satellite images. The result is a machine-learning model that has learned to identify trees just using four-band high-resolution (~1 meter) satellite or aerial imagery—no lidar required!

Former New York Times cartographer Tim Wallace describes how his current firm, Santa Fe-based Descartes Labs, has built a machine learning model to identify tree canopy from satellite imagery thus making accurate mapping of trees and urban forests far more accessible to cities worldwide.

San Francisco Open Forest Map tree inventory (point data) in comparison with the Descartes Labs tree canopy layer (image data). Image: Tim Wallace/Descartes Labs

"The ability to map tree canopy at a such a high resolution in areas that can’t be easily reached on foot would be helpful for utility companies to pinpoint encroachment issues—or for municipalities to find possible trouble spots beyond their official tree census (if they even have one)," writes Wallace. "But by zooming out to a city level, patterns in the tree canopy show off urban greenspace quirks. For example, unexpected tree deserts can be identified and neighborhoods that would most benefit from a surge of saplings revealed."

New York...

Autonomous crane technology receives investment boost

2018-09-27T14:52:00-04:00

With industrial robotics forecast to be worth $71.72 billion by 2023, it’s no wonder entrepreneurs are turning their attention to increasingly lucrative sectors, like warehouse automation, order fulfillment, and manufacturing. Tel Aviv-based Intsite is one of the latest examples. The startup today announced a $1.35 million pre-seed round led by Terra Venture Partners and the Israel Innovation Authority to fund what it claims is the world’s first autonomous crane technology.

Image: Intsite

AI-powered autonomous construction technology is poised to see enormous growth in the coming years, promising to significantly increase efficiency, cut costs & realization time, and reduce human errors as well as workplace-related injuries.

"According to McKinsey, about 98 percent of construction mega-projects go significantly over budget, in part as a result of heavy equipment inefficiencies," reports Venture Beat.

This AI-powered & LiDAR-equipped robot could soon help detect construction errors early

2018-02-23T15:39:00-05:00

Using lidar-equipped robots, Doxel scans construction sites every day to monitor how things are progressing, tracking what gets installed and whether it’s the right thing at the right time in the right place. You’d think that construction sites would be doing this by themselves anyway, but it turns out that they really don’t, and in a recent pilot study on a medical office building, Doxel says it managed to increase labor productivity on the project by a staggering 38 percent.

"You could send in some humans with lidar backpacks, but that would be more expensive," IEEE Spectrum explains. "The company is also using drones in a limited capacity right now, since they require human supervision, but it’s easy to imagine how much more efficient this process could get as robotic autonomy improves."

AutoDraw, Google's newest A.I. Experiment, guesses what you're trying to draw

2017-04-17T13:29:00-04:00

You simply start drawing your best version of a pizza, or house, or dog, or birthday cake and the algorithms try to figure out what it is that you’re trying to draw. It then tries to match your squiggles with drawings in its database, and if it finds any possible matches, it’ll show them in a list at the top of your virtual canvas. If you like one of those options, you simply click on it and AutoDraw replaces your amateurish creation with something a bit slicker.

The new AutoDraw tool is part of Google's A.I. Experiments sandbox and pairs machine learning with artist drawings from a growing, crowd-sourced library. "AutoDraw’s suggestion tool uses the same technology used in QuickDraw, to guess what you’re trying to draw," states the tool's About page. "Right now, it can guess hundreds of drawings and we look forward to adding more over time."

While simplicity and intuitive accessibility are the striking features of this early version, AutoDraw—and a universe of variations—show a lot of potential for architectural sketching and prototyping.

Amazon's newest venture promises to shake up retail design—and eliminate the need for cashiers altogether

2016-12-06T12:42:00-05:00

“No lines. No checkouts. No registers. Welcome to Amazon Go.” The newest “disruption” offered by Silicon Valley promises to radically shake up retail design in the name, per usual, of increased efficiency. Located in Seattle, the Amazon Go store is a market without cashiers. Instead, shoppers simply download an app that gives them access to the store. As they pick up items, their virtual shopping cart is automatically adjusted. When they leave the store, a bill is sent to their Amazon account. Amazon Go uses computer vision, sensor fusion, and deep learning—a technology bundle Amazon has dubbed ‘Just Walk Out’—to track the shopping experience.

Of course, this newest innovation could radically change the way we design retail environments. It also means Amazon will know even more about our shopping preferences and be able to tailor advertisements ever more precisely. Perhaps most importantly, Amazon Go threatens to eliminate the need for cashiers. Currently, 3.4 million Americans work ...

Jobs are for machines, and life is for people.

2016-03-25T11:44:00-04:00

These exponential advances, most notably in forms of artificial intelligence, will prove daunting for as long as we continue to insist upon employment as our primary source of income. The White House, in a stunning report to Congress this week, put the probability at 83 percent that a worker making less than $20 an hour in 2010 will eventually lose his job to a machine. Even workers making as much as $40 an hour face odds of 31 percent.