Robots in the forest: sustainable timber houses for all

Robots in the forest: sustainable timber houses for all

Timber houses made entirely by robots could make construction more accessible and less capital and resource intensive.

At a factory in East London, timber sheets are processed into joists and cladding panels.

Holes are drilled and glue applied to turn the panels into a block. The flooring is fastened, walls erected and a roof put into place. Window frames are structured and glazing units inserted into the frame. All these components are then assembled to build houses of different sizes, from one-storey garden studios and office pods to two-storey homes.

But there’s no human builder to be found anywhere.

This is because these modular dwelling units, designed by tech start-up Automated Architecture (AUAR), are constructed entirely by robots.

The use of Lego-like prefabricated plywood blocks – function agnostic with each role established after assembly – helps break down what is often a lengthy and complicated building process into small tasks which pre-programmed robots can perform.

This, in turn, cuts costs, delivery time and the environmental impact.


An AUAR house costs 20 per cent less than the market average and can be delivered in just six weeks. The system reduces “embodied” carbon emissions by 80 per cent compared with traditional construction methods. They are the emissions that occur in the first couple of years – the construction phase of a project that involves extraction of raw materials, manufacturing, transportation, installation, and disposal of waste. The reduction is thanks to the use of timber and other bio materials, local manufacturing and a short production chain.

Faster, more sustainable construction techniques such as those developed by AUAR could soon become the default choice.

Not only because the world needs to build an estimated 2 billion new homes by the end of the century, but also because the construction industry is responsible for 40 per cent of global carbon emissions.

“We need a fundamental rethink around housing… As a society, we can shoot rockets to the moon, but are no longer able to imagine that we can build better,” Gilles Retsin, co-founder and chief architect of AUAR, tells mega in an interview.

“As building is slow and expensive, only a limited number of actors can take the risk to construct. This in turn keeps the market limited and scarce in supply of housing and puts the decision-making of housing construction in the hands of the few – government and large developers.”


AUAR – a spin-off from the Bartlett School of Architecture at University College London – uses artificial intelligence and machine learning to optimise design, increase material efficiency and cut costs.

It is an approach Retsin says should help transform what is an outdated and resource-intensive industry that suffers permanent labour shortages.

Construction is in many ways stuck in a time warp. The average building is constructed using over 7,000 different parts which need to be assembled into a functional whole. Globally, productivity growth in the construction industry has flatlined in the past two decades, in stark contrast to manufacturing whose productivity expanded by 3.6 per cent in the same period.1

“(The use of prefabricated blocks) could make construction more accessible, faster and therefore less capital intensive, opening the market to a larger group of housebuilders,” says Retsin, who is also an associate professor at the UCL Bartlett School of Architecture.

“We hope to empower developers, contractors and architects to build better and encourage a new group of people to start to build homes.”

Robots and AI ultimately bring us back to how we used to build in the past, with local materials such as timber, extracted directly from the forest.

Assemble, disassemble, repeat

Another distinctive aspect of AUAR’s building process is that it uses an environmentally friendly engineered wood product called cross-laminated timber (CLT).2

AUAR’s prefabricated sheets are also designed to be repurposed, allowing buildings to continuously adapt and meet future demands – thus embracing the concept of a circular economy.

“While timber is an obvious choice in the context of the global climate emergency, it also has the potential to transform construction into an automated industry. Timber currently already has a high degree of automation throughout its entire production chain,” Retsin says.

The market for CLT is expected to grow to a USD2.5 billion globally by 2027 from USD1.1 billion in 2021, an annual increase of some 15 per cent.3

Some experts reckon it could grow even more rapidly after landmark revisions to the International Building Code lifted the height limit for wood buildings to 18 stories in 2021 from the previous seven.

AUAR is setting up a factory close to Ghent, Belgium where five robots will build 40 houses over the next year.

Its London factory doubles as a research lab where robots are programmed and the building systems updated and tested for delivery.

The company is working with strategic investors in the property sector to roll out the project to other parts of Europe and North America and expand the offering to six-storey multi-family homes. It aims to deliver 10,000 net zero homes per year by 2032.

“Robots and AI ultimately bring us back to how we used to build in the past, with local materials, such as timber, extracted directly from the forest,” Retsin says.

“We no longer need the long, polluting, wasteful, global supply chains from the past that have eventually led to the climate crisis. Robots and AI will help us build a future that is local, sustainable and beautiful.”

[1] McKinsey research
[3] Markets and markets
Please confirm your profile
Please confirm your profile to continue
Or select a different profile
Confirm your selection
By clicking on “Continue”, you acknowledge that you will be redirected to the local website you selected for services available in your region. Please consult the legal notice for detailed local legal requirements applicable to your country. Or you may pursue your current visit by clicking on the “Cancel” button.

Welcome to Pictet

Looks like you are here: {{CountryName}}. Would you like to change your location?