ROBOTIC BUILDING CONSTRUCTION
A number of individuals and companies across the world are developing robotic construction systems. While these usually grab a lot of headlines, the reality of the situation needs closer study.
Lets' begin by understanding why people are interested in robotic construction:
Lets' begin by understanding why people are interested in robotic construction:
advantages of construction by robots
Robotic building construction has the following advantages:
All these factors make them much cheaper and more reliable than conventional construction in the long run. They eliminate the potential risks from accidents and lawsuits, assuming there are no humans on site.
- Robots can be extraordinarily precise in construction
- Robots can work 24 x 7 in a wide range of conditions
- They eliminate human problems in construction: not showing up for work, showing up drunk, mistakes, laziness, injury, the need for safety, the need for health insurance, fights with co-workers
- Specialized robots could work in small, dark, airless spaces that are unsafe for humans, or on the facades of tall buildings without facing vertigo
- They tend not to form unions
All these factors make them much cheaper and more reliable than conventional construction in the long run. They eliminate the potential risks from accidents and lawsuits, assuming there are no humans on site.
so do they actually work?
Lets look at a few building robots:
Construction Robotics is a US based company that has developed a bricklaying robot. Called SAM, or semi-autonomous mason, the robot has been used to lay bricks on a number of projects.
Another bricklaying robot from Aussie company Fastbrick Robotics
MIT Robot that 3D print or build foam structures that can in turn be used to make concrete walls.
Chinese architecture studio uses bricklaying robot to build facade with interesting wave-like ripples for an arts centre in Shanghai.
As you can see, most robots consist of long arms that can place brick, foam, or other materials accurately. While this is no doubt impressive, there is a long way to go before robots will completely replace construction by humans.
Let's now talk about at some of the problems faced by robots at construction sites:
Let's now talk about at some of the problems faced by robots at construction sites:
- Robots were originally developed for use in factories. They operate in highly controlled, clean, clutter-free environments, most often without humans in close proximity. A building site is another matter altogether. First of all, each site is laid out differently, so a robot would have to be intelligent enough to recognize where it is on the site and then move to the right location. Secondly, building sites are often chaotic, disorganised spaces with materials, tools, debris, and wires spread about. Many areas of the site will have unpaved soft soil, into which a robot will sink if it steps off the beaten path. There will be dust, rain, ice, and storms. There will be humans walking around. Getting a robot to work in these conditions is very difficult.
- Buildings nowadays are complex entities, and construction entails many different trades coming together to work in perfect sync with each other. The robots that you see now each perform one specialised task. It is impossible for a bricklaying robot to perform delicate tasks such as installing electrical cabling. Human dexterity, intelligence and situational awareness, developed over hundreds of millions of years, is currently far superior. Services like power cables, data and information cables of many kinds (voice, security, fire alarm, BMS) have to be woven through the fabric of the building. So a robo-mason must also be able to lay conduits for electrical wires, plumbing pipes, and must be able to perform waterproofing tasks.
so when will i move in to a building made by robots?
In our (sometimes humble) opinion, this will not happen before 2040 or so, when robots become a lot more street-smart, and develop some kind of 'general intelligence'. Robots will need enough situational awareness to be able to move around a construction site, and enough multi-tasking ability to work with a range of materials and techniques as described above.
The current crop of robots represent the industry taking its 'baby steps' towards solving the problem.
Until then, one is more likely to see architects using robots in novel ways, to fabricate interesting shapes and forms that cannot be done in the traditional way. Enjoy the examples below:
The current crop of robots represent the industry taking its 'baby steps' towards solving the problem.
Until then, one is more likely to see architects using robots in novel ways, to fabricate interesting shapes and forms that cannot be done in the traditional way. Enjoy the examples below:
ETH University Zurich develops robot construction
University of Stuttgart students develop web-like structure built by drones and robots
Artist Jenny Sabin builds canopy woven by robots in MOMA courtyeard
self assembling buildings?
Its possible that robotic construction will enable an entirely new approach. Rather than trying to make a robot emulate a human, what if swarms of robots were each given a repititive task, and limited intelligence, to build structures, just like ants build anthills?
There is no "master architect" or "project manager" in anthills. Each ant repeatedly adds material, following a very clear set of rules, and the anthill emerges from the collective efforts of the entire swarm of ants.
If we look at human cells, the information or blueprint to construct the cell is embedded within the structure of the cell, in the DNA. This enables the cells to make copies of themselves. It may be possible to develop building components that have some built-in geometry that allows them to link together to form buildings in a number of different ways, and perhaps allow humans to program them to link together two or three different ways, thereby allowing a variety of structures from the same module. In this approach, the robot does not build the structure, but is the structure, as the building is built of thousands of identical mini-robots. Watch MIT professor and head of MIT's self-assembly lab Skylar Tibbits explain his work in the following TED talk:
There is no "master architect" or "project manager" in anthills. Each ant repeatedly adds material, following a very clear set of rules, and the anthill emerges from the collective efforts of the entire swarm of ants.
If we look at human cells, the information or blueprint to construct the cell is embedded within the structure of the cell, in the DNA. This enables the cells to make copies of themselves. It may be possible to develop building components that have some built-in geometry that allows them to link together to form buildings in a number of different ways, and perhaps allow humans to program them to link together two or three different ways, thereby allowing a variety of structures from the same module. In this approach, the robot does not build the structure, but is the structure, as the building is built of thousands of identical mini-robots. Watch MIT professor and head of MIT's self-assembly lab Skylar Tibbits explain his work in the following TED talk: