Hyperedge- . IoT, Embedded Systems, Artificial Intelligence,
AI is beginning to be employed to help manage a mix of robots, drones, cameras, and sensors on construction sites, as the industry rolls out more automation. (Credit: Getty Images) 

By AI Trends Staff 

Robots and drones directed and managed with the help of AI are appearing on construction sites, as the construction industry gains experience with uses ranging from improving site security to management of multi-year projects.   

Such large scale, multi-year projects require the coordination of many complicated tasks and moving parts including designs and blueprints, permits, and unexpected delays and changes. These can quickly challenge the ability of humans to manage without the assistance of technology, according to a recent account in Forbes.  

For scheduling, the use of advanced AI can help to prevent costly delays on sites or among suppliers, vendors, and others involved in the process. AI is also being applied to contingency planning, running through multiple scheduling scenarios for example if a permit is delayed or an incident happens.   

Drones are being used for surveying and taking overhead images of construction sites through stages of construction. Robots are being used to help with tasks such as bricklaying, pouring concrete, or installing drywall, augmenting the labor force to help with labor costs and to keep the project on schedule.  

Increasingly, construction sites are being equipped with cameras, IoT devices, and sensors that monitor many aspects of construction operations. AI-enabled systems are able to watch what is going on 24/7 without distraction.   

Startups Eye Opportunity; Built Robotics Moves Earth 

Hyperedge- . IoT, Embedded Systems, Artificial Intelligence,
Gaurav Kikani, VP of Strategy, Built Robotics

A number of startups are taking advantage of this opportunity. Built Robotics, for example, got its start using autonomous skid-steer loaders to move sand or gravel on construction sites. Today the company has autonomous systems for bulldozers and 40-ton excavators. 

 “We have a software platform that actuates the equipment that takes all the data being read by the sensors on the machine every second to make decisions and actuate the equipment accordingly,” stated Gaurav Kikani, VP of strategy for Built Robotics, in an account in VentureBeat.  

Built has focused on earth moving projects at remote job sites in California, Montana, Colorado, and Missouri that are far removed from human construction workers. Autonomous heavy equipment monitored by a human overseer tills the earth in preparation for later stages of construction, when human crews arrive to do things like build homes or begin wind or solar energy projects. In the future, the startup wants to help with more infrastructure projects. Built raised $33 million last fall in a series B round, making its total funding $48 million, according to Crunchbase.  

Built started out focusing on standalone activities at a site, with one machine working alone to complete a job, and later moved into excavators and smaller dozers working together. The dozers would push material away or create space for the excavator to be more productive. 

Software Needed to Manage “Morphologies” on Construction Sites  

“The fleet coordination element here is going to be critical. Realistically, to get into the heart of construction, I think we’re going to start to coordinate with other types of equipment,” Kikani stated. “The trickiest thing about construction is how dynamic the environment is. Building technology that is pliable or versatile enough to account for the changing conditions, and being able to update in real time to plan to accommodate for that is going to be the key here.”  

Computer vision systems are being used to track progress on construction sites. Startup Indus.ai, among a handful of companies in the business, offers cameras to track the flow of trucks entering a site, the number of floors completed in a building and the overall pace of progress. It is capable of following daily work production and helps supervisors determine whether the work of individuals and teams follows best practices.  

Hyperedge- . IoT, Embedded Systems, Artificial Intelligence,
Matt Man, CEO, Indus.ai

“We can observe and use a segmentation algorithm to basically know every pixel—what material it is—and therefore we know the pace of your concrete work, your rebar work, your form work and [can] start predicting what’s happening,” stated Indus.ai CEO Matt Man to VentureBeat. 

He envisions a mix of working humans and machines collaborating on construction sites. “There could be armies of robot-building things, but then there is an intelligent worker or supervisor who can manage five or 10 robotic arms at the same time,” Man stated.  

Software for directing the on-site activity will become more critical as contractors embrace robotics, in his view. “Having all these kinds of logistical things run together really well, it’s something I think AI can do,” Man stated. “But it’s definitely going to take some time for the whole orchestration to be done well.”  

Hyperedge- . IoT, Embedded Systems, Artificial Intelligence,
Brian Ringley, construction technologist, Boston Dynamics

Boston Dynamics, known for years as the maker of cutting-edge robots, also entered construction sites last year as part of its transition from an R&D outfit to a commercial company. Boston Dynamics construction technologist Brian Ringley echoed the notion that software platforms will emerge to coordinate multiple machines on construction sites. 

“In  the same way we use lots of different people and lots of machines on sites now to do things, I believe there will be multiple morphologies on construction sites and it will be necessary to work together,” Ringley stated.  

That seems to be happening. For example, the International Union of Operating Engineers, which has over 400,000 members, last spring established a multi-year training partnership with Built Robotics.   

Read the source articles in Forbes and VentureBeat. 

Source: AI Trends