From cars to construction: Automobile technologies could make your job site safer
By now you have surely driven in cars that illuminate your side view mirror when someone is in your blind spot, vibrate your steering wheel when you stray out of your lane and beep when you’re about to back up over your trash can. Your car might even have cruise control functions that automatically regulate your speed and braking based on how close you are to other vehicles. Cars use a combination of cameras and sensors to determine how far an object is from your bumper. The sensor is constantly analyzing the camera’s video feed in real time and alerts you with a vibration, beep or flashing light when you are too close to something.
If having this technology in your car has become novel, maybe it’s time to incorporate it on your construction site. While cameras and sensors are mostly used on job sites for security, these car technologies could monitor a whole range of things to maintain quality, efficiency and safety. But let’s focus on safety for now since one in five worker fatalities occur in construction. Cameras and sensors strategically placed on buildings, vehicles and vests, gloves and hard hats could help minimize the Fatal Four:
- Falls: Sensors could warn a worker if they are about to walk into a hole or sense when a guardrail is broken or missing. It could let someone know when a ladder is being used too far from the work that needs to be done so somebody doesn’t overreach and fall. Workers could also be reminded when they should be tied off and that they shouldn’t jump across scaffolding.
- Electrocutions: Sensors could tell electricians when an unsafe electric current is running through scaffolding near them or if there’s a live wire on site. They could notify someone if an electrical panel was ajar or if wire nuts or electrical tape aren’t appropriately adhered. Sensors and backup cams could also alert a crane operator when they are working too close to power lines.
- Struck by object: Wearable technology that uses sensors and cameras could vibrate when the worker is in the path of a moving object or vehicle. They could even cut the ignition switch if that vehicle was about to hit something or someone. This technology could also alert a crane or excavator operator when someone or something was in their blindspot. Sensors can also make sure cranes and other machinery are safely grounded.
- Caught in/between objects: This technology could automatically turn off a scissor lift that was about to trap someone against a ceiling. They could also alert someone if they are between two objects that could potentially pin them.
At the same time, 360-degree cameras with sensors could be mounted to a safety manager’s hardhat to literally give them eyes in the back of their head. The sensors would not only alert them if something outside their periphery was amiss or dangerous, but they would have the ability to record and survey the site to review later.
Still not convinced that this is ready for primetime? Well, the biggest proof point that these car technologies can be incorporated into wearable technologies for construction is Toyota’s Project BLAID. Worn over the shoulders, this device for blind people uses cameras and sensors to detect objects in the user’s surroundings the same way cars do. BLAID has speakers and vibration motors that help users locate bathrooms, escalators, stairs and doors. Given the fact that Toyota successfully migrated these cameras and sensors from cars to wearables, it’s easy to imagine how this technology could be used on a construction worker to help make the job site safer.
Driverless cars to construction
Toyota says Project BLAID will soon incorporate technologies being tested and perfected in driverless cars such as object recognition software. But what can object recognition do for construction?
Well, object recognition has the potential to make a job site safer in all the ways we’ve already mentioned, but do it much more effectively and intelligently. Cameras and sensors alone can only alert someone when they are too close to a dangerous object. But they don’t know what that object is. Object recognition software, however, works hand-in-hand with cameras and sensors to differentiate between objects such as people and piles of debris. If the software detected that a person and not a pile of dirt was in the blind spot of an excavator, it could cut the engine or maneuver around the person.
The software could also be programed to recognize objects such as hard hats and tethers in order to alert a safety manager if someone wasn’t tied off or was missing Personal Protective Equipment. Object recognition learns exactly what and where items should be in relation to humans, materials and vehicles. And all this information could be plugged into a BIM model so that a safety manager would have a clear picture of where all the hazards on the job site are in real time. So while the cameras and sensors in your car today could provide major benefits on your construction site, the cutting-edge technologies in autonomous vehicles would be even more groundbreaking.
Some final thoughts
One of the biggest challenges of object recognition software is that the computer has to learn how to identify the objects you want it to recognize. It has to learn when it should react to unsafe circumstances. This requires physically scanning hard hats, tethers and other items in a variety of positions so the sensors and cameras know when something is awry. This cataloging of objects so to speak is much simpler for automobiles to achieve because cars travel on fixed paths with fairly predictable hazards such as moving cars and objects jutting out in front of the vehicle. Cars can easily measure the distance between themselves and the curb so they know how to react. A construction site is much more dynamic. The materials, equipment and personnel change from day-to-day at an amazing pace. The building itself is constantly evolving too.
But not all of these objects need to be entered into the image recognition software immediately. You could start with hardhats and tethers, for example, to get the project off the ground. Then you can keep adding items such as toe boards and guardrails to your library as you move forward. Alternatively, you could start by utilizing sensors that are in our cars today for many of these functions and wait for object recognition technology to mature into the mainstream.
Another reason it’s crucial to roll something like this out incrementally is that cameras on a construction site are likely to be met by skepticism and privacy concerns. But people will be more willing to support the changes if the benefits are clearly demonstrated through a pilot program. And the process of testing and evaluating the pilot’s results should be shared with everyone. In the end, these technologies should only be used for the betterment and safety of all. This technology could be life saving, after all, so it’s easy to imagine scenarios in which the benefits outweigh the disadvantages.
Which brings us to one final thought: Google and the ultra competitive car manufacturers currently have a stranglehold on many of these technologies. But while they haven’t shared their trade secrets with companies in other industries to-date, eventually someone will replicate these systems for others to use. When they do, there’s no reason construction can’t be one of the first industries to adopt this transforming technology.
This post was a collaboration between Suffolk Construction’s Executive Vice President & Chief Innovation Officer Chris Mayer, and Content Writer Justin Rice. If you have questions, Chris can be reached at firstname.lastname@example.org and Justin can be reached at email@example.com or follow him on Twitter at @JustinAlanRice.