When we think “robot”, we often imagine humanoid figures like R2D2 from Star Wars or biologically manufactured machines like the replicants in Blade Runner 2049. And long before the science fiction genre existed, humans have been fascinated by artificial humans — from Frankenstein’s monster to Pygmalion’s Galatea.
We fall in love with fictional robots like WALL-E and follow the news about real life robots with artificial intelligence like Sophia. We look forward to the day when robots become the norm but the truth is, they already are.
Robots are already used in many industries like agriculture, retail and even healthcare. In fact, the market is booming. According to a report by McKinsey & Company, the total market for robotic systems amounted to $48 billion in 2017. This figure has grown and is expected to continue growing exponentially.
As machine learning has advanced, so has the use of robots in various industries. You may be imagining a humanoid-type robot wandering around a factory floor or picking apples from a tree, but if you’re trying to visualize the robot of today, think simpler and perhaps even look closer to home.
Your Google Home is a robot. Your Mi Robot Vacuum Cleaner is exactly what its name says it is. These may have been new additions to our households, but robots have been part of the industrial workforce for years.
An unnoticed team of workers
Industrial robots could be stand-alone arm-like structures that pick, weld and assemble items on a production line. They might be mobile robots that transport pieces from one side of the warehouse to the other. They are sometimes exoskeletons that can be strapped onto a human to boost strength.
These robots have contributed to increased efficiency, improved quality of products and increased profitability. They’ve also helped to create better working environments, being able to do tasks that are too risky for humans or providing support so that humans can do more valuable work.
Robots, when combined with machine learning algorithms, can be valuable team members to have. Imagine a robot that’s able to identify components or other objects, that can sort and pick these out in assembly lines and warehouses.
While it may be possible to program a robot based on precise item location, vision sensors and machine learning make it possible for the robot to “see” the correct items to pick without having to program it to account for precise location. This means that a particular robot can be utilized in a range of settings.
It could be used in a warehouse, picking specific items to be packed. Or it could be used on a production line — where its vision systems would allow it to fit one piece to another. It would be able to put the pieces into the right orientation without human assistance or over-specific programming.
Machine learning would also enable these robots to be used in the context of quality control. Where before it was a manual process that was both laborious and inaccurate, robots equipped with vision sensors and machine learning algorithms would be able to do real-time quality control. They would be able to identify inefficiencies on the factory floor, detect anomalies in products, errors in standard operating procedures and lead to overall improvement of work quality.
Robots with artificial intelligence
With the advancement in robotics and machine learning, there’s been an expansion to the industries, as well as the tasks that robots can be used to accomplish. A very important segment that they’ve been deployed in is the waste management industry.
As waste piles up, it becomes more inefficient and potentially dangerous for processes like waste separation to be done manually. This process is a vital part of maintaining the health of humans, animals and the environment as unsegregated waste would mean that all of it would end up at landfills, contaminating the land as it decomposes.
The industrial robots of today are extremely suited to perform waste separation tasks at a much higher efficiency. Equipped with computer vision and other sensors, powered by artificial intelligence, these robots are able to identify and separate waste materials, even being able to determine appropriate methods for disposal or recycling.
As robotics technology advances, they could even be used at landfills to recognize greenhouse gas emissions and help to prevent gas leaks. Or they could be the smart trash cans of the future, picking rubbish off the streets.
Robots could take over the waste management industry by 2030 and this would be good news. They would be able to handle the higher risk and more taxing activities in waste management. Humans would no longer have to choose between risking their health and protecting the environment.
The extent to which robots are able to improve human lives is boundless and it’s exciting to see the robotics industry growing at a rapid pace. Current developments in machine learning and artificial intelligence will fuel that growth even more. We’ll begin to see a feedback loop where growth in robotics creates a demand for machine learning models that will in turn provide insights into new applications for robotics.