Skip to Main Content

Strategic Business Insights (SBI) logo

Old Robot, New Tricks P1745 February 2022

Author: David Strachan-Olson (Send us feedback.)

Industrial robots still underpin the robotics industry; new control technologies could make them useful in more aspects of manufacturing.

Abstracts in this Pattern:

Despite the media's focus on cutting-edge robots, the core of the robotics industry remains industrial robots. Industrial robots have existed for decades but continue to find new applications. Shipments of industrial robots have increased at a significant pace during the past decade, and the prevalence of industrial robots in manufacturing environments is increasing. Development of new AI algorithms for robot perception and planning could help further accelerate this trend.

The International Federation of Robotics (IFR; Frankfurt, Germany) studies the use of robots worldwide. One metric the IFR uses is robot density—the number of robots to the number of manufacturing employees. In a recent report, the IFR states that the global density of industrial robots has nearly doubled during the past five years, increasing from 66 robots per 10,000 manufacturing employees in 2015 to 126 robots in 2020. China accounted for a major portion of this change, increasing its robot density from 49 to 246 robots per 10,000 manufacturing employees. This increase in robot density highlights the broader adoption of automation technologies and the expanding uses for industrial robots.

Ongoing developments in the algorithms that control robots will further expand the uses for robots in manufacturing environments; algorithms for object manipulation are particularly important. Recently, researchers from MIT (Massachusetts Institute of Technology; Cambridge, Massachusetts) created a framework to allow a simulated robotic hand to manipulate and reorient more than 2,000 objects without prior knowledge of their shape. The researchers are hopeful that the algorithm will transfer to a real robotic hand.

A better understanding of animal decision-making could also help researchers make better algorithms for robots. A behavior study by researchers from the University of Konstanz (Konstanz, Germany), the Max Planck Institute of Animal Behavior (Max Planck Society for the Advancement of Science; Munich, Germany), and other institutions has uncovered a common algorithm that helps animals make decisions. The research suggests that animals break down complex decisions into a series of two‑choice decisions until only one option remains. The possibility exists that this bifurcation process could see application to robots.