Robots With Smooth Moves Are Up to 40% More Efficient

Image: Chalmers University of Technology

Robot arms (and robots in general) are all about following orders. Your orders, merciless human overlord. You tell them to jump, and they say “I’m an arm, I don’t jump, but I can move around a little bit if you want.” And then they do, as best as they can. As far as the arm is concerned, its entire reason for existing is to move where you tell it to as fast as possible, I guess because it figures (usually quite wrongly) that you have better things to do than sit there and mind it.

These fast, precise movements are one of the reasons that we like robots as much as we do, but as it turns out, they’re not particularly energy efficient. This might not be something that you think about after dropping tens of thousands of dollars (or whatever) on a robot arm, but energy use adds up, especially if you have tens of thousands (or whatever) of arms.

Swedish researchers at Chalmers University of Technology, working as part of the European Union’s AREUS Project (Automation and Robotics for European Sustainable Manufacturing), have taken a crack at robot arm efficiency, and come up with an optimization algorithm that tweaks acceleration and deceleration to reduce energy consumption by up to 40 percent.

Here’s the algorithm in action; what is most remarkable about this is how little it changes the speed of the arm, while keeping the same trajectory. The end of this video has an overlay of the original and optimized paths, and it’s almost impossible to tell the difference.

As the video shows, those very slight changes result in a 30 to 35 percent energy reduction, which is staggering. All the algorithm does is tell the arm to accelerate and decelerate a little bit more calmly, and huge energy savings result. I’d imagine that it’s also better for the electric motors, improving the longevity of the robot.

An additional 5 to 10 percent improvement in efficiency can be achieved if you’ve got multiple robots all working together, tweaking the motions of each to minimize downtime. As an added bonus, the robots don’t smash into each other, either:

This is all happening in a lab, of course: as the researchers point out, further studies in industry are required to really figure out how much of a difference this makes. Fortunately, the AREUS Project has backing from companies like Daimler and Kuka, so we’re pretty sure they’ll be able to execute on the necessary real-world testing.

Current results will be presented at the IEEE International Conference on Automation Science and Engineering (IEEE CASE) in Sweden later this month.

[ AREUS Project ]



IEEE Spectrum's award-winning robotics blog, featuring news, articles, and videos on robots, humanoids, drones, automation, artificial intelligence, and more.
Contact us:

Erico Guizzo
Senior Writer
Evan Ackerman
Jason Falconer
Angelica Lim

Newsletter Sign Up

Sign up for the Automaton newsletter and get biweekly updates about robotics, automation, and AI, all delivered directly to your inbox.