Researchers develop universal controller for exoskeleton

A photo of a Georgia Tech researcher demonstrating exoskeleton technology.

Former Georgia Tech Ph.D. student Dean Molinaro (now an applied scientist at the Boston Dynamics AI Institute) walks up an adjustable ramp while wearing an experimental exoskeleton, demonstrating how researchers collected data in their effort to develop a unified control framework for robotic assistance devices. [Photo by Candler Hobbs/Georgia Tech]

Researchers at Georgia Tech say they developed a way to allow for the deployment of exoskeleton technology in homes, workplaces and more.

Plenty of engineers have sought to develop real-life robotic assistance that could protect workers from painful injuries and help stroke patients regain mobility. But those developments require extensive calibration and context-specific tuning, which keeps them largely limited to research labs.

However, a team of researchers in Georgia Tech Associate Professor Aaron Young’s lab developed a universal approach to controlling robotic exoskeletons. It requires no training, no calibration and no adjustments to complicated algorithms. The team says users can just wear the exoskeleton and go.

Young’s lab developed a deep learning AI algorithm to autonomously adjust how the exoskeleton provides assistance. The method used an existing algorithm trained on force and motion-capture data. Subjects representing different genders and body types wore the hip exoskeleton and walked at varying speeds on force plates, climbed height-adjustable stairs, walked up and down ramps and transitioned between those movements.

They demonstrated that this works seamlessly in supporting walking, standing and climbing up stairs or ramps. The team published their work in Science Robotics.

A photo showing Georgia Tech Associate Professor Aaron Young adjusting the experimental exoskeleton worn by one of the researchers.

Georgia Tech Associate Professor Aaron Young adjusts the experimental exoskeleton worn by Dean Molinaro. [Photo by Candler Hobbs/Georgia Tech]

The team’s controller delivered assistance through their own exoskeleton. Through this, they found that they could reduce users’ metabolic and biomechanical effort. Users expended less energy and it alleviated the work done by joints when not wearing the device. The exoskeleton provided benefits even with the extra weight added by the device.

The team also hopes to use their exoskeleton to help stroke patients regain mobility in the future.

“What’s so cool about this [controller] is that it adjusts to each person’s internal dynamics without any tuning or heuristic adjustments, which is a huge difference from a lot of work in the field,” Young said in a post at the unversity’s website. “There’s no subject-specific tuning or changing parameters to make it work.”