A Robotic Leg Learns to Walk

A Robotic Leg Learns to Walk

Researchers announced that they might be the first ones to create an AI-controlled robotic leg that learns to walk on its own.

Thomson Reuters Launches A Start-Up Incubator
Deloitte Insights: The expansion of AI
Leap.ai Provides A AI Service For Job Seekers

Researchers announced that they might be the first ones to create an AI-controlled robotic leg that learns to walk on its own.

The robotic limb is reportedly being driven by animal-like tendons that can even be tripped up and then recover. The most interesting part is that the robot was not programmed to do so. It learned it on its own, as Science Daily reports in an article on the subject.

The researchers, which include Francisco J. Valero-Cuevas, a professor of Biomedical Engineering, have managed to create a bio-inspired algorithm that can learn a new walking task by itself after 5 minutes of unstructured play, and then adapt the tasks it learned without being programmed to do so.

Their research was published in Nature Machine Intelligence and is hoped to open exciting new possibilities for understanding human movement and disability. The technology can be used to create responsive prosthetics, and robots that could interact with complex and changing environments.

“Nowadays, it takes the equivalent of months or years of training for a robot to be ready to interact with the world, but we want to achieve the quick learning and adaptations seen in nature.”, explained the senior author Valero-Cuevas.

The lead author of the paper and a doctoral candidate in the Department of Biomedical Engineering at USC, Marjaninejad, said that the robot was first allowed to understand its environment in a process of free play.

“These random movements of the leg allow the robot to build an internal map of its limb and its interactions with the environment.”, as she said.

Up to now, robots were usually learning by prior or parallel computer simulations. The authors of the study explain that their robots learned by doing and without the necessity of these simulations.

“Because our robots can learn habits, they can learn your habits, and mimic your movement style for the tasks you need in everyday life — even as you learn a new task, or grow stronger or weaker. “, said Valero-Cuevas.

The potential use of the technology may be in the sector of assistive technology, where robotic limbs can respond to the needs of people who have lost the use of their limbs.

The research is available here.

COMMENTS

WORDPRESS: 0