Exoskeleton with eyes and AI – scientists are developing special robotic legs for disabled people (Video)

A University of Waterloo research project called ExoNet is developing a special exoskeleton, which is capable of thinking and making decisions based on its environment. This kind of technology one day could help paralyzed people walk again.

Exoskeletons could one day allow paralyzed people to walk independently again, but they need to be smart. Image credit: University of Waterloo

You are constantly adjusting your walking parameters based on the feedback you’re getting from your environment. You walk differently on a soft surface, you prepare yourself before using stairs. Meanwhile robots cannot really do that, especially exoskeletons. These robotic legs could help disabled people walk again on their own, but how could they prepare to stop, climb stairs, make a sharp turn? Scientists believe that in the future exoskeletons are going to be smart thanks to cameras and artificial intelligence.

Currently exoskeletons need to be controlled manually via smartphone applications or joysticks. This is less than ideal, because the disabled person can’t walk as intuitively as an able-bodied person can. And his or her hands are always occupied with these controls. That kind of a cognitive load is extremely tiring and can be dangerous over time. Could you imagine needing to take out your phone every time you want to climb a set of stairs or walk through a strip of sand? Scientists want to borrow a page from a book about autonomous cars and therefore are optimizing AI computer software to process the video feed to accurately recognize stairs, doors and other features of the surrounding environment.

Brokoslaw Laschowski, leader of the ExoNet research project, said: “Our control approach wouldn’t necessarily require human thought. Similar to autonomous cars that drive themselves, we’re designing autonomous exoskeletons that walk for themselves.”

Simply put, these new robotic legs would have several cameras, which would provide information to the computer vision system. A deep-learning AI algorithm would then try to mimic how able-bodied people walk by seeing their surroundings and adjusting their movements. System would be able to switch between different modes of locomotion to ensure optimal setting without the user needing to take manual controls. For example, exoskeleton would stop and prepare itself to climb the stairs. Users would have to wear wearable cameras, but the  optimized system would be more convenient to use.

We take a lot of things for granted. Such as our own walking – do you ever notice how many little adjustments you make without even thinking about it? Like when your angle rolls a bit and you instinctively compensate to avoid a fall? These kind of abilities are difficult to program. But maybe someday exoskeletons will provide that freedom to disabled people.


Notify of
Inline Feedbacks
View all comments
Would love your thoughts, please comment.x