Smart Robotic Exoskeletons

The time when exoskeletons and prosthetic legs are able to think and control movement on their own using artificial intelligence (AI) is here.

The ExoNet project at the University of Waterloo combines computer vision and deep learning AI to mimic how able-bodied people walk by seeing their surroundings and adjusting their movements.

The research team is giving robotic exoskeletons a sense of vision to control themselves. While exoskeletons legs operated by motors already exist, users of the exoskeletons currently control them via smartphone applications or joysticks. However, the overall inconvenience of taking out a smartphone has prompted the team to go further with their innovation.

The exoskeleton users are now fitted with wearable cameras and are optimising AI computer software to process the video feed so that surroundings such as stairs, doors and other objects can be detected.

Future work on the system will see the robotic exoskeletons climb stairs, avoid obstacles and take actions based on the user’s movements and the upcoming terrain. The control approach would not necessarily require human thought, much like an autonomous car.

Next steps in the project are to improve the energy efficiency of motors for robotic exoskeletons and prostheses by using human motion to self-charge the batteries. Read more here.