Computer Vision for Robotic Welfare Technology

The use of robotics in welfare technology is increasingly more common as these devices becoming both more capable and readily available. For instance; robotic arms or exoskeletons which could help an otherwise paralyzed person move. The potential for such assistive robotic devices is especially big for quadriplegics, i.e. persons with all four limbs paralyzed.

The objective of this project is to apply computer vision to make it easier for a paralyzed person to control an upper-body exoskeleton through a tongue-based interface. Computer vision would enable the system to become aware of its immediate environment in order to behave in more intelligent way. For instance; detecting nearby objects and how to interact with them. Another focal point of the project is the HRI (human robot interaction) as the system needs to arbitrate between input from the computer vision and input from the user in a sensible way.