Vision-based system for grasp assistive devices, that can be controlled through the human gaze.
We humans naturally grip on objects before moving their arms to clutch the object. Amyotrophic lateral sclerosis (ALS) which is a progressive nervous system disease that affects nerve cells in the brain and spinal cord, causing loss of muscle control. Patients with this ALS disorder can select an object with their integral eye movement, but due to lack of voluntary muscle control cannot move their limb.
The 2 important parameters for controlling the wrist angle and the grasp space of the assistive device to replicate a human-like stable hold are object orientation and object width. To control the grip with a prosthetic hand, vision systems have been evolved that can measure the geometrical characteristics of the object. Many of the now existing vision systems are integrated with electromyography and to some extent requires a slight amount of voluntary muscle movement to control the vision system. As a result, these systems cannot be used or are not beneficial for those using assistive devices controlled using the brain.
Now the team came up with this new concept of a vision system which can be controlled through the human gaze.
Keeping the object of interest in focus and at the center of the picture, the team measured the vertical and horizontal electro-oculogram signals and controlled the pan and tilt of the cap-mounted webcam. To reduce the Algorithmic complexity and system storage capacity, a simple ‘signature’ extraction procedure was also utilized.
The team worked on the object’s size, fixed approximate wrist orientation angle and the grasp space size within 22ms. The combined accuracy surpassed 75%. The increasing number of grasps is expected to offer more natural guidance for ALS patients with integration of the proposed system with the brain-controlled grasp assistive device.
Though several research works have already achieved success in generating the correct grasp type from their brain measurement, the team was searching for fine control over an object with a grasp’s assistive device (orthosis/exoskeleton/robotic arm).
The new invention of Electro-oculogram Based vision system can provide the above-mentioned assistive devices with a new outlook and aspiration to the users. So far, the team with the new technology has been tested on 10 Healthy participants with better results.
Keep browsing iSkulltech to get more interesting news from the tech world. You can follow us on Facebook and Instagram to get the notifications instantly.