- Human Machine Interfaces have allowed the development of two robots: a cap that controls a wheelchair and glasses that recognize eye movement and allow a patient with quadriplegia to communicate.
- The international journal Sensors recognized the researchers’ work by publishing an article.
- In Mexico alone, it is estimated that there are 6 million people with a movement disability.
With the purpose of offering solutions to improve the quality of life of patients with quadriplegia or some type of motor paralysis, researchers from the Universidad del Valle de México, developed robots in the form of Human Machine Interfaces. These are assistance and rehabilitation systems through physiological signals. Thanks to this, it has been possible to manufacture a cap that controls a wheelchair and lenses that, through eye movements, allow people to communicate.
Innovation made in Mexico
Dr. Francisco David Pérez Reynoso, researcher at the Center for Research, Innovation and Technological Development (CIIDETEC-UVM) and the leader of this development, explained how it works. Through the training of artificial intelligence sensors fabrication of these devices is possible. They are non-invasive and are customized according to a patient’s need for muscle, eye or mechanical movements.
He pointed out that one of the main problems of the advances that exist is the degree of customization to apply rehabilitation therapy. Also to adapt an assistance system to the individual characteristics of the patients. Hence the importance of this work.
On the other hand, he recalled that according to the World Health Organization (WHO), in 2020 alone there were more than one billion people with some form of disability globally. Of these, about 190 million have operating difficulties that required support services.
While in Mexico, the 2020 Population and Housing Census revealed that there are more than six million people with some type of disability. Of this figure, 48% have mobility problems to walk, go up or down.
The role of Artificial Intelligence
The researcher indicated that to get a machine or device to adapt to the characteristics or needs of a person, they use artificial intelligence. In other words, the HMI is trained in such a way that regardless of the signal or sensor that is placed on the interface, it can interpret it as a control command or movement command. In this way the machine interprets the movements made by the user and not the user has to adapt to the machine.
This allows a significant time reduction. Now it takes a couple of weeks instead of months to adapt the assistive robot.
An example of this is the manufacture of a Wheelchair, which consisted of a controlled car with head movements and a neural network, to later implement it in the wheelchair. The neural network is trained in such a way that it understands the commands in order to be in control.
Another of the works carried out was the development of some assistive lenses. Through muscular signals through the eye, the patient could write text, for example, water or could communicate through symbols or colors. This device was made for a young man who suffered a motorcycle accident and consequently became quadriplegic, he only had movement over his eyes.
Dr. Francisco David Pérez Reynoso highlighted that, regarding this research work, his article was recently published Recognition of EMG signal patterns by machine learning for the control of a manipulator robotby international scientific and technological magazine Sensors.
In said article, the researcher confirmed that the design of the HMI made it possible to classify muscle signals according to the time of contraction. The model implemented through a neural network achieved the personalization and classification in real time for the generation of movement commands of a virtual robot.