A team of researchers from the University of California in San Francisco, has developed a system that combines artificial intelligence and robots, which allowed a man with quadruple paralysis to control an automatic arm using their ideas only.
The man was able to perform complex tasks such as the holding and release of things, the opening of a wardrobe and the release of a cup and placing it under the drinking distributor, without the need for fundamental adjustments to the system for a period of 7 months, which is an unprecedented development in this area.
The system is based on advanced artificial intelligence algorithms to match brain signals with the corresponding movements, as part of the BCI (BCI) interface. This approach allows a man to see robotic arm movements in real time while imagining them, allowing errors to correct quickly and reach higher precision in the control of movements.
“This integration between human learning and artificial intelligence represents a new step in the development of brain and computer facades, which we must reach more advanced motor functions that imitate natural capacities,” explains neurologist Carronish Gangoli, of the University of California.
The study revealed that the brain activity models linked to the movement remained fixed, despite a slight change in its positions over time, which is the result of the nervous learning process. Thanks to artificial intelligence, the system was able to keep the pace of this slight change in the brain signal site without having to reset it several times. This means that the user continued to control the robotic arm with precision and regularly for a long time, without the need for technical intervention to restore calibration.
The researchers stressed that the nervous side was entirely subject to a voluntary control, without any automatic assistance. Visual aid can contribute to improving performance, especially to manage complex things.
Despite the complexity of this system and its high cost, due to its dependence on brain implants and electric planning technology (ECOG) to read nervous activity, its results prove the possibility of identifying the nervous models associated with motor ideas and following them even if their positions in brain change.
This innovation comes in the context of accelerated progress in brain and computer facades techniques, where similar systems have been developed who have helped people who have lost the ability to speak to restore their voices and have enabled a quadruple paralysis to practice activities such as chess.
Despite the need for subsequent development, the continuation of technological progress will allow the implementation of more complicated tasks in the future. Healthy alart
