Using a mathematical algorithm, bioengineers from the University of Illinois in Chicago have come up with an AI that can estimate a person’s intents by analyzing previous actions. In an experiment, research subjects reached for an object on a virtual desk, but their hand was pushed in the opposite direction. These actions were tracked and analyzed by the researchers, leading to the development of an advanced algorithm that predicted the movement intended by the subject’s hand. The algorithm could still determine the way the person wanted to move despite the interruption in the action. "We call it a psychic robot,” says Justin Horowitz, first author of the study published in the journal PLOS ONE. “If you know how someone is moving and what the disturbance is, you can tell the underlying intent."
Real life applications
The algorithm developed by the researchers could have many useful real life applications. One area is the advancement in "smart" prosthetics where stroke patients suffering from tremors or muscle spasms could be aided by the algorithm to calculate their intents and help them complete their action smoothly even with the physical shaking. The algorithm could also be used to make semi-autonomous vehicles safer such as when a car goes out of control, the algorithm could correct its course by analyzing the driver’s previous actions. Horowitz explains: "The computer has extra sensors and processes information so much faster than I can react. If the car can tell where I mean to go, it can drive itself there. But it has to know which movements of the wheel represent my intention, and which are responses to an environment that’s already changed."