Over the past few years, handful of labs have demonstrated remote control of animals, typically insects. However, these techniques are usually invasive and require operations, such as implanting electrodes in the animal’s brain or nervous system.
Led by Phill-Seung Lee of the Mechanical Engineering Department and Sungho Jo of the Computing Schoo, the KAIST concept aims to guide the animal subject by tapping into its instincts, including its fight-or-flight response.
Turtles make for ideal subjects due to their relatively high cognitive capacity, ability to differentiate between varying wavelengths of light, and instinctual escape behaviors. For example, turtles predictably move toward white light and away from obstacles in their environment.
The turtle wears a “cyborg system” on its shell that consists of a camera, Wi-Fi transceiver, Raspberry Pi control module, battery, and a blinder that swings from left to right to obstruct the turtle’s view. The human operator wears a brain-computer interface system that picks up signals from the brain to move the blinder accordingly.
By watching a feed from the turtle’s camera, the human operator can guide the turtle simply by thinking about moving. The system picks up on the three mental states of left, right, and idle movement. If the human operator thinks “left,” the blinder will swing to the turtles right side, causing the turtle to move away from the obstruction.
Though the immediate applications of a mind-controlled turtle seem a little far-fetched, the researchers think their technology could be used to inform positioning and augmented reality systems, offering support for rescue, reconnaissance, and surveillance missions.
