For astronauts in space, working in zero gravity and trying to accomplish tasks while wearing a bulky spacesuit is challenging. The lack of gravity slows down a person's motor skills. And it's not easy to operate equipment while wearing gloves and a helmet or other cumbersome gear. Wouldn't it be easier just to control everything with a thought?
A group of researchers led by Riccardo Poli, a computer science professor at the University of Essex, is working on that idea. For the first time, they used a brain-computer interface (BCI) to control a spacecraft simulator -- although, by their own admission, in a highly simplified environment.
Although brain-computer interfaces in space remain theoretical, the scientists discovered that their BCI was far more effective when two people were hooked up to it and had to collaborate on a task in space. That kind of enhanced decision-making ability, while good in space, could be applied to a number of high-stress situations on Earth.
The team set up at NASA's Jet Propulsion Laboratory and began by putting a cap containing 66 electrodes on a human subject. This has the advantage of being a non-invasive way to pick up brain signals, but Poli pointed out that trying to read EEG signals from the scalp is like trying to listen to a concert hall by standing in the street outside the venue. Traffic and noise make it hard to hear.
To help amplify the brain signals, the team used a computer that generated special visual stimuli on a screen. This helped the human subject produce brain signals that could be analyzed a little more easily. Then the scientists made a simulation and presented their subject with a challenge: Steer a spaceship so that it passes within a certain distance from the sun.
The "spaceship" was actually a large circle on a screen and the "sun" was a large white sphere that got bigger as the spaceship was guided closer. A set of eight gray dots arranged in a circle was the cursor for moving the spaceship, with each dot representing a different direction. Then the dots lit up either green or red at random.
In order to steer the spaceship in a particular direction, the human subject had to focus his mind on one of the cursor dots and mentally identify the color of the dot every time it flashed. Concentrating on colors causes the brain to produce stronger brain signals for their system to detect, Poli explained.
When the subject concentrated on moving the spaceship along the right trajectory, several computers worked together to read the brain signals, analyze them, and display the simulated spaceship's movement in real-time.
Initial results were promising. Then, when two people worked collaboratively on moving the spaceship, the trajectory improved considerably. Working with a brain-computer interface is such an intense experience that one person naturally has lapses in attention and two people aren't likely to have a simultaneous lapse, Poli explained.
Joint decision-making might not sound as impressive as controlling a spaceship with your brain, but such a set-up might have more immediate applications in a military setting. For example, intelligence officers tasked with sifting through large volumes of satellite images for anomalies could potentially work more effectively if two of them were hooked up to a brain-computer interface.
Real spaceship control is unlikely to be attempted with such a simple set-up, though. "Certainly a kid with a joystick can do better than any number of adults with a brain-computer interface," Poli said. His research is more of an attempt to move BCI from the lab to the real world and give space agencies a better understanding of BCI in a space context.
This month Poli and his colleagues are presenting their work at the International Conference on Intelligent User Interfaces in Santa Monica, California. Next, Poli said they plan to continue collecting data for their system.
Deniz Erdogmus is an associate professor of electrical and computer engineering at Northeastern University who directs the Cognitive Systems Laboratory there. His team focuses on researching brain interfaces to develop tech that helps people with severe motor and speech disabilities.
"This paper presents yet another exciting exploratory work in the domain of brain-controlled machines," he said. When properly designed and used, such a system might reveal an astronaut's true thoughts. "Brain interfaces could, in the near future, measure attention and vigilance, motivation, fatigue, cognitive load, affective state," he added.
Poli is quick to point out that others have done more groundbreaking work developing brain-computer interfaces. The key for his team was using one to control a spacecraft simulator for the first time.
"What we were trying to do is see where the limits are," he said.