As I sit down in Nissan’s simulator, I prepare myself for the fact that a cohort of researchers could scrutinize my skills as a wheelman with more rigor than the most aggravating backseat driver. And, I accept that this process involves wearing what looks like a too-small, sideways bicycle helmet, which holds 11 electrodes poking through my hair.
“For each corner, there’ll be an evaluation of your driving smoothness,” says Lucian Gheorghe, the Nissan researcher in charge of this rig. The electrodes tickling my scalp measure EEG activity, those very faint electrical signals in my (and your) brain. Gheorghe is interested in motor related potentials, a specific pattern of activity the brain creates as it prepares to move a limb.
It takes half a second for the body to translate that signal to the wave of an arm or kick of a leg, and Nissan wants to exploit the gap. If a computer reading the signals knows I’m going to turn hard right, before I do it, it can help out by turning the wheels earlier. At high speeds, every fraction of a second counts. Those 500 milliseconds of advance notice is enough for active safety systems to kick in, and hopefully make driving smoother and safer.
The driving simulator, which Nissan demonstrated at CES in Las Vegas last week, includes a seat and basic steering wheel, but no pedals (the sim controls speed), in front of three wide screens giving an almost wrap-around view of a Norwegian mountain road. I don’t have much chance to take in the fake scenery before the “car” gets going, and it takes me a moment to adjust to the sensitiveness of the wheel. An oncoming car looms as I drift toward the wrong side of the road. “Don’t go over there!” says Gheorghe. I jerk back. The computer monitors it all, and compares my actions to my brainwaves, learning how I drive. I can feel it judging me.
On a big screen above me, Nissan has rigged up a graphic display showing some of the inner workings of what it calls the Brain-to-Vehicle, or B2V, system, with colorful pictures of gray matter. A red patch flashes right on top of my brain, in the motor cortex, when I plan to move. “You have control of feet there, and your hands right next to it, so we always look at this area,” says Gheorghe.
It’s a little creepy that a computer knows what I’m thinking, but at least I’m comfortable. The system is not invasive, and Nissan has worked to make the helmet easy to wear. Older systems looked like a shower cap with 64 electrodes and squirts of conductive gel. With this, there are just the few pins touching my head, held in place with gentle elastic cords. The headgear is wireless, with a bluetooth transmitter on one side, and a battery on the other. Comfy, but hardly cool.
Despite the production-ready looking headgear, this is very much a research project, and Nissan’s still figuring out all the use cases. Its researchers have proven the basic tech of capturing brainwaves, but how helpful are they in real-world driving? Yes, the driver is about to move, but what is he going to do? Brake? Swerve? Or just turn up the volume?
That’s where “smart” autonomous driving systems come in. They can use sensors and built-in maps to assess what’s happening outside the car, and make educated guesses like “If this is person is about to steer, has to be a steer left—going right would mean hitting that car or jumping the curb.” Then the computer can start to steer in that direction before the driver even twitches. If that steer left is a swerve to avoid an obstacle in the road, starting it a little sooner gives you a little better chance of missing it. Gheorghe says he can differentiate between movement signals to control the radio and the larger movements related to driving.
Any system that tries to predict what a driver wants will have to make that interpretation perfectly to be safe, let alone useful. “Think autocorrect, or Google-auto-complete, but at 60 mph,” says Anuj Pradhan, who studies human-car interaction at the University of Michigan Transportation Research Institute. “If the two systems, human and vehicle, are not on the same wavelength, then there could be severe consequences.”
Nissan isn’t the only company that thinks it can improve transportation by poking at our brains. In 2016, I flew a small plane using just my mind (an no experience of any kind). Honeywell Aerospace, which created the system, says eventually it could help pilots with non-critical tasks like flicking through checklists by thought alone, leaving their hands free for more important controls.
At Stanford, researchers are measuring brain signals and other physiological measures like heart rate and pupil size to see how much thinking effort, or cognitive load, driving tasks take, and when human drivers really have to concentrate. That could lead to smart systems that inhibit notifications like text messages at critical times (sudden traffic, bad weather, busy intersections), but will let them through there’s not much going on.
And semi-autonomous cars, like Autopilot-equipped Teslas, could monitor the mental state of the human driver, and know if it’s OK to ask them to take control, or if they’re so wrapped up in Instagram on their phones that they’ll need a few seconds to figure out what’s happening. “That could be the real power use case for this system—something that currently is being done only via eye or head tracking,” says Pradhan.
Nissan is already testing its setup on the road. “I drive everyday wearing this from work to home, and home to work,” says Gheorghe. So do dozens of his team members in Japan and Switzerland, recording their daily drives, gathering data to analyze later, and refining the readings and predictions. “I’m digitizing my driving life,” Gheorghe says.
Navigating the winding mountain road, I try to hit each corner smoothly. I couldn’t feel any computer interference, and, Gheorghe says, I never will. The input should be imperceptible—as long as the car and the human do the same thing. Along with helping me drive, it could make the car feel sportier, because it is more responsive, more quickly. And if I’ll let a convertible mess up my hair for a more thrilling drive, why not a goofy helmet?