Brain-machine interfaces (also known as brain-computer interfaces) have been around for some time.

In particular, scientists have made some impressive strides in developing invasive BMIs, which record brain activity from an implant inside the brain. For example, in 2012, a paralyzed woman used a BMI developed by the BrainGate consortium to pick up a cup with her mind. But most of these efforts involved controlling computers or robotics limbs.

Meanwhile, other researchers have shown that wheelchairs can be controlled using BMIs that sit on a person's scalp and measure "EEG" brain waves through the skull. But these signals only allow for crude control which is often slow and easily interrupted.

"We wanted to have something that is closer to the way we move for a human who is completely paralyzed, because you can control each movement continuously," study leader and Duke University neuroscientist Miguel Nicolelis told Business Insider.

Training monkeys to drive wheelchairs

In the new study, Nicolelis and his colleagues developed a wireless BMI that they implanted in the brains of two monkeys, in areas responsible for planning and controlling limb movements.

First, they trained the animals to control a wheelchair by first having them passively driven around, while their brain activity was recorded. The recorded signals were used to train an algorithm that converted them into instructions that could steer a wheelchair.

Shawn Rocco/ Duke HealthHere's what a monkey's brain signals look like.

Then, they switched the wheelchair to brain control, and allowed each monkey to drive itself toward a bowl of fruit. The food was always in the same position, but the animal's starting position was changed randomly.

The animals were able to drive the wheelchair within the first day of training, and they got better after training for several days, Nicolelis said.

The advantage of a wireless system, of course, is that it doesn't need to be plugged in to a computer. However, the batteries only last for about two hours on a single charge, according to Nicolelis.

Other researchers are also developing wireless BMIs and wheelchairs controlled by limb movements, but this is the first attempt to combine these to achieve brain control of whole-body movements.