Engineers at University of Minnesota in Minneapolis developed a head cap with electrodes that captures and interprets thoughts to direct autonomous devices. The team led by biomedical engineering professor Bin He demonstrated the system with a remote-controlled flying device and published its findings yesterday in the Journal of Neural Engineering.
The cap worn by the subjects resembles a swim cap that fits snugly over the skull, with 64 electrodes to capture brain signals with electroencephalography (EEG), a technique to diagnose disorders such as epilepsy or brain injuries. The cap captures signals from the motor cortex that controls voluntary movements, which in this case, represent thoughts about movements triggering a different set of neurons in the brain. Unlike other systems that capture brain signals, the cap is non-invasive; no chips or electrodes or implanted in the wearer.
In earlier research, Prof. He was able to map the areas of the brain activated when a person imagines physical movements. Prof. He followed up that earlier work, with experiments showing the system could control the movements of a virtual helicopter on a computer screen.
In the new study, Prof. He and colleagues applied the system to controlling the movements of a model helicopter, an AR Drone autonomous flying device with four rotors, made by the company Parrot SA. The AR Drone was preset to fly forward, unless instructed to fly in another diection by the operator.
The EEG system captures thoughts by the individual imagining use of the right hand, left hand, or both hands together. Signals from these thoughts were translated into instructions to turn the AR Drone right or left, or in vertical movements up or down, and were transferred to the model helicopter over a local WiFi network.
In the study, the researchers tested the ability of five test subjects wearing the EEG system to control the AR Drone device, including navigating the model helicopter through two large balloon rings suspended from the ceiling of a gymnasium. The test subjects faced away from the flying model, but could view its movements over a computer screen.
The researchers measured the performance of the test subjects against another group using a standard keyboard to control the AR Drone. The results showed the test subjects were able to perform from 69 to 92 percent of the required tasks, without crashing the AR Drone into the gymnasium walls. Using a higher performance standard — directing the AR Drone through the ring targets without colliding with the rings — the test subjects succeeded between 54 to 85 percent of the time.
Prof. He believes the findings can be applied to help people with paralysis or neurodegenerative diseases, for controlling wheelchairs or prosthetic devices, as well as victims of stroke. “We’re now studying some stroke patients,” adds Prof. He, “to see if it’ll help rewire brain circuits to bypass damaged areas.”
The following video tells more about the EEG system and shows some of its use with the AR Drone.
- FDA Approves Robotic Device for Stroke Rehabilitation
- Tiny, Winged Manueverable Robotic Insects Created, Tested
- Robotic Bat Wing Developed to Simulate Flight Dynamics
- Miniature Robots Being Developed to Work in Swarms
- Algorithm Improves Brain-Controlled Cursor Movements
* * *
You must be logged in to post a comment.