Researchers are developing a mind-controlled robotic arm that does not necessarily need an invasive (possibly life-threatening) brain implant to work efficiently for patients with disabilities or, as they say, for “everyday practical use.”
The robotic arm is a project by a team from the Carnegie Mellon University, in collaboration with the University of Minnesota. The team is working on a robotic arm that could benefit people with paralysis and other disabilities or movement disorders.
Robotic arms have been in development for quite some time now due to the benefits it poses for people who need them the most — people with disabilities. The most effective and most practical way to come about with robotic arms is through brain-computer interface (BCI) technology. In other words, the robotic arms can be controlled by the user’s mind as if it is their regular arm.
“BCIs have been shown to achieve good performance for controlling robotic devices using only the signals sensed from brain implants. When robotic devices can be controlled with high precision, they can be used to complete a variety of daily tasks,” Carnegie Mellon University said in a post.
Currently, there are two significant ways of approaching BCI: one of which involves an invasive procedure, where doctors have to surgically place a brain implant that allows the user’s thoughts or consciousness communicate with the arm. However, this procedure often poses health and side-effects due to the complexity of the procedure. That is already on top of the expensive costs (which only a very few can afford) for the entire operation and the robotic arm.
The other, more desired approach is through less-invasive or, even, non-invasive BCIs. This technology can be attached to the skin to detect the brain’s signals. However, this kind of BIC technology is still far from being considered as effective since based on previous evidence, it does not pick up brain signals as fast as invasive brain implants do. It results in jerky, delayed, or even stops during motion while in use.
Fortunately, researchers from the Carnegie Mellon University and the University of Minnesota have developed the technology to bridge the gap between the benefits of invasive and non-invasive BCI robotic arms.
Bin He, the Trustee Professor and Head of the Biomedical Engineering Department at Carnegie Mellon, sees that non-invasive BCIs is the best approach to achieving efficient and highly precise robotic arms.
“There have been major advances in mind-controlled robotic [arm] using brain implants. It’s excellent science,” He said. “But noninvasive is the ultimate goal. Advances in neural decoding and the practical utility of noninvasive robotic arm control will have major implications on the eventual development of noninvasive neurorobotics.”
The Carnegie Mellon University and the University of Minnesota research team developed a breakthrough, where a system can deal with the lower signal quality that comes from using sensors that are used outside of the body or are applied to the skin instead. They were able to employ a combination of new sensing and machine learning technologies to grab signals from the user that are from deep within the brain and process through “dirty” signals often associated with non-invasive sensors.
In a paper published on Wednesday in the journal Science Robotics, it details how the researchers were able to demonstrate smooth operational functions by letting a set of participants wear EEG head caps that monitors brain activity and a simple test run.
During the testing of the system, they asked the participants to try and control the arm with new sensors and follow a mouse cursor displayed on a computer screen. The robotic arm was able to continuously track the cursor in real-time with no jerky movements — an exciting first for a noninvasive BCI system.
In the past, non-invasive BCI robotic arms used to appear like it was playing catch-up on similar tests all the while, lagging and moving sporadically. Today, researchers announced that it follows a “smooth, continuous path.”
“Despite technical challenges using noninvasive signals, we are fully committed to bringing this safe and [economical] technology to people who can benefit from it,” He said.
“This work represents an important step in noninvasive brain-computer interfaces, a technology that someday may become a pervasive assistive technology aiding everyone, like smartphones.”
Currently, the team behind the project is looking to push with more clinical trials. The technology has been tested on 68 able-bodied human subjects (up to 10 sessions for each subject), including virtual device control and controlling of a robotic arm for continuous pursuit.