Researchers from Carnegie Mellon University and the University of Minnesota have used the non-invasive Brain-Computer Interface (BCI) technology to successfully develop the first brain-controlled robotic arm with the ability to continuously track computer cursors. The results were published on Science Robotics.
Relying on brain consciousness to achieve non-invasive control of robotic devices will have enormous potential for application, especially to change the lives of patients with paralysis and dyskinesia.
Non-invasive brain-computer interface is the ultimate goal
In sci-fi works, the realization of mind control can be easily and easily carried out. In the real world, the supporting technology behind it - brain-computer interface technology has been nearly a hundred years old.
Scientists have long hoped to achieve breakthroughs in scientific research and application technology in the field of brain-computer interface, providing better exploration tools for many problems that are still unanswerable, helping humans to further understand their brains, preventing, diagnosing and treating brains. Department of disease, and this technology is widely used in the fields of sleep management, intelligent life and rehabilitation of disabled people.
Today, brain-computer interface technology has proven to be a good use of brain implant signals to control robotic devices. When robotic devices can be controlled with high precision, they can be used to perform a variety of everyday tasks.
However, to date, those brain-computer interfaces have successfully controlled the study of robotic arms, using invasive brain implants. These implants require extensive medical and surgical expertise to properly install and operate, not to mention the cost and potential risks to the subject. Therefore, their use is limited to a small number of clinical cases.
On the other hand, there are a number of problems with non-invasive external sensing, such as the "more complex" signals received by brain-computer interface devices, resulting in lower signal resolution and less precise control. The development of invasive or even non-invasive brain-computer interface technology, enabling paralyzed patients to use their "thoughts" to control robotic limbs is a major challenge in the field of brain-computer interfaces. If this non-invasive brain-computer interface technology is successful, it will bring huge changes to many patients and even ordinary people.
He Bin, director of the Department of Biomedical Engineering at Carnegie Mellon University, said that brain-controlled robotic devices using brain implants have made significant progress, and this is an excellent science. But non-invasiveness is the ultimate goal. The advancement of neural decoding and the practicality of non-invasive robotic arm control will have a major impact on the ultimate development of non-invasive neural robots.
Non-invasive brain control robot arm
In the latest paper, He Bin and colleagues made breakthroughs in the development of low-invasive or non-invasive brain-computer interface technology. They use novel sensing and machine learning techniques to overcome noisy EEG signals through non-invasive neuroimaging and a new continuous tracking paradigm, significantly improving EEG-based neural decoding for real-time continuous robots equipment control.
Using non-invasive brain-computer interface technology, the researchers successfully controlled the robot arm for continuous tracking of the cursor on the computer screen for the first time in human subjects. Prior to this, robots that were non-invasively controlled by humans would follow the moving cursor to make unstable, discontinuous motions, and now the arm follows the cursor along a smooth, continuous path.
The research team has also established a new framework to address and improve the "brain" and "computer" components of the brain-computer interface by increasing user participation and training and imaging the spatial resolution of non-invasive neural data from EEG sources. The results of the study show that the team's unique approach to solving this problem not only increases BCI learning by nearly 60%, but also increases the continuous tracking capability of computer cursors by more than 500%.
The technology can also help a wide variety of people by providing secure, non-intrusive "consciousness control" devices that allow people to interact with the environment and control their environment.
To date, the technology has been tested in 68 physically fit human subjects (up to 10 tests per subject), including virtual device control and robotic control for continuous tracking. Next, the research team also plans to conduct relevant clinical trials on patients.
“Despite the technical challenges of using non-intrusive signals, we are fully committed to bringing this safe and economical technology to those who can benefit from it,†He Bin said. “This work represents an important step in the non-invasive brain-computer interface, and this technology may one day become a universal aid to everyone.â€
reference:
Https://engineering.cmu.edu/news-events/news/2019/06/20-he-sci-robotics.html?
Source: DeepTech Deep Technology
Gas Chromatography,Lab Method For Gas,Capillary Column Gas Chromatograph,New Gas Chromatograph Detectors
Guangdong Widinlsa International Co.Ltd , https://www.gzwidinlsa.com