Scientists of the National Research Nuclear University "MEPhI" have developed a method of decomposition of multi-channel control system based on advanced BCI (neurocomputer interface), which is designed to help users of robotic wheelchairs. The results of the study are published in the collection Biologically Inspired Cognitive Architectures Meeting.

The use of a robotic wheelchair with multiple control channels is very important for people with severely limited mobility. Patients are often able only to perform small gestures with their hands, give voice commands and produce "thought images".

Different control channels have their drawbacks: for example, a traditional joystick may be not suitable for people with severely limited motor activity. Therefore, voice control channel, gesture control channel and BCI-based channel are also in demand. The aggregate of all these channels is called extended BCI.

"As a rule, control channels can be used simultaneously, but at the same time give conflicting commands. For example, the voice channel recognized the command "right", and the gesture channel determined the command "left". Therefore, there is a problem of choosing the "right" team in case of a conflict situation," said one of the authors, Associate Professor at the Institute of Intelligent Cybernetic Systems (ICIS) of MEPhI Konstantin Kudryavtsev.

To resolve such conflicts, scientists of ICIS MEPhI have developed a method to decompose the control system. It is based on a statistical assessment of the quality of each of the control channels on the basis of a specially selected quality factor and the assignment of certain channels to certain commands. In case of a conflict, a command from the priority channel is selected.

Next, the scientists intend to experimentally test the robotic chair, improve the methods of assessing the quality methods of channels assessment and minimize erroneous situations in determining the "channel commands".