Stimulus-free brainwave-initiated and learning-based brain robotics

Brain-computer interfaces (BCIs) using signals acquired with intracortical implants have achieved successful robot control useful for completing daily tasks.

However, non-invasive BCI-controlled robotics is challenging due to constraints associated with external stimuli, inherent complexity of neural signals, and difficulties in online neural decoding.

The scientific problem of this project is how to use semantics-based stimulus-free brain signals to achieve efficient brain-robot control and interactions, and I will develop a new interdisciplinary approach to achieve bidirectional brain-robot systems.

The scientific novelty lies in 1) a new semantics-based stimulus-free approach to read task-relevant brain signals with real-time neural decoders, 2) precise interpretation of brain signals as text-based control commands using a graph-based learning algorithm, and 3) an advanced control scheme to a bidirectional brain-robot system.

Dividing into three work packages in a grant period of three years, I will conduct the project at three top-ranking departments under the supervision of three highly renowned professors with closely relevant research interests to the project.

The theoretical contributions of the project will offer technically feasible solutions for brain-robot systems while the new technological tools developed will enable task-oriented and realistic applications in the real world. The results will notably advance the frontiers of Swedish research on brain robotics.