SBIR-STTR Award

It is hypothesized that the BioLink, a brain actuated control device produced by Brain Actuated Technologies, Inc can be used as a controller to enhance the human/machine interface. Two technical objectives of this proposed effort will be to determine if there are ways to improve the accuracy and responsiveness of the BioLink interface through judicious use of selected control frequencies and through implementation of adaptive calibration schemes. Rule-based and Neural-network calibration schemes will be evaluated. A third objective will be to evaluate the effectiveness of various modes of BioLink control. The fourth objective will be to quantify the levels of accuracy and responsiveness achievable with the BioLink. The proposed study will be performed in three parts. Part I will be designed to evaluate subjects' abilities to control the BioLink at a number of selected control frequencies and to evaluate two different modalities of control at the selected control frequencies. Part II of the study will involve developing rule based and neural network based calibration systems which optimize the subjects' BioLink system parameters such as amplifier gain, window centering and thresholding. The effectiveness of the optimization schemes will be evaluated in Part III.

A Phase I study was undertaken to test the feasibility of using forehead biopotentials for brain actuated computer control. Findings indicate that operators are able to use their forehead biopotentials for both continuous and intermittent computer control. Operators reported an immediate sense of control, performed successfully within minutes and improved with training. The measurement site was easy to access and reliable signals were always obtained. In other studies operators were able to demonstrate simultaneous independent control of at least two control signals derived from the single forehead signal. Continuous and intermittent control with under 0.2 second response times were measured. The Phase I findings indicate that the approach is feasible. Therefore the following Phase II effort, designed to transition the technology to a preproduction prototype, is proposed. Physiological modelling, based upon literature review and experiments, will be performed. Results will identify forehead biopotential sources and will be used for task sensitive controller design. A dry electrode system will be developed. A low noise bioamplifier/microcontroller signal acquisition system will be developed. An operator controller interface, using a system theoretic/task sensitive approach drawing upon human operator manual control and decision making results, will be developed. An automatic interface adjustment scheme, sensitive to individual operator characteristics, will be implemented.