SBIR Phase I: A Hybrid Brain-Computer Interface for Virtual and Augmented Reality

这个小企业创新研究（SBIR）第一阶段项目的更广泛的影响/商业潜力解决了对非侵入性脑机接口（BCI）和免提技术控制的需求，包括人工和虚拟现实（AR/VR）和智能设备。拟议的多用途BCI预计将立即应用于包括制造业和医药在内的多个行业。目前，现有的系统要么太昂贵，要么限制了实时控制。所提出的BCI是专门为三维环境设计的，并且旨在利用来自人体的多个（“混合”）信号，以允许使用负担得起的硬件来提高性能。它还被设计为并有望允许AR/VR控制，这可以实现生产力应用，以及在现实世界场景中使用模型BCI。长期目标是使用户能够滚动菜单，选择对象，甚至使用他们的大脑活动打字。该平台使用现有的AR/VR头戴式设备的外形尺寸来整合大脑传感电极，并将与流行的设备兼容，独立或与现有的控制器并行。电极设计为安全、无创和干燥（需要导电凝胶或生理盐水）。本项目的高风险、高回报的研究将极大地推进脑机接口系统的应用，重点是AR/VR技术。本项目是通过定制机器学习平台将眼电信号和脑电图（EEG）信号相结合，从而实现一种新型的混合脑机接口。BCI检测和解释神经信号，从而能够控制各种技术。然而，目前的BCI在其适用性方面仍然非常有限。它们要么需要昂贵的设备、侵入性手术，要么在使用经济实惠的非侵入性硬件时性能太低。该BCI旨在提供三维场景中的实时控制（例如，AR/VR/现实世界的智能设备），同时使用负担得起的硬件。该SBIR第一阶段项目旨在将联合收割机三个不同的创新结合起来：高性能EEG信号分析、高速眼动分类和自定义多信号集成分类技术。具体来说，该项目旨在使用一种基于生理学的定制机器学习和人工智能方法，将联合收割机视觉和EEG信号结合起来，专门实现3D AR/VR控制。最终目标是开发一个高性能的BCI系统，提供灵活的用户控制的硬件，软件和移动的应用程序。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project addresses the need for non-invasive brain-computer interfaces (BCIs) and hands-free control of technologies, including artificial and virtual reality (AR/VR) and smart devices. The proposed multi-purposed BCI is expected to have immediate applications for several industries, including manufacturing and medicine. Currently, existing systems are either too expensive or limited for real-time control. The proposed BCI is specifically designed for 3 dimensional environments, and is intended to leverage multiple ('hybrid') signals from the human body to allow increased performance using affordable hardware. It is also designed for and expected to allow AR/VR control, which can enable productivity applications, as well as model BCI use in real-world scenarios. The long term goal is to enable users to scroll menus, select objects, and even type using their brain activity. The platform uses the existing form-factor of AR/VR headsets to incorporate brain-sensing electrodes, and will be compatible with popular devices, independently or in parallel with their existing controllers. The electrodes are designed to be safe, non-invasive, and dry (requiring conductive gel or saline). The high-risk, high-reward research to be conducted under this project will significantly advance the applications of BCI systems in general, with an emphasis on AR/VR technologies.The proposed project concerns a novel hybrid BCI by combining oculomotor and electroencephalography (EEG) signals via a custom machine learning platform. BCIs detect and interpret neural signals enabling control over a variety of technologies. However, current BCIs remain extremely limited in their applicability. They either require expensive equipment, invasive surgery, or have too low performance when using affordable noninvasive hardware. This BCI aims to provide real-time control in 3-dimensional scenarios, (e.g., AR/VR/real-world smart devices), while using affordable hardware. This SBIR Phase I project seeks to combine three distinct innovations: high-performance EEG signal analysis, high-speed eye movement classification, and custom multi-signal ensemble classification techniques. Specifically, the project seeks to use a custom machine learning and artificial intelligence approach informed by physiology to combine oculomotor and EEG signals to specifically enable3D AR/VR control. The ultimate goal is to develop a high performance BCI system that affords flexible user control across hardware, software, and mobile applications.