基于异步脑机接口系统的远程低时延控制研究

Research on Remote Low latency Control Based on Asynchronous Brain computer Interface System

针对脑机接口目前应用在控制领域存在着自主性较低、实时性较差等问题,展开了基于异步脑机接口系统的远程低时延控制的研究。异步脑机接口系统采用滤波器组典型相关分析算法实现特征识别,并通过检测连续滑动窗口阈值实现状态识别;在此基础上,搭建两级C/S架构解耦原始系统,通过分析传统控制流程存在的时间和性能损耗,设计了基于半同步半异步线程池的新型控制系统架构。通过自研移动采样机器人进行了实验验证,实验结果表明,该系统具备可远程、低时延、高稳定性和高识别率的特点。

In view of the problems of low autonomy and poor real time performance in the current application of brain computer interface in the control field,research on remote low latency control based on asynchronous brain computer interface systems is carried out.The asynchronous brain computer interface system uses the filter bank typical correlation analysis algorithm to realize feature recognition,and realizes state recognition by detecting continuous sliding window thresholds.On this basis,a two level C/S architecture is built to decouple the original system.By analyzing the time and performance losses of traditional control processes,a new control system architecture based on semi synchronous and semi asynchronous thread pools is designed.Experimental verification is carried out through a self developed mobile sampling robot.The experimental results show that the system has the characteristics of remote control,low delay,high stability and high recognition rate.