A method of brain computer cooperative navigation combined with simultaneous localization and mapping

一种结合SLAM的脑机协同导航方法

¹ School of Electronic and Information, Northwestern Polytechnical University, Xi’an 710072, China
² Institute of Medical Research, Northwestern Polytechnical University, Xi’an 710072, China

Received: 18 June 2021

Abstract

Introducing human brain intelligence into robot system is an effective means to improve robot's cognition and decision-making ability. Aiming at the problems of human brain fatigue and the need of multi lead information in brain robot control, a brain computer cooperative navigation method combining synchronous localization and mapping (SLAM) is proposed in this paper. Through the steady-state visual evoked potential based on three leads, the image of the target area of interest of human brain is selected, and the brain computer cooperative navigation task is completed by combining SLAM and artificial potential field. The test results show that the average accuracy of the target area image selection method based on steady-state visual evoked potential is 94.17%, which proves that the three leads are effective. On this basis, the brain computer cooperative navigation method combined with SLAM is tested. The results show that the completion rate of navigation task is as high as 92.5%. This method alleviates the fatigue of human brain and reduces the hardware requirements of EEG acquisition.

摘要

在机器人系统中引入人脑智能，是提高机器人认知、决策等能力的有效手段。针对脑-机器人控制存在着人脑疲劳、需要多个导联的信息等问题，提出了一种结合同步定位与地图构建(simultaneous localization and mapping，SLAM)的脑机协同导航方法。通过基于3个导联的稳态视觉诱发电位，实现人脑感兴趣目标区域图像的选取，并结合SLAM和人工势场方法，完成脑机协同导航任务。测试结果表明，基于稳态视觉诱发电位的目标区域图像选取方法，平均正确率为94.17%，证明3个导联选取目标区域图像是有效的。在此基础上，测试结合SLAM的脑机协同导航方法，结果表明导航任务完成率为92.5%。所提方法缓解人脑疲劳的同时，降低了脑电采集的硬件要求。