Brain-Controlled Robotic Arm Grasping System Based on Adaptive TRCA

摘要

The SSVEP-BCI system used in actual scenarios usually uses fixed time window length decoding to analyze EEG signals, which cannot achieve decoding window length adaptation, resulting in low overall system operating efficiency. To solve this problem, this article uses an adaptive TRCA-based SSVEP EEG decoding algorithm. It can establish an individualized spatial filter based on the collected training data to maximize the extraction of effective EEG signals, at the same time, Bayesian estimation is used to dynamically find the optimal data length. In this paper, a brain-controlled robotic arm grasping system based on adaptive TRCA is constructed. Ten subjects were recruited for system testing, and 400 online tests were performed on the system using fixed window and adaptive methods. The test results show that the system is stable and reliable, and the accuracy of decoding EEG signals can reach 95% within 1.5 seconds, which can be used to assist people with disabilities in their daily lives.