Brain Computer Interface(BCI) System Research Of Upper Limb Rehabilitation Training Robot

Brain-computer interface(BCI)can convert electrical activity signals of cerebral cortex into computer or other machine language to directly control external devices,independent of peripheral nervous system and muscle tissue.Patients with severe motor disability(stroke,spinal cord injury,etc.)can communicate with the outside world through BCI,so that these patients with dyskinesia and normal brain function can independently control the external equipment and improve the quality of life.Traditional BCI usually only uses a single control signal,which is difficult to meet the actual operation requirements.The results show that the performance of hybrid BCI is better than that of traditional BCI,such as recognition accuracy,information transmission rate and so on.Nowadays,mixed BCI,which combines different types of EEG,has become one of the main directions of brain development.Although brain-computer interface has achieved fruitful results in recent years,there are still a series of problems in the application of rehabilitation robot based on brain-computer interface,such as long recognition time,difficulty for users to control the rhythm of rehabilitation training independently and low recognition rate of many kinds.Therefore,this paper presents a set of upper limb rehabilitation training robot system based on Steady State Visual Evoked Potential(SSVEP)and Alpha Wave Mixed BCI.The following three aspects are done:1)Alpha wave and SSVEP signal are combined to realize the asynchronous control of the system.Both SSVEP and Alpha belong to the signal with obvious rhythm and high signal-to-noise ratio.The feasibility of controlling the upper limb rehabilitation training robot is realized by combining two different types of signal characteristics.This system uses the blocking phenomenon of Alpha wave to transform idle/working state;through the analysis of SSVEP signal,the control command of upper limb rehabilitation training robot is obtained,and the multi-thread concurrent upper computer software is designed to ensure the real-time performance of the system and improve the performance of hybrid BCI.2)The analysis and processing of SSVEP and Alpha wave signal is the most important link for the system to successfully control the upper limb rehabilitation robot.In this paper,we first use independent component analysis(ICA)to preprocess EEG signals,and then extract and classify EEG features.The frequency identification methods of SSVEP signal are mainly introduced,which are canonical correlation analysis(CCA),multivariable synchronization index(MSI)and total task correlation analysis(TRCA).Off-line experiments are designed based on two performance indicators of data length and channel number.The performance characteristics of the three algorithms are compared,and the most suitable algorithm is found to be applied to the system to improve the stability of the system.3)On-line experimental verification of an asynchronous brain-computer interface upper limb rehabilitation training robot.In the end of this paper,we design an on-line experiment to prove that the system has a high accuracy(accuracy > 93%)in the analysis of EEG signals of users,and it takes a short time.Subjects can control the upper limb rehabilitation training robot system based on hybrid brain-computer interface according to their own rhythm,and control the training action and training time according to their own rhythm,which provides a good theoretical and practical basis for the development and application of asynchronous brain-computer interface.