Brain–computer interfaces (BCIs) have emerged as a technology for controlling devices like drones, robotic arms, and wheelchairs. This project focuses on understanding the connections between time data in BCI-controlled wheelchair systems. The emphasis is on comprehending how the timing and sequencing of various data streams and processes impact the overall functioning of these systems. We investigate the stages, real-time data acquisition, preprocessing, and feature extraction to machine learning-based classification. EEG acquisition of ten subjects was acquired three times, including forenoon, afternoon, and evening, of two different motor imagery actions: left wrist movement and right wrist movement, with eight and two reference electrodes. The data is then preprocessed, which involves several steps to clean and enhance the data. The main steps are filtering, artifact removal, and segmentation. Features extraction techniques are applied to the three target classes of ten subjects. These features are utilized in training machine learning algorithms, including neural networks, decision trees, and support vector machines. Five datasets of three timelines are classified. The medium neural network classification model outperformed other models across various datasets and timelines. The results suggested that the BCI-controlled wheelchair systems are influenced by the timing and sequencing of data acquisition, emphasizing the need for careful consideration of temporal factors in system design. While our study primarily delves into the influence of time dependency on BCI-controlled wheelchair systems, it’s crucial to note that other dependencies must also be considered.
Brain–Computer Interfaces for Mobility Assistance: A Comparative Analysis of EEG-Based Classification Models for Brain-Controlled Wheelchairs
Smart Innovation, Systems and Technologies
Congress on Control, Robotics, and Mechatronics ; 2024 ; Warangal, India February 03, 2024 - February 04, 2024
Proceedings of the Second Congress on Control, Robotics, and Mechatronics ; Kapitel : 20 ; 245-256
31.10.2024
12 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch