Farklı performans ölçümleri kullanılarak kararlı durum görsel uyarılmış potansiyel kontrollü quadcopter yolunun değerlendirilmesi

Abstract

This study focuses on controlling a quadcopter system using a steady-state visual evoked potential (SSVEP)-based brain-computer interface system. In the literature, researchers report the accuracy and information transfer rate. However, these measures do not provide sufficient information about the predicted and target path similarity. The drone is expected to follow a certain square-shaped path and return to its starting position. We calculated the final and mean distances as additional outcome measures using several classifiers. The results emphasize the importance of having a balanced confusion matrix in the performance of quadcopter control and provide a more complete picture in the evaluation of the quadcopter’s performance. Focusing on the relationship between classification accuracy and spatial deviation might create a new perspective for BCI-based control systems.