BCIwiki (bciwiki.com) — Jointly optimizing signal quality, decoding algorithms and shared control strategies is key to making brain-controlled robotic arms more natural and reliable, Hao Kang of the University of Glasgow concluded in a review published in Theoretical and Natural Science on September 8, 2026. The journal is published by EWA Publishing (DOI: 10.54254/2753-8818/2026.hz36492).
Brain-computer interfaces (BCIs) that decode motor cortex activity can give patients with severe motor disorders a way to bypass damaged nerve pathways and control external devices, the review says. Research has advanced in continuous movement, multi-degree-of-freedom control and sensory feedback, but long-term stability, control accuracy and clinical safety remain limited.
The paper analyzes signal acquisition in invasive, non-invasive and intravascular BCIs, outlines the preprocessing, feature extraction and neural decoding pipeline, and summarizes methods for generating continuous, discrete and multi-degree-of-freedom robotic arm commands.
Future work should focus on stable neural interfaces, adaptive individual decoding, computer vision assistance and two-way sensory feedback to support long-term clinical and everyday use, the author wrote.