Brain-Controlled Robotic Arm System Based on Dynamic Real-Scene Steady-State Visual Evoked Potentials

Le résumé fourni par la source

Abstract Brain-controlled robotic arm systems represent a crucial pathway to enhancing the autonomy of individuals with severe motor disabilities. However, traditional systems are constrained by abstract artificial stimulation interfaces and unidirectional control modes, suffering from issues such as low ecological validity, high user cognitive load, and weak environmental interaction capabilities. This paper innovatively proposes a brain-controlled robotic arm system based on Dynamic Real-Scene Steady-State Visual Evoked Potential (DR-SSVEP). Our system utilizes a depth camera integrated into the robotic arm to perceive the environment, mapping images and spatial coordinates of real objects into SSVEP stimulation sources, thereby achieving a “what you see is what you control” dynamic real-scene interaction. To further reduce the user’s cognitive load, we designed an asynchronous state recognition mechanism based on multi-feature fusion, allowing users to initiate interactions voluntarily at will. Experimental results demonstrate that the system achieves a robotic arm grasping accuracy of 92.3%, while significantly improving interaction naturalness and user satisfaction. This provides an effective solution for establishing an embodied interaction paradigm integrating “brain-machine-environment” fusion.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.