-Assisted Shape-Memory-Alloy Tentacle Arm

Abstract

Repetitive manual tasks in industrial production environments expose workers to sustained physical loads that elevate fatigue and the incidence of musculoskeletal disorders. Full automation addresses these risks but carries integration costs prohibitive for many small and medium manufacturers. This paper proposes a BCI-assisted tentacle-like mechanical arm actuated by shape memory alloy (SMA) wires as a worker augmentation device: a lightweight, modular soft-robotic limb controlled via non-invasive EEG-based motor imagery and supported by a compressed AI inference model running offline on a Raspberry Pi class embedded processor.

PDF