Researchers built a detailed computer model of a tiny, H-shaped robot powered by living heart cells and found that ventricular cells give much stronger and more steerable motion than atrial cells. The work asks whether the type of cardiomyocyte (heart muscle cell) used to power a biohybrid microrobot matters for autonomous navigation toward low-oxygen (hypoxic) tissue.

The model robot has two soft rubber arms (Ecoflex) 5 mm long and 1 mm thick, joined by a thin gold rod to keep the cells electrically connected. Each arm was divided into five rigid segments joined by flexible joints. The body rests on five small hemispherical legs (radius 150 µm) that create an anisotropic friction “ratchet” — in plain terms, the surface provides low resistance when sliding forward and much higher resistance when sliding backward so a travelling wave of contractions can produce net forward motion.

Actuation was driven by prescribed calcium signals that mimic neonatal rat cardiomyocyte activity. The authors used standard values for systolic and diastolic calcium (1.0 µM and 0.1 µM) and a phase-shifted activation that creates a posterior-to-anterior travelling wave with 0.25 s offsets between joints. This peristaltic, wave-like contraction combined with the directional friction produces locomotion. The model was run in two scenarios: straight-line motion under normal oxygen, and steering in an applied oxygen gradient to test autonomous heading towards hypoxia.

The simulations produced clear differences between cell types. At the same energetic cost of transport, ventricular cardiomyocytes gave a 4.35× increase in locomotion speed compared with atrial cells. Under a simulated oxygen gradient, the model with ventricular cells showed a 2.91× larger heading response than the atrial one. The authors highlight that choosing cardiomyocyte phenotype is an underused design choice for biohybrid locomotion.