Model Predictive Control for Magnetically Actuated Cellbots

This article presents a control strategy for, which are magnetically actuated, micron‐scale robots embedded into living cells. Our approach combines model predictive control, which provides low computational complexity and adaptability to changing environments, with Gaussian Processes (GPs), which are used as disturbance estimators for precise path tracking. To address the challenges posed by unmodeled dynamics, we integrate data‐driven modeling with model‐based control to accurately track desired trajectories using relatively small data. To the best of our knowledge, this is the first work to integrate data‐driven modeling with model‐based control for magnetically actuated cellbots. The GP effectively learns and predicts unmodeled disturbances, providing uncertainty bounds as well. The experimental results show that our controllers perform very well, even in environments cluttered with cells.