Auto-Tuning of Model Predictive Control for Bilateral Teleoperation with Bayesian Optimization
Abstract
Model Predictive Control (MPC) is becoming a popular control method for teleoperation due to its ability to ensure safety constraints. However, tuning MPC is a nonintuitive process that requires significant expertise and effort. In this work, we propose a method for auto-tuning a model predictive controller in bilateral teleoperation settings. We use the Bayesian Optimization algorithm (BO) to seek the optimal weights of the MPC cost function for precise teleoperation. Our simulations and experiments show the effectiveness of the proposed tuning method.
| Origin | Files produced by the author(s) |
|---|