Soft robots achieve better orientation control with new modeling method
Orientation Control of Soft Robots via Adiabatic Spectral Submanifolds
Robotics
Summary
Controlling soft robots accurately, especially their position and direction, is hard because they bend and move in complex ways. The authors improved a modeling technique that simplifies these complex movements, making it easier for control systems to guide the robot. They tested their method on a detailed computer simulation of a soft robotic arm and found it reduced mistakes in movement by over 60% compared to earlier methods. This helps soft robots work more precisely in delicate settings.
What this means in practice
- •For robotic system developers: Control soft robots more precisely by using improved data-driven reduced models in predictive controllers for delicate tasks.
- •For automation engineers: Enhance the accuracy of orientation and position control in soft robotic arms used in manufacturing by adopting advanced spectral submanifold methods.
Authors
Aron Karakai, Roshan S. Kaundinya, Mike Yan Michelis, Robert Katzschmann, George Haller
Abstract
Soft robots are commonly sought for safety-critical interactions in delicate environments, where accurate position and orientation control is imperative. Model predictive control (MPC) offers a solution, but it requires a model of the robot's infinite-dimensional nonlinear dynamics that is at once accurate and computationally cheap. Recent theory on adiabatic spectral submanifolds (aSSMs) and their applications to soft robots provide data-driven model-reduction methods to construct such models. Here, we extend these methods to identify aSSMs from enlarged observable datasets and upgrade the currently available aSSM-MPC schemes. Evaluated on a high-fidelity finite-element simulation of a pressure-actuated soft arm, our controller reduces position and orientation tracking error by more than 60% compared to existing data-driven baselines.