A Robust Controller based on Gaussian Processes for Robotic Manipulators with Unknown Uncertainty

TL;DR

This paper introduces a Gaussian Process-based robust control method for robotic manipulators that effectively manages unknown uncertainties, ensuring accurate trajectory tracking with high probability.

Contribution

It presents a novel learning-based feedback linearization approach that incorporates Gaussian Process regression to estimate and compensate for model mismatch in robotic systems.

Findings

Guarantees asymptotic trajectory tracking with high probability.

Successfully tested on a 2-DOF planar robot.

Enhances robustness without prior bounds on model mismatch.

Abstract

In this paper, we propose a novel learning-based robust feedback linearization strategy to ensure precise trajectory tracking for an important family of Lagrangian systems. We assume a nominal knowledge of the dynamics is given but no a-priori bounds on the model mismatch are available. In our approach, the key ingredient is the adoption of a regression framework based on Gaussian Processes (GPR) to estimate the model mismatch. This estimate is added to the outer loop of a classical feedback linearization scheme based on the nominal knowledge available. Then, to compensate for the residual uncertainty, we robustify the controller including an additional term whose size is designed based on the variance provided by the GPR framework. We proved that, with high probability, the proposed scheme is able to guarantee asymptotic tracking of a desired trajectory. We tested numerically our strategy on a 2 degrees of freedom planar robot.