Uncertainty Averse Pushing with Model Predictive Path Integral Control

TL;DR

This paper presents a method for robust robot pushing by integrating learned forward models with uncertainty estimates into a model predictive control framework, enabling the robot to avoid uncertain regions and improve manipulation robustness.

Contribution

It introduces a novel approach combining learned uncertainty-aware models with MPPI control for robust manipulation, outperforming physics simulation in real robot experiments.

Findings

Uncertainty-aware models improve push robustness.

The method outperforms physics simulation in real robot tasks.

Simulation results show effective uncertainty-averse path planning.

Abstract

Planning robust robot manipulation requires good forward models that enable robust plans to be found. This work shows how to achieve this using a forward model learned from robot data to plan push manipulations. We explore learning methods (Gaussian Process Regression, and an Ensemble of Mixture Density Networks) that give estimates of the uncertainty in their predictions. These learned models are utilised by a model predictive path integral (MPPI) controller to plan how to push the box to a goal location. The planner avoids regions of high predictive uncertainty in the forward model. This includes both inherent uncertainty in dynamics, and meta uncertainty due to limited data. Thus, pushing tasks are completed in a robust fashion with respect to estimated uncertainty in the forward model and without the need of differentiable cost functions. We demonstrate the method on a real robot, and show that learning can outperform physics simulation. Using simulation, we also show the ability to plan uncertainty averse paths.