Distributionally Robust Control via Stein Variational Inference for Contact-Rich Manipulation

TL;DR

This paper introduces a distributionally robust control method using Stein variational inference to improve reliability and robustness in contact-rich robotic manipulation tasks, especially under uncertainty.

Contribution

It proposes a novel deterministic formulation for robust control that explicitly models task-sensitive uncertainty, enhancing performance and reliability.

Findings

Up to 3× improved robustness in manipulation tasks

Outperforms existing model-based control methods

Explicitly models task-sensitive parameter uncertainty

Abstract

Reliable robotic manipulation requires control policies that can accurately represent and adapt to uncertainty arising from contact-rich interactions. Modern data-driven methods mitigate uncertainty through large-scale training and computation, and degrade significantly in performance with limited number of training samples. By contrast, classical model-based controllers are computationally efficient and reliable, but their limited ability to represent task-relevant uncertainty can hinder performance in contact-rich interactions. In this work, we propose to expand the capabilities of model-based manipulation control through more flexible uncertainty modeling that retains performance while exactly adapting to uncertainty. Our approach casts the manipulation problem as a distributionally robust control optimization and proposes a novel deterministic formulation based on Stein variational inference that preserves performance while explicitly modeling task-sensitive parameter uncertainty. As a result, the derived controllers are more aware of task sensitivities to uncertainty, yielding high reliability without compromising performance. Experimental results demonstrate up to 3$\times$ improved robustness across a range of contact-rich manipulation tasks under broad parametric uncertainty, outperforming existing model-based control methods.