VISION-SLS: Safe Perception-Based Control from Learned Visual Representations via System Level Synthesis

TL;DR

VISION-SLS introduces a scalable control method using learned visual features and System Level Synthesis to ensure safety and constraint satisfaction in complex visuomotor tasks with high-resolution images.

Contribution

It combines learned low-dimensional visual representations with an efficient SLS-based output-feedback control approach for safe, scalable visuomotor control.

Findings

Successfully applied to simulated and real hardware tasks with high-resolution images.

Enables safe, information-gathering behavior with reduced uncertainty.

Outperforms baselines in safety and solve times on hardware.

Abstract

We propose VISION-SLS, a method for nonlinear output-feedback control from high-resolution RGB images which provides robust constraint satisfaction guarantees under calibrated uncertainty bounds despite partial observability, sensor noise, and nonlinear dynamics. To enable scalability while retaining guarantees, we propose: (i) a learned low-dimensional observation map from pretrained visual features with state-dependent error bounds, and (ii) a causal affine time-varying output-feedback policy optimized via System Level Synthesis (SLS). We develop a scalable, novel solver for the resulting nonconvex program that leverages sequential convex programming coupled with efficient Riccati recursions. On two simulated visuomotor tasks (a 4D car and a 10D quadrotor) with >= 512 x 512 pixels and a 59D humanoid task with partial observability, our method enables safe, information-gathering behavior that reduces uncertainty while guaranteeing constraint satisfaction with empirically-calibrated error bounds. We also validate our method on hardware, safely controlling a ground vehicle from onboard images, outperforming baselines in safety rate and solve times. Together, these results show that learned visual abstractions coupled with an efficient solver make SLS-based safe visuomotor output-feedback practical at scale. The code implementation of our method is available at https://github.com/trustworthyrobotics/VISION-SLS.