Local Conformal Calibration of Dynamics Uncertainty from Semantic Images

TL;DR

OCULAR is a conformal prediction algorithm that calibrates uncertainty in dynamics models using perception data, providing guarantees for unseen environments and aiding safe planning.

Contribution

It introduces a perception-aware conformal calibration method that guarantees uncertainty quantification without environment-specific data, even under model mismatch.

Findings

OCULAR provides guaranteed prediction regions with user-set likelihood.

It effectively distinguishes between high and low uncertainty regions in state-action space.

The method performs well on systems with model mismatch and out-of-distribution data.

Abstract

We introduce Observation-aware Conformal Uncertainty Local-Calibration (OCULAR), a conformal prediction-based algorithm that uses perception information to provide uncertainty quantification guarantees for unseen test-time environments. While previous conformal approaches lack the ability to discriminate between state-action space regions leading to higher or lower model mismatch, and require environment-specific data, our method uses data collected from visually similar environments to provably calibrate a given linear Gaussian dynamics model of arbitrary fidelity. The prediction regions generated from OCULAR are guaranteed to contain the future system states with, at least, a user-set likelihood, despite both aleatoric and epistemic uncertainty -- i.e., uncertainty arising from both stochastic disturbances and lack of data. Our guarantees are non-asymptotic and distribution-free, not requiring strong assumptions about the unknown real system dynamics. Our calibration procedure enables distinguishing between observation-velocity-action inputs leading to higher and lower next-state-uncertainty, which is helpful for probabilistically-safe planning. We numerically validate our algorithm on a double-integrator system subject to random perturbations and significant model mismatch, using both a simplified sensor and a more realistic simulated camera. Our approach appropriately quantifies uncertainty both when in-distribution and out-of-distribution, being comparatively volume-efficient to baselines requiring environment-specific data.