LASER: Learning Active Sensing for Continuum Field Reconstruction

TL;DR

LASER introduces a reinforcement learning framework for adaptive sensing that actively navigates sensors to high-information regions, significantly improving continuum field reconstruction from sparse measurements.

Contribution

LASER is a novel active sensing method using a POMDP and latent world model to adapt sensor placement dynamically for better field reconstruction.

Findings

LASER outperforms static sensing strategies in diverse scenarios.

It achieves high-fidelity reconstruction with sparse measurements.

The framework effectively navigates sensors to informative regions.

Abstract

High-fidelity measurements of continuum physical fields are essential for scientific discovery and engineering design but remain challenging under sparse and constrained sensing. Conventional reconstruction methods typically rely on fixed sensor layouts, which cannot adapt to evolving physical states. We propose LASER, a unified, closed-loop framework that formulates active sensing as a Partially Observable Markov Decision Process (POMDP). At its core, LASER employs a continuum field latent world model that captures the underlying physical dynamics and provides intrinsic reward feedback. This enables a reinforcement learning policy to simulate ''what-if'' sensing scenarios within a latent imagination space. By conditioning sensor movements on predicted latent states, LASER navigates toward potentially high-information regions beyond current observations. Our experiments demonstrate that LASER consistently outperforms static and offline-optimized strategies, achieving high-fidelity reconstruction under sparsity across diverse continuum fields.