LASER: Learning Active Sensing for Continuum Field Reconstruction

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.

Measuring physical phenomena like fluid flows is difficult because we can only afford a few sensors, and most existing methods keep them fixed in place even as conditions change. We introduce LASER, a system that learns to predict how a physical field will evolve and uses this foresight to proactively move sensors toward the most informative locations. Across both simulated fluid problems and real-world data, LASER reconstructs physical fields more accurately than methods with fixed sensor layouts, which could benefit applications like environmental monitoring and climate science.