LASER: Learning Active Sensing for Continuum Field Reconstruction

2026-04-21

AI Summary

This work addresses the challenge of high-fidelity reconstruction of dynamic continuous physical fields under sparse and constrained sensing conditions, where conventional approaches struggle due to their reliance on fixed sensor configurations. The paper introduces LASER, a novel framework that uniquely integrates closed-loop active perception with an implicit world model. The implicit world model learns the spatiotemporal dynamics of the physical field, while active sensing is formulated as a partially observable Markov decision process (POMDP). Leveraging reinforcement learning in the latent space, LASER evaluates hypothetical sensing actions and adaptively plans observation trajectories that maximize information gain. This approach overcomes the limitations of static or offline sensing strategies and achieves significantly superior reconstruction accuracy compared to existing methods across diverse physical field tasks, enabling high-fidelity recovery even from highly sparse observations.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Search and Optimization: Learning to Search
• Computer Vision: Low Level & Physics-based Vision

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

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.

Problem

• continuum field reconstruction
• active sensing
• sparse sensing
• sensor placement
• physical field measurement

Innovation

• active sensing
• continuum field reconstruction
• latent world model
• reinforcement learning
• POMDP