A Foundation Model for Virtual Sensors

2026-01-28

AI Summary

This work proposes the first lightweight foundation model tailored for virtual sensing, addressing key limitations of existing approaches that rely on manual input selection, lack task coordination, and incur prohibitive computational costs in time-series modeling—particularly for multi-output prediction. The proposed architecture enables joint forecasting of multiple target signals through a unified framework, integrating adaptive input selection and multi-task learning to automatically discover input relevance and exploit inter-task synergies. Notably, the model’s parameter count remains nearly constant regardless of the number of sensors. Evaluated on a dataset exceeding 18 billion samples, it achieves up to 415× faster computation and 951× lower memory usage compared to current baselines, while matching or surpassing their prediction accuracy, demonstrating efficient scalability to hundreds of virtual sensors.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Other Foundations of Computer Vision

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Virtual sensors use machine learning to predict target signals from available measurements, replacing expensive physical sensors in critical applications. Existing virtual sensor approaches require application-specific models with hand-selected inputs for each sensor, cannot leverage task synergies, and lack consistent benchmarks. At the same time, emerging time series foundation models are computationally expensive and limited to predicting their input signals, making them incompatible with virtual sensors. We introduce the first foundation model for virtual sensors addressing both limitations. Our unified model can simultaneously predict diverse virtual sensors exploiting synergies while maintaining computational efficiency. It learns relevant input signals for each virtual sensor, eliminating expert knowledge requirements while adding explainability. In our large-scale evaluation on a standard benchmark and an application-specific dataset with over 18 billion samples, our architecture achieves 415x reduction in computation time and 951x reduction in memory requirements, while maintaining or even improving predictive quality compared to baselines. Our model scales gracefully to hundreds of virtual sensors with nearly constant parameter count, enabling practical deployment in large-scale sensor networks.

Problem

• virtual sensors
• foundation model
• time series
• sensor networks
• computational efficiency

Innovation

• foundation model
• virtual sensors
• time series
• computational efficiency
• sensor fusion