MuViS: Multimodal Virtual Sensing Benchmark

TL;DR

MuViS is an open-source benchmarking platform that standardizes evaluation of multimodal virtual sensing methods across diverse datasets, highlighting the lack of a universally superior approach.

Contribution

Introduces MuViS, a domain-agnostic benchmarking suite for multimodal virtual sensing, enabling standardized comparison and fostering development of more generalizable models.

Findings

No single approach outperforms others across all datasets

Existing methods lack a universal advantage in virtual sensing

MuViS facilitates reproducible evaluation and future research

Abstract

Virtual sensing aims to infer hard-to-measure quantities from accessible measurements and is central to perception and control in physical systems. Despite rapid progress from first-principle and hybrid models to modern data-driven methods research remains siloed, leaving no established default approach that transfers across processes, modalities, and sensing configurations. We introduce MuViS, a domain-agnostic benchmarking suite for multimodal virtual sensing that consolidates diverse datasets into a unified interface for standardized preprocessing and evaluation. Using this framework, we benchmark established approaches spanning gradient-boosted decision trees and deep neural network (NN) architectures, and show that none of these provides a universal advantage, underscoring the need for generalizable virtual sensing architectures. MuViS is released as an open-source, extensible platform for reproducible comparison and future integration of new datasets and model classes.