Data Fusion in Digital Twin

Data Fusion in Digital Twin

Data fusion copes with multisource data through synthesis, filtering, correlation, and integration.

Data fusion categories:

• raw-data-level fusion,
• feature-level fusion, and
• decision-level fusion.

Data fusion methods:

• random methods. Random methods (e.g., classical reasoning, weighted average method, Kalman filtering, Bayesian estimation, and Dempster-Shafer evidence reasoning,) are applicable for all three levels of data fusion.
• artificial intelligence. Artificial intelligence methods (e.g., fuzzy set theory, rough set theory, neural network, wavelet theory, and support vector machine) are applicable for the feature-level and decision-level data fusions.