Towards intelligent industrial systems: A comprehensive survey of sensor fusion techniques in IIoT

Abstract

Industrial Internet of Things (IIoT) is systems aim to facilitate human monitoring and the direction of efficient production of goods in industrial settings by linking a wide variety of intelligent devices such as sensors, actuators, and controllers. This is achieved by utilizing Internet of Things (IoT) to diagnose a problem with a specific IIoT part is to employ a basic diagnostic technique that's based on models and data. Physical models, signal patterns, and machine-learning strategies must be adequately built to account for system challenges. Another factor that could lead to an exponential rise in complexity is the ever-increasing interconnections between different electronic hardware. The knowledge-based defect diagnosis methods boost interoperability in the operation. Users don't need to be experts in the field to benefit from the system's high-level thinking and response to their queries. So, in advanced IIoT systems, a knowledge-based fault diagnostic approach is favored over traditional model-based and data-driven diagnosis methods. The goal of this study is to evaluate recent improvements in the design of knowledge-based defect detection in the context of IIoT systems, deductive and inductive reasoning, and many other forms of logical reasoning. IIoT-based systems have revolutionized industrial settings by connecting intelligent devices such as sensors, actuators, and controllers to enable efficient production and human monitoring. In this survey paper, we explore machine learning-based sensor fusion techniques within the realm of Industrial Internet of Things (IIoT), addressing critical challenges in fault detection and diagnosis.