Intelligent integrated sensing systems possess capabilities such as perceiving external information, data processing. It has a wide range of applications in Internet of Things (IOT), smart manufacturing, and other fields. However, if the information is acquired incorrectly, it will definitely affect the correctness of data processing. Though there have been many researches about sensor data processing, only a few focused on how to self-detect and self-diagnosis the working status of the sensors themselves. Based on the above, this paper proposes to evaluate the accuracy and reliability of sensor output by predicting the sensor’s real-time output with multi-sensor fusion. The proposed system can perform self-diagnosis to find out the cause or type of the error when failure occurs. Furthermore, this system can also perform real-time fault-tolerant conveniently. Experimental results show that this method has high recognition accuracy with small amount of calculation complexity. The intelligent integrated sensing system can be more reliable with the proposed self-diagnosis technologies for the smart sensors.

Self-Detection and Self-Diagnosis Methods for Sensors in Intelligent Integrated Sensing System

In this work, a hybrid neural network model (HNNM) is proposed, which combines the advantages of genetic algorithm, multi-agents and reinforcement learning In order to generate networks with few connections and high classification performance, HNNM could dynamically prune or add hidden neurons at different stages of the training process Experimental results have shown to be better than those obtained by the most commonly used optimization techniques.

Kui Zhang

Papers

A hybrid neural network model based reinforcement learning agent