Radar-infrared sensor track correlation algorithm based on neural network fusion system

Simultaneous capture of the texture and shape of a moving object in real time is expected to be applicable to various fields including virtual reality and object recognition. There are several difficulties must be overcome to develop a sensor able to achieve this feature: fast capturing of shape and the simultaneous capture of texture and shape. This paper presents a new type of neural network base fusion system infrared sensor, which is compatible to standard process. The proposed infrared sensor adopts a suspended n-well containing several p+/n-well diodes as infrared sensing element. The thermal analysis indicates that the sensor exhibits excellent steady-state and transient thermal property. In order to predigest the calculation, the measurement data are to be classified and selected. The fuzzy neuron network information fusion based on the T-S model is used to avoid obstacle for the mobile robot, which fully utilized the information coming from the sensors. Finally, the experiment with the autonomous robot proved that the method is really feasible and efficient.