DefinePK

Abstract

In this article is presented the design and implementation of a prototype detector antipersonnel mines using neural networks and fuzzy logic, supported by the metal detector Garrett ACE 250. This prototype generates a signal source, which is amplified both voltage and current and entered to the primary winding of the sensor. When the emitted signal collides with a metallic object is returned to the sensor and becomes the signal to be processed. The offset angle is used to determine the type of metal and the approximate distance to which the detected object is located. Using a graphical interface in Matlab in real time the signal is shown to the user. As a result, a discriminator prototype metals, programmed in Matlab and the graphical user interface are obtained. The prototype was validated by experimental field testing and a better result was achieved with fuzzy logic achieving the expected discrimination of the types of metal used and the distance at which they are found.