Anti balance Load Control System Applied to an Overhead Crane Prototype Activated by Voice Commands
Keywords:Artificial Neural Networks, fuzzy control, mel-frequency cepstral coefficients, voice commands
This paper presents a proposal of anti-balance control applied to a didactic overhead crane, with the possibility of its activation by voice commands throw a Graphical User Interface. It was opted for the application of fuzzy intelligent control for the load anti-balance and Proportional-Derivative, PD, classic control with Ziegler-Nichols tuning for the car sidelong movement. The whole control is embedded in an ATmega 2560 microcontroller, responsible for commuting between the two forms of control, as well as accomplish the communication with the interface. Moreover, it was used the Mel-Frequency Cepstral Coefficients for the characterization of the voice signals and a Multilayer Perceptron artificial neural network for the classification of the extracted coefficients. After the accomplishment of the car movement tests and load balance analysis, it was verified that the fuzzy controller for the sidelong movement allied to the Ziegler-Nichols tuning for the anti-balance PD control and the speech commands recognition by artificial neural network were effective for the studied application.