Regresar

Using open source embedded hardware and software tools in automatic control from mathematical model

Abstract:

This paper exposed the results for an automatic control implemented into a Raspberry Pi 2 B+ platform, using the Octave tool for mathematical modeling and the controller running. Linux operating system was used for this purpose, being installed the mathematical modeling tool and the complimentary toolboxes for automatic control. The distribution of Linux chosen was Raspbian, which is the vendor's official recommended distribution with Debian bases. It was implemented a DC brushed motor control to test the hypothesis that is possible to build applications for classical automatic control using this platform, taking the motor velocity as a variable control interpreted like a differential of potential, showing an example of control. It was necessary to use additional devices to complete the system, like an external Analog to Digital Converter (ADC), due to the selected embedded board lacked this component. It was also implemented additional software libraries for communication between the control system and the real external world across the pins of the board. These libraries were PIGPIO and Oct2Py. The first library generates PWM pulses using C++ and Python, and the second one is used to open a session between Octave and Python, and can be used Octave in the Python environment. In the last stage, the test was running in both languages, taking several measurements for the control system. With the results of the experiments was determined that the implemented controller in Python with the Oct2Py was not enough reliable to run continuously in a control system. On the other hand, the test on the controller implemented in C++ resulted in better outputs allowing classical control systems to be executed permanently without any problem and satisfying the hypothesis.