Industrial automation facing the parallel shift challenge

World Congress on Industrial Automation

July 20-22, 2015 San Francisco, USA

Marius M Balas

Posters-Accepted Abstracts: Adv Robot Autom

Abstract :

The presentation points a new technological boom, following the 2008 global financial crisis, triggered by the development and dissemination of the FPGA technology in electronics and IT. Some outstanding achievements in avionics, biomedical/health care equipment and data flow com-puting are briefly analyzed, revealing a lush superiority of the FPGA systems comparing to the conventional bus oriented computing devices: microprocontrollers, DSP, etc. in most respects: speed, energy consumption, size, weight, reliability, etc. The industrial automation must take this wave, sooner the better. The FPGA revolution is due to the FPGA parallel programmed architecture. How-ever, the parallel shift comes with a price: the conception and the programming of the all parallel operating devices/embedded systems are totally different from the conventional ones, demanding new skills and dedicated training. Most of the existing FPGA applications are just emulating conventional programmed operation and lose such way almost all the potential advantages of the parallel computing. One of the adequate theoretical tools, the fuzzy interpolative methodology, able to bring the linguistic representation of knowledge into the embedded systems camp is exposed. This approach is able to implement fuzzy expert systems by means of look-uptables with linear interpolation, and it also can incorporate specific knowledge about the applications by means of internal models. Such systems are using a classical time analysis by phase trajectories. Some applications are eventually provided: the constant time to collision optimization of the traffic, the imposed distance braking, the bronze casting with controlled cooling rate, etc.