Automatic generation of VHDL for controller of inverted pendulum in consideration of robustness

In this study, to automate electronic controller design, we apply evolutionary computation, starting with the targeted inverted pendulum and task definition, followed by the framework of applying a genetic algorithm to controller design automation, focusing on its robustness. We use complex programmable logic device (CPLD) for the controller data processing, and optimize very-high-speed integrated circuit hardware description language (VHDL), which describes the logical circuit in CPLD as the controller, using evolutionary computation. We present the fitness function we use to develop the controller automatically. Evolutionary simulation verified the feasibility of our framework.