Dynamic tunable and reconfigurable hardware controller with EKF-based state reconstruction through FPGA-in the loop

This work investigates the dynamic tunable and reconfigurable control of an inverted pendulum model implemented in MATLAB by using FPGA in-the-loop (FIL). A state-based controller is implemented to maintain the pendulum in the upright position. In order to reduce the number of sensors, an Extended Kalman Filter (EKF) is implemented which reconstructs the missing states. Initially, the pendulum is in the equilibrium position and a PID-Controller is required to move the pendulum to the upright position. Different variants of a feedback controller like P/PI/PD/PID-Controller have been designed in order to fulfill this change of position which can be added or removed from the design by using dynamic partial reconfiguration. The main benefit of the proposed technique is that FIL can be used to validate the controller with the simulation model before using it with a real inverted pendulum system. Furthermore, the hardware controller can be fine-tuned and reconfigured at runtime for time-variant systems. In case of sensor malfunction, the methodology can still be used to control the model because of the state reconstruction through EKF.