Development of a microcontroller-based adaptive fuzzy controller for a two-wheeled self-balancing robot

In this paper, an intelligent system use an adaptation fuzzy controller using Mamdani algorithm modified by relation models for a two wheeled self balancing robot is developed. Hardware model of the robot and sensor signal processing are described. The signals from sensors are filtered by a discrete complementary filter. A mathematical model of the robot is derived based on Newtonian mechanics. The proposed control system comprises two loops for regulation of the pitch angle and tracking the desired position of the robot. The inner loop uses a PD controller for position tracking. The outer loop is designed with an adaptive fuzzy controller to regulate balancing of the robot. The proposed controllers are tested in simulations using the mathematical model. These controllers are also designed and implemented in the real time system using a STM32F4 DISCOVERY kit which is equipped with a 32-bit ARM7 microprocessor. Simulations and experimental results show advantages of the adaptation fuzzy controller. Using the adaptive fuzzy controller for stability of the robot will allow a more effective and robust control to be implemented.